diff --git "a/designv11-84.json" "b/designv11-84.json" new file mode 100644--- /dev/null +++ "b/designv11-84.json" @@ -0,0 +1,8848 @@ +[ + { + "image_filename": "designv11_84_0002250_amr.850-851.317-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002250_amr.850-851.317-Figure1-1.png", + "caption": "Fig. 1 Overall view of the designed shape", + "texts": [ + " Thus it can be concluded that a circular cylindrical shape with streamlined seepy nose and tail fairings is appropriate for the pressure hull. Cylindrical shell can cooperate with a variety forms end closures, which can be roughly divided into flat one, spherical one, elliptical one, dished one and so on. For the ease of manufacturing and cost efficiency, pure monocoque construction or the one with ribs are in consideration. A circular cylindrical shape with flat end caps is considered in this paper due to its properties and uncomplicated fabrication process. After the overall shape (see in Fig.1) was determined, the next step is to estimate the internal deployment space. Based on the size of the attitude and buoyancy adjusting mechanism, it is established that the required deployment size with some allowance is 1.5 m in length and with an inner diameter of 0.22 m. The successful development of such system which will go down to the depth of 1500m would depend on the availability of a suitable material for construction. An advanced material with diverse properties will certainly be required" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002883_s00542-011-1333-8-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002883_s00542-011-1333-8-Figure6-1.png", + "caption": "Fig. 6 Schematic of the sensor structure", + "texts": [ + " Considering the calculated standard deviations (error bars), no distinct relationship can be concluded between the activity and the pH of the solution. Further analysis is therefore needed to determine the exact mechanism of immobilization. 3 Photoresist-based biosensor 3.1 Methods To prove the applicability of the immobilization scheme in functional device structures on common MEMS substrates we fabricated a simple biosensor based on a three-electrode setup (working, counter and reference electrode) (Shum et al. 2009), surrounded by large areas of NR71 (Fig. 6) for the inclusion of enzyme functionality. Sensor structures were fabricated on 500 nm PECVD SiO2/Si substrates. The electrodes were deposited via e-beam evaporation of 100 nm Pt/100 nm Pd/10 nm Ti and patterned via subsequent lift-off in acetone. The diameter of the counter electrodes was *1 mm with the width of the metal lines being *75 lm. The width of the working electrode and the spacing between features were *50 lm, respectively. Thereafter, the NR 71 was spin-coated onto the fabricated structures and patterned to expose the electrodes" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002993_s10846-013-9932-5-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002993_s10846-013-9932-5-Figure1-1.png", + "caption": "Fig. 1 Quad-rotor", + "texts": [ + " Finally, Section 7 gives conclusions of this work. In order to show the behavior of the GPA scheme, we will implement the algorithm in a formation of quad-rotors. The generalized coordinates for the quad-rotor are the following q = (x, y, z, \u03c6, \u03b8, \u03c8) \u2208 R 6 (1) where \u03be = (x, y, z) \u2208 R 3 denotes the position of the center of mass of the rotorcraft, relative to the inertial frame I, and \u03b7 = (\u03c6, \u03b8, \u03c8) \u2208 R 3 represents the Euler angles (roll, pitch and yaw respectively) that describe the rotorcraft orientation. Figure 1 shows a quad-rotor with its generalized coordinates. The dynamic model of the quad-rotor used in this paper is an Euler-Lagrange approach [5] mx\u0308 = u(cos\u03c8 sin \u03b8 cos\u03c6 + sin\u03c8 sin\u03c6) (2) my\u0308 = u(sin\u03c8 sin \u03b8 cos\u03c6 \u2212 cos\u03c8 sin\u03c6) (3) mz\u0308 = u cos \u03b8 cos\u03c6 \u2212 mg (4) \u03c8\u0308 = \u03c4\u03c8 (5) \u03b8\u0308 = \u03c4\u03b8 (6) \u03c6\u0308 = \u03c4\u03c6 (7) 2.1 Yaw and Altitude Control To maintain the yaw angle constant \u03c8 = 0 \u2200t > 0, the control used is \u03c4\u03c8 = \u2212kp\u03c8\u03c8 \u2212 kv\u03c8 \u03c8\u0307 (8) where kp\u03c8 and kv\u03c8 denote the proportional and derivative gains, respectively, of the PD yaw control" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003771_amm.532.320-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003771_amm.532.320-Figure2-1.png", + "caption": "Fig. 2 Herringbone gear Fig. 3 Herringbone gear meshing assembly", + "texts": [ + " First involute equation is derivated: b b b b x r cos( ) r rad( ) sin( ) y r sin( ) r rad( ) cos( ) 0 k k k k k k z \u03b8 \u03b8 \u03b8 \u03b8 \u03b8 \u03b8 = \u00d7 + \u00d7 \u00d7 = \u00d7 \u2212 \u00d7 \u00d7 = The helix equation over the involute starting point is: cos sin b b i s i i i i i Bt p X r Y r Z Bt \u03c0 \u03b8 \u03b8 \u03b8 = = \u00d7 = \u00d7 = The overall modeling process is shown in figure 1. Applied Mechanics and Materials Vol. 532 (2014) pp 320-323 Online: 2014-02-27 \u00a9 (2014) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMM.532.320 All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 141.211.4.224, University of Michigan Library, Media Union Library, Ann Arbor, USA-09/07/15,03:16:11) Modeling results are shown in figure 2. After modeling, it is needed to model another herringbone gear which matches with the first herringbone gear. The modeling process is the same as above. The meshing assembly results are shown in figure 3 by using UG gear assembly module assembly. 3 Kinetic Model Construction 3.1 Model Import Since there is no special interface between UG and ADAMS, therefore, the file format convert is needed to transfer UG model to ADAMS model. Commonly used convert formats include five different formats, i.e" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003715_robio.2014.7090476-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003715_robio.2014.7090476-Figure5-1.png", + "caption": "Fig. 5. Coordinate transformation", + "texts": [ + " When sufficient points are attained, the calibration can be completed and sufficiently accurate result may be obtained at one time. This procedure is programmed and embedded into our offline programming system. The important advantage of this calibration method lies in that no expensive equipment is needed. Now the coordinates of the sphere center are known in the base coordinate frame of the robot. After converting the coordinates of the trajectory points on the spherical surface into the base coordinate frame, the target positions of TCP may be easily computed. As shown in Fig.5, OP = OO\u2032 +O\u2032P . When the TCP of a robot moves on a spherical workpiece for some tasks such as welding, polishing and carving, the orientation of the tool is usually constrained. In general, there are two types of constraints for the control of tool orientation: 1) the tool axis is normal to the surface at the desired position of TCP, and 2) the tool axis keeps a constant angle with one of the three axes of the object frame located at the object center. For the first constraint, the axis of the tool should be in the normal direction of the surface at the desired point, or in other words, the tool axis should direct to the center point of the sphere. To satisfy this constraint, according to the kinematics, the Z axis direction of the tool coordinate frame should be \u2192 z = (xp \u2212 xo\u2032 , yp \u2212 yo\u2032 , zp \u2212 zo\u2032), where (xp, yp, zp) and (xo\u2032 , yo\u2032 , zo\u2032) are the coordinates of the TCP and the sphere center in the base frame of the robot, respectively, refer to Fig.5. For the second orientation constraint, suppose that the tool axis is to keep a constant angle \u03b8 with the Z axis of the object frame, refer to Fig.6(a). Let P (x\u2032, y\u2032, z\u2032) denote the coordinates of TCP in the object frame. Then the coordinates of the intersecting point of the tool axis and the Z axis of the object frame can be easily found as (0, 0, \u221a x\u20322 + y\u20322 \u2217 cot\u03b8 + z\u2032). The Z axis orientation of the tool coordinate frame is thus (in the object frame) \u2192 z = ( \u2212x\u2032,\u2212y\u2032, \u221a x\u20322 + y\u20322 \u2217 cot\u03b8 ) . Since x\u2032 = xp \u2212 xo\u2032 , y \u2032 = yp \u2212 yo\u2032 , and z\u2032 = zp \u2212 zo\u2032 , we get \u2192 z = ( xo\u2032 \u2212 xp, yo\u2032 \u2212 yp, \u221a (xo\u2032 \u2212 xp)2 + (yo\u2032 \u2212 yp)2 \u2217 cot\u03b8 ) " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002114_amr.291-294.1195-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002114_amr.291-294.1195-Figure2-1.png", + "caption": "Fig. 2. Diagram of Stress Analysis of Cycloid Gear", + "texts": [ + " Thus, under the condition of that the basic dimension is not changed, by conducting optimization of profile modification and parameters optimization on traditional structure, the axial dimension is increased by width of one cycloid gear, transmitting torque is increased about 50%, and the goal of large capacity with small size is realized. The structure of three cycloid gears is suitable for those cases of large speed ratio. Japan FA adopts the structure of three cycloid gears, of which the driving ratios are i=29, 59, 89, 119 respectively. See Fig. 1. No-clearance meshing of standard gears and analysis of meshing between pin gear and cycloid gear Discussion of Deformation Coordination Conditions Fig. 2 is a sketch of force analysis. Suppose the pin gear is stable, a moment cT is stressed on the cycloid gear; the cycloid gear rotates by an angle of \u03b2 due to the elastic deformation of driving part. If the deformation of cycloid gear, pin gear sleeve (or pin gear) and rotating arm is ignored, the bending of gear pin and total deformation of gear contacting and squeezing are calculated. As the ith gear, its deformation is: \u03b2\u03b4 .ii l= (1) All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002011_2011-01-1691-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002011_2011-01-1691-Figure10-1.png", + "caption": "Figure 10. Transmission paths in upper body", + "texts": [ + " The rear-side vibration paths are in right- and left- hands on the rear quarter windows, center pillars and rear pillars. The vector diagram of the contributions of each path at 118Hz is shown in Figure 9. The front-side path is principal. The other paths whose vectors are almost orthogonal to the response are negligible to the front-side path. Therefore, the front-side path is focused to analyze. In order to analyze the front-side path effectively, the upper part of the body model is cut out from whole body model and is loaded on the cutting section as shown in Figure 10. The response of the cut body model is confirmed to be equal to that of the whole body model. For the reduction of the roof response, nine VT paths at the upper body structure shown in Figure 5 are focused. Figure 11, 12, 13 shows the VT characteristics that are calculated and visualized by the proposed techniques. Figures 11, 12 (a)-(d) and 13 (a) show the influence degree for translational forces in the directions of xyz-axis, and Figures 13 (b) shows the influence degree for moment around the yaxis because other influence degrees are so small to be omitted" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001312_icuas.2015.7152309-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001312_icuas.2015.7152309-Figure1-1.png", + "caption": "Figure 1: Device design.", + "texts": [ + " Objects which are purpose built to be aerodynamically stable in flight and free fall. Although stable flight was required for safe operation, too much stability decreases the controllability, thus limiting possible maneuvers. Therefore dual actuation vanes, allow for maneuverability whilst still maintaining stable flight. Carbon-fiber has been used for the shell, with internal framework out of MDF. Ensuring all structures above the seam line are as lightweight as possible thus aiding to ensure the C.o.G will be in maintained at the seam. Figure 1 identifies the coordinate system for the device. Where the origin is coincident with the center of gravity of the device and the axes of motion in which the behavior of the device is to be analyzed. The drag on the device is increased by deflecting the vanes by the collective deflection angle, \u03b2c, shown in figure 2. In this way, the device can adjust its descent velocity. Maximum deflection of the vanes is 67\u00b0. To achieve the nominal descent velocity of 54ms-1, the device is designed that the vanes need only to deflect 20\u00b0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002291_s1064230713020056-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002291_s1064230713020056-Figure7-1.png", + "caption": "Fig. 7. SN and SA methods.", + "texts": [ + " ND and AD Methods The possible presence of impulse modes I and V on the winding is a specific feature of the ND and AD methods. Therefore, the mathematical description of the speed\u2013torque characteristics of the ND and AD methods (Fig. 6) is a combination of the corresponding expressions for impulse modes I and V: 7.5. SN and SA Methods When the SN and SA methods are used, there may be impulse modes I\u2013IV on the winding depending on the load torque, the PWM duty cycle, and the sequence of alteration of voltage pulse signs within the PWM period. Thus, the speed\u2013torque characteristics (Fig. 7) are described by the expression Tavg* 2\u03b3 1\u2013 \u03a9avg* for UM\u2013 +UPOW UPOW\u2013,{ },= 2\u03b3 1 \u03a9avg*+ \u03c4a \u2013 2e \u03b3\u03c4a 1\u2013 \u03a9avg*+ 1 \u03a9avg*+ for UMln +UPOW UPOW\u2013 , ,{ },= 2\u03b3 1 \u03a9avg*\u2013 \u03c4a + 2e \u03b3\u03c4a 1\u2013 \u03a9avg*\u2013 1 \u03a9avg*\u2013 for UMln +UPOW UPOW\u2013 , ,{ },=\u2013 2\u03b3\u2013 1 \u03a9avg* for UM\u2013+ UPOW\u2013 +UPOW,{ }.=\u23a9 \u23aa \u23aa \u23aa \u23aa \u23a8 \u23aa \u23aa \u23aa \u23aa \u23a7 = Tavg* \u03b3 \u03a9avg* for UM\u2013 +UPOW 0,{ },= \u03b3 \u03a9avg* \u03c4a \u2013 e \u03b3\u03c4a 1\u2013 \u03a9avg*+ \u03a9avg* for UMln +UPOW 0 , ,{ },= \u03b3 \u03a9avg* \u03c4a \u2013 1 e \u03b3\u03c4a \u03a9avg*+\u2013 \u03a9avg* for UMln UPOW\u2013 0 , ,{ },=\u2013 \u03b3\u2013 \u03a9avg* for UM\u2013 UPOW\u2013 0,{ }.=\u23a9 \u23aa \u23aa \u23aa \u23aa \u23a8 \u23aa \u23aa \u23aa \u23aa \u23a7 = 262 JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL Vol" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001968_2013-01-0424-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001968_2013-01-0424-Figure2-1.png", + "caption": "Figure 2. Disc-Brake System", + "texts": [ + " To feed the ABS ESP and ARC controller with input signals from the vehicle model, we will add virtual transducer signals in the body templates(see Table 1.). The disc-brake system template represents a device that applies resistance to the motion of a vehicle. The caliper part is mounted to the suspension upright, while the rotor is mounted to the wheel. The toe and camber values that the suspension subsystem publishes define the spin axis orientation. In addition, A rotational SFORCE is applied between the two parts (Fig.2). The braking torque is expressed as a function of a number of parameters. For a detailed description of the force function, see [7]. To feed the ABS and ESP controllers with input signals from the vehicle model, we will add virtual transducer signals in the brake system templates (see Table 2.). Then, all necessary actuator signals will be created which control the brake torque in the brake system (see Table 3.). The front and rear suspension models are double-wishbone suspension, made up of upper and lower control arms, coil springs, shock absorbers and other components" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001641_ist.2015.7294539-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001641_ist.2015.7294539-Figure1-1.png", + "caption": "Figure 1. The structure profile of the electromagnetic measurement sensor array", + "texts": [ + " Under such considerations, in this paper, in accordance with the mathematical basis of tomography - Radon transform and its inverse transform, through a combination of the physical meaning of sensitivity field and the characteristics of electromagnetic holographic measurement approach for flow imaging, a holographic measurement sensitivity field is established, and its rationality and applicability are validated by the numerical test and the measurement data. II. ELECTROMAGNETIC HOLOGRAPHIC MEASUREMENT MODEL There are 16 electrodes equally and azimuthally distributed in the sensor array of the electromagnetic holographic measurement, as is shown in Figure 1. Through circumferential scanning, 11*16=176 amplitude/phase data could be obtained in the measurement procedure for each of the flow pattern. These are the whole data of one holographic measurement procedure. In practical applications, through mathematical transformation, the amplitude and phase data of electric potential are usually fused into complex potentials which contain both the real and imaginary parts, so as to overcome the dimensional variances. 978-1-4799-8633-0/15/$31.00 \u00a92015 IEEE Cai et al" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002090_j.disopt.2013.07.004-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002090_j.disopt.2013.07.004-Figure1-1.png", + "caption": "Fig. 1. Unloading boxes off a gravity conveyor.", + "texts": [ + " Montr\u00e9al (Qu\u00e9bec), Canada, H3C 3A7 b D\u00e9partement d\u2019informatique et de Math\u00e9matique, Universit\u00e9 du Qu\u00e9bec \u00e0 Chicoutimi, Chicoutimi (Qu\u00e9bec), Canada, G7H 2B1 a r t i c l e i n f o Article history: Received 4 October 2012 Received in revised form 10 July 2013 Accepted 19 July 2013 Available online 22 August 2013 Keywords: Dynamic programming Gravity conveyor Boxes Unloading a b s t r a c t In this paper, we study the problem introduced by Baptiste et al. (2011) [3] of minimizing the number of steps to unload a set of boxes off a gravity conveyor. We show that this problem can be solved in polynomial time with a dynamic programming algorithm that runs inO n3A log F time,where n is the number of boxes initially lined up on the conveyor, A is the size of the accessible zone, and F is the forklift capacity. \u00a9 2013 Elsevier B.V. All rights reserved. 1. Introduction A finite number of boxes are placed on a gravity conveyor, as shown in Fig. 1. Each box has a destination, and boxes with different destinations are depicted with different colors. A forklift driver needs to unload all these boxes from the conveyor andmove them to their corresponding destinations. At each step, he picks a contiguous group of boxes with the same color. The number of such boxes is restricted by the forklift capacity. At any given step, only some of the boxes \u2013 those in the accessible zone, shown in Fig. 1 \u2013 can be removed from the conveyor. As a contiguous group of boxes is unloaded, a gap is momentarily formed. It is immediately filled by the boxes to its right (which are positioned at a higher height), which slide down. This movement leads to new boxes entering the accessible zone. The order in which boxes are unloaded can greatly influence the amount of time needed before all of them are moved to their corresponding destinations. Our goal is to develop strategies which unload the gravity conveyor in as fewmoves as possible" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001215_sibcon.2015.7147304-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001215_sibcon.2015.7147304-Figure1-1.png", + "caption": "Fig. 1. Piston machine", + "texts": [ + " However, the use of the kind of maintenance requires the application of means of non-destructive control. Vibration-based diagnostic is the most informative means of non-destructive control for pumps and rotary compressors. At the same time the vibration, examination doesn\u2019t make it possible to come to one-valued conclusion about the condition of the isolated assemblies for pumps and piston compressors. That's why the development problem of the effective means of non-destructive control for pumps and piston compressors appears [2]. II. CHECK POINTS In piston compressors (Figure 1) the most vulnerable from the point of view of wear and at the same time hard diagnosed assemblies are such as: a main bearing and a rod bearing, a finger connection of a rod and a piston, walls of a piston and a cylinder, a valve gear link. In addition, the critical wear of any listed assemblies can bring to the compressor destructive or to the contact of work body with oil in the compressor case and atmosphere. The increased wear of both main or connecting rod bearings and a finger connection is accompanied by the local heat not practically having an influence on machine temperature" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002867_icsenst.2011.6137012-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002867_icsenst.2011.6137012-Figure1-1.png", + "caption": "Fig. 1. The measurement setup consists of a blade above a sensor array which lies within a reference plane.", + "texts": [ + " A fast and simple way to determine these geometric properties is to apply voltage to the blade in order to generate an electric field between the blade and sensing electrodes at a reference potential. Changes in the blade geometry cause variations in the electric field which can be measured by suitably positioned electrodes. In order to achieve an optimal sensitivity of the measurement setup knowledge about the structure of the electric field is crucial. The measurement setup is illustrated in Fig. 1. In our case the absolute value of the electric field is measured using capacitance sensors. In order to be able to perform a sensitivity analysis the structure of the electric field needs to be known analytically. The electric field can be obtained by numerical analysis, e.g. by finite element methods. Unfortunately, numerical methods are not suited for sensitivity analysis where a representation by field equations is preferable. In general the equations describing the field structure in a measurement setup are difficult to obtain. Conformal mapping methods offer the possibility of obtaining field equations for certain types of problems, for examples see e.g. [1], [2], and [3]. The main restriction of this method is that only 2D problems can be analyzed. For our setup we can assume that fringe effects do not influence the electric field because the blade is extended along the z-axis. Thus, in this case conformal maps can be utilized to obtain an equation which defines the electric field in the xy-plane of Fig. 1. The basics of conformal mapping are given in [4] and [5]. A particularly useful method for obtaining conformal maps, the Schwarz-Christoffel Transform (SCT) is presented in detail in [6]. In the first part of this paper we use conformal mapping to obtain a mapping function which maps a reference domain to the section plane of the measurement setup. The field structure in the reference domain is known which enables us to analyze the original problem. The aforementioned mapping function represents the field equation. In the second part this field equation is used to compute the absolute value of the electric field in the measurement setup. Based on the absolute value the sensitivity with respect to the various blade parameters is computed. Finally, the influences of the blade parameters are determined with respect to the position of the sensor array. In this chapter we construct an equation representing the electric field between the blade and the sensor array in the xy-plane, see Fig. 1. This is achieved by developing 978-1-4577-0167-2/11/$26.00 \u00a92011 IEEE 417 a function which conformally maps a reference domain to the measurement setup. An ideal infinite plate capacitor is used for the reference domain because its field structure is known. Fig. 2 illustrates the idea. The ideal plate capacitor is positioned in the w-plane where lines parallel to the u-axis represent equipotentials, and lines parallel to the v-axis represent the lines of force. The coordinates are interpreted as components of a complex number w = u+ jv which are mapped to the z-plane with z = x + jy via a function z = f(w)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002641_s12206-013-0504-1-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002641_s12206-013-0504-1-Figure4-1.png", + "caption": "Fig. 4. Sketch of sensor locations.", + "texts": [], + "surrounding_texts": [ + "Fig. 1 depicts our rotordynamic test rig. A slender steel shaft is supported by two steel oil journal bearings lubricated by ISO VG32 turbine oil. The middle of the shaft holds a steel sleeve coupled with labyrinth seals with a diameter of 180 mm. Two steel balance disks near the bearings are used to regulate the original vibration of the rotor and provide unbalanced exciting forces. The magnitudes and angles of unbalanced excitation forces can be changed by changing the location of unbalance blocks, which are fixed on the balance disks. Six rings of copper labyrinth seals are fixed on the interior wall of the cylinder, as shown in Fig. 2. The cylinder is hung by springs (which can be exchanged to adjust the stiffness) from the vertical and horizontal directions. The cylinder is ensured to be excited easily only by seal force, which has the same frequency as the rotor rotation. At the free end of the shaft, a 15 KW dc motor drives the shaft through a 4:1 ratio speedincreasing gearbox via a disc-type coupling. The rotor speed can be adjusted from 500 rpm to 6,000 rpm. An electromagnetic shaker is used to apply dynamic loads to the cylinder with loads up to 500 N to obtain the impedance function of the cylinder. The dynamic loads applied to the cylinder are measured with the load cell located between the stinger and the shaker frame. Figs. 3 and 4 show that two sets of eddy current sensors, which are 90\u00b0 apart, are secured through the wall of the cylinder, facing the outer diameter of the steel sleeve from the hori- zontal and vertical directions. The sensors record the relative displacements of the rotor and cylinder along the two orthogonal directions. Each set includes two sensors to record the relative displacement between the cylinder and rotor from vertical and horizontal directions. Four magnetoelectric velocity transducers are located on the two ends of the cylinder to monitor the absolute vibrations from horizontal and vertical directions. One eddy sensor is fixed on the bearing box to work as a keyphasor transducer. The phases of all vibration signals are based on this keyphasor signal. A pressurized air line, which feeds into the middle cavity of the cylindrical vessel, is outfitted with multiple pressure taps, a metal tube rotameter, a static pressure transducer, and a thermocouple. All measurements are conducted at room temperature (28\u00b0C). Table 1 and Fig. 5 show the details of the test seal dimensions and operating conditions for measurements of seal force coefficients." + ] + }, + { + "image_filename": "designv11_84_0003600_12.2072609-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003600_12.2072609-Figure6-1.png", + "caption": "Figure 6 The flaw for SLM manufacturing overhang structure", + "texts": [ + " If the designed motion clearance is too large, the stability of movement will be affected, if the designed motion clearance is too small, the processability of SLM manufacturing will be hard. Therefore, when design the non-assembly abacus, the manufacturing factors should be considered in the design process. One of the important factors of SLM manufacturing for the motion gap can be described as geometric features resolution. In order to test the min. size of motion clearance and the min. size of rods, different size of clearance and rods was manufactured by SLM, as shown in the figure 6. The results proved that larger than 150\u03bcm clearance could be manufactured. The rods with diameter ranged from 0.15mm to 5mm all could be manufactured well. Considering the smaller rods result in lower mechanical properties, so this experiment did not study on the tiny rods geometrical features with size less than 0.15mm. Table2 The chemical compositions of powdered material C Cr Ni Mo Si Mn O Fe Proc. of SPIE Vol. 9295 929510-3 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 05/18/2015 Terms of Use: http://spiedl", + " When manufactured inclined plane, the \"step effect\" will occur, shown in the figure5(c), the ladder error will increase with the reducing the inclined angle. Learning the minimum building size when manufacturing tiny clearance and rods, the 0.18mm motion clearance and larger than 0.5mm rods were designed in copper cash abacus and collapsible abacus During the manufacturing processing, \"step effect\" happened, because SLM is a layer additive manufacturing technology. Meanwhile, overhang parts would emerge warping effect. Warping effect is due to the thermal stress formed by rapid solidification of melting pool during SLM process, as shown in the Figure 6(a). Proc. of SPIE Vol. 9295 929510-4 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 05/18/2015 Terms of Use: http://spiedl.org/terms Many researchers, such as Kruth et al8 investigated that temperature gradient mechanism could explained the warping phenomenon. At the same time, when overhang parts were manufactured , lase irradiate powder-support zone, as Figure 6(b), point b, where the heat conduction rate is only 1/100 of laser irradiate solid-supported zone, can absorb more energy input than laser irradiate solid-support zone. Therefore, the melt pool at overhang parts become too bigger and sinks into the powder as the result of gravity and capillary fore. The dross will be formed. In the actual SLM process, wrapping effect and dross happened at the same time, and influenced the processability of SLM process. Resolution is that support structure should be used to prevent producing the dross, warping and other defects for overhang pars", + " The overhang angle \u03b1 is described by the length of N +1 layer' overhang portion S and the layer thickness H, as shown in the Figure5\uff0cand the relationship as the following equation shows: )/tan( HS=\u03b1 (1) In this article, when S is larger than one melt pool, the support structures should be added. when S is smaller than 1/2 melt pool, the support structures can be avoid to be added. According to this theory, the overhang angle can be divided into three regions at different thickness, including stable fabrication zone, critical fabrication zone, and hard fabrication zone as shown in Figure 7. During SLM manufacturing processing\uff0clayer thickness was set as 0.035mm. Following the Figure6, when overhang angle smaller than 45 degree, the overhang could be stably manufactured, but it is hard to be manufactured when overhang angle larger than 63.4 degree. Therefore, in this article, the manufacturing direction of copper cash abacus and collapsible abacus were display at 30 degree of overhang angle, as shown in the Figure 8. When the overhang angle larger than 63.4 degree, the support structures were added. When the overhang angle between 45-63.4 degree, whether the support structure is necessary or not, can be freely adjusted according to the actual situation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002036_978-94-007-2069-5_11-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002036_978-94-007-2069-5_11-Figure2-1.png", + "caption": "Fig. 2 Mounting conditions (a) general case, (b) prototype design", + "texts": [ + " Torsion bar spring is mounted at a determined preload torque, and thus act on meshing pinion wheels in opposite directions. When transmissing a rotary motion, then under the direction of rotation of input shaft performance by either one or the other path of gearing [1]. Similar methods of backslash elimination are described by [2] and [3]. The gear by the Fig. 1a is closed loop of gears. Therefore, it is necessary in design to comply with geometric and kinematic constraints, which relate to axis distances and gear circumferential speeds. In general case (Fig. 2) must be satisfied condition for pitch diameters d32 d31 D d22 d12 d11 d21 : (1) Assuming the same module of all the gears, pitch diameters can be replaced by the number of teeth z32 z31 D z22 z12 z11 z21 : (2) The advantage of this type of gearing is high variability in dimensions and therefore large area for optimization of specific applications. For the prototype design and subsequent calculations was chosen case, when d11 D d12 D d31 D d32 D d1 and d21 D d22 D d4 D d2 (Fig. 2). Table 1 lists technical parameters of the gearbox, which are used for subsequent calculations. The gearing on Fig. 1a is a combined six-member mechanism (including frame). On Fig. 3 is release of each member. It should be determine 38 unknown forces and torques, torques Mtk (torsion-bar spring preload) a M2 (load on output shaft). are parameters of equation system. There are 30 equilibrium equations, the remaining eight equations resulting from the geometry of toothing (four meshes, for each one two equations)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure11-1.png", + "caption": "Figure 11 INPUT GEAR AND CUTTER", + "texts": [], + "surrounding_texts": [ + "This example considers a spur bevel gear set for motion transmission between intersecting axes to introduce the developed process. The tooth profile is a standard involute tooth profile. The nominal gear pair data is presented in Table 2 whereas the nominal cutter data is presented in Table 3. Figures 9 and 10 show the gear elements in mesh with the hyperboloidal cutter elements. No geometric, rating, or manufacturing data are generated." + ] + }, + { + "image_filename": "designv11_84_0002557_ever.2014.6844147-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002557_ever.2014.6844147-Figure1-1.png", + "caption": "Fig. 1: The common coaxial connected model of the dynamic test system", + "texts": [ + " In particular, when the shaft torsional stiffness of the motor, coupling and dynamic torque meter are not in the same order of magnitude, but their moment of inertia are in the same order of magnitude, the torque value of the load machine and dynamic torque meter are not the actual value of the tested motor. The correction of the experimental results should be give full consideration to the effects of flexible and inertia. The proposed model can calculate the actual torque of the tested motor during the dynamic test. In the finally, this paper also carry out an experiment to verified the correctness and accuracy of the model. II. COMMON MODEL OF DYNAMIC TEST SYSTEM The common coaxial connected model of the dynamic test system is shown in figure 1. Every part of the system from left to right is the shaft of tested motor, coupling 1, dynamic torque meter, coupling 2, the shaft of load machine. In this paper, a vertical plan which taken through the axis is used as a reference plane for the rotary system. The points both on the reference plane and two couplings are marked as point 1, point 2, point 3 and point 4. During the rotation, the angle between reference plane and the vertical line from points to the axis marked as \u03b81, \u03b82, \u03b83, and \u03b84", + " Its measured value is the actual torque value of the tested motor in steady state conditions. But in the process of actual dynamic test, the dynamic torque meter as the key part of the torque measuring apparatus can considered as a flexible by the principle [9] of rotor dynamics. In this case, the measured value of the dynamic torque meter is not the actual torque value of the tested motor. It is just the value of transmittable torque by the meter during the testing. III. THE MATHEMATICAL MODEL OF DYNAMIC TESTING SYSTEM In the coaxial connected system of figure 1, assuming coupling 1 is made by flexible material. The transmittable torque of coupling 1 in the dynamic test can get from [9] as follow: 2 2 1 2 2 1 2( - ) c ( - )T k \u03b8 \u03b8 \u03c9 \u03c9\u0394 = + (1) where k2 is the shaft torsional stiffness of coupling 1, c2 is damping coefficient of coupling 1, \u03b81 and \u03b82 is the angle between reference plane and the vertical line from point 1 and 2 to the axis, \u03c91 and \u03c92 are angular velocity of point 1 and 2. For most dynamic test system, the shaft torsional stiffness of the motors is relatively large and the shaft can be considered rigid. The rest of the system considered as flexible link. We suppose that the average angular acceleration of the two end points of flexible parts in the system can represent their acceleration. So based on the dynamic model of figure 1, the mathematical model was proposed by the rotor dynamics from right to left is given in follows: 4 4 5L dT T J dt \u03c9\u0394 \u2212 = (2) 34 3 4 4 2 dd dt dtT T J \u03c9\u03c9 + \u0394 \u2212 \u0394 = (3) 3 2 2 3 3 2 d d dt dtT T J \u03c9 \u03c9+ \u0394 \u2212 \u0394 = (4) 1 2 1 2 2 12e d d ddt dtT T J J dt \u03c9 \u03c9 \u03c9+ \u2212 \u0394 = + (5) 3 4 3 44 4 4 3 3 2 3 3 2 3 2 2 1 2 2 1 2 0 0 0 0c = 0 0 + c 0 0 0 0 c 0 0 T k T k T k \u03b8 \u03b8 \u03c9 \u03c9 \u03b8 \u03b8 \u03c9 \u03c9 \u03b8 \u03b8 \u03c9 \u03c9 \u2212 \u2212\u0394 \u23a1 \u23a4 \u23a1 \u23a4\u23a1 \u23a4 \u23a1 \u23a4 \u23a1 \u23a4 \u23a2 \u23a5 \u23a2 \u23a5\u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5\u0394 \u2212 \u2212\u23a2 \u23a5 \u23a2 \u23a5\u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5\u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5\u0394 \u2212 \u2212\u23a3 \u23a6 \u23a3 \u23a6 \u23a3 \u23a6\u23a3 \u23a6 \u23a3 \u23a6 (6) ( 1,2,3,4)i i d i dt \u03b8\u03c9 = = (7) where J1, J2, J3, J4 and J5 is the inertia of the tested motor, coupling 1, dynamic torque meter, coupling 2 and loading machine, \u03b83 and \u03b84 is the angle between reference plane and the vertical line from point 3 and 4 to the axis, \u03c93 and \u03c94 are angular velocity of point 3 and 4, k3 and k4 is shaft torsional stiffness of dynamic torque meter and coupling 2, c3 and c4 is damping coefficient of dynamic torque meter and coupling 2, \u0394T2, \u0394T3 and \u0394T4 is the transmittable torque of coupling 1, dynamic torque meter and coupling 2, Te and TL is the mechanical torque of the tested motor and load machine. When the test system worked on steady state, the angular velocity is a constant. 1 =0d dt \u03c9 (8) It can be drawn from the mathematical model: =e LT T (9) In the case of every part of figure 1 is made by rigid materials, the angular velocity of the entire shaft connected system is equal. So it is only to consider the effect of the moment inertia. 1 2 3 4\u03c9 \u03c9 \u03c9 \u03c9= = = (10) 1 1 2 3 4 5e L dT T J J J J J dt \u03c9\u2212 = + + + +\uff08 \uff09 (11) As can be seen from the above mathematical model, in order to accurately obtain the torque value of the tested motor during the test only by the value of torque meter is not enough. It should be know the material properties and moment inertia of every part, the angular velocity of point 1, 2, 3 and 4 should also be measured during the dynamic test" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002841_sta.2014.7086691-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002841_sta.2014.7086691-Figure1-1.png", + "caption": "Figure 1. :Dual star induction machine", + "texts": [ + " In addition, in order to reduce the size of the program and the simulation time, a connection matrix between six-phase inverter-machine was proposed. Experimental results was carried out on a prototype of a squirrel-cage double star induction machine of 0.5 kW supplied by VSI- inverter. The study encompasses the following sections. After the introduction. The model of DSIM (or six phase induction machine SPIM), is described in section 2. The fed of SPIM by a voltage source inverter are exposed in sections 3. Main results are discussed in section 4. The paper ends, of course, by a conclusion. II. NATIVE MATHEMATICAL MODEL As shown in Fig. 1, the machine has two stator windings sets (a1, b1, c1) and (a2, b2, c2) spatially shifted by \u03b3, with isolated neutral points and an equivalent three-phase squirrelcage rotor. \u03b8\u2019 is the angle shift between the rotor and the star 1, and \u03b8\u2019 - \u03b3 is the one between the rotor and the star 2. For this machine, we adopt the following assumptions: - Stator windings are sinusoidally distributed, - Windings are identical within each three phase set. - Magnetic saturation was neglected. 978-1-4799-5907-5/14/$31" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003581_detc2014-34444-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003581_detc2014-34444-Figure1-1.png", + "caption": "Figure 1: The nodal coordinates of the gradient deficient beam element", + "texts": [], + "surrounding_texts": [ + "As shown in Fig. (1), the vector of nodal coordinates of the proposed gradient deficient beam element is defined as ( ) ( ) ( ) ( ) ( ) ( )[ ]TB y B x BA y A x At )( TTTTTT rrrrrre = (1) The position vector of any point in the proposed ANCF gradient deficient beam element can be defined as )( ),(),,( tyxtyx eSr = (2) where e is the vector of nodal coordinates that is defined by Eq. (1) and S is the shape functions matrix that can be written as [ ]IIIIIIS 654321 ),( SSSSSSyx = (3) where I is a 3x3 unit matrix and S1, S2, S3, S4, S5, S6 are the shape functions that are defined as ( ) ( ) ( ) =+\u2212= \u2212=\u2212= +\u2212=+\u2212= , , ,23 ,1 ,2 ,231 6 32 5 32 43 32 2 32 1 \u03b7\u03be\u03be\u03be \u03be\u03be\u03be\u03b7 \u03be\u03be\u03be\u03be\u03be lSlS SlS lSS (4) where lx=\u03be , ly=\u03b7 and l is the element length in the reference configuration. The equation of motion of the proposed ANCF gradient deficient beam element can be written as (Shabana 2008) 0QQeM =\u2212+ es&& (5) 2 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/31/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use where M is the constant mass matrix, eQ is the external force and sQ is the elastic force that will be discussed in details in the following section. The mass matrix of the proposed ANCF element can be obtained from the kinetic energy of the element that is defined as eMeeSSerr &&&&&& T V TTT V dVdVT 2 1 2 1 2 1 0 00 === \u222b\u222b \u03c1\u03c1 (6) where V is the element volume, V0 is the initial element volume, \u03c1 is the element density and 0\u03c1 is the initial element density. From Eq. 6, the mass matrix is defined as (Shabana 2008) \u222b \u222b\u222b \u2212 == 2/ 2/ 0 000 0 b b l T V T dydxhdV SSSSM \u03c1\u03c1 (7) where b is the element dimension in the y direction of the element coordinate system, h is the element initial thickness, \u03c1 is the element density. Like most of ANCF elements, the proposed ANCF element has constant mass matrix and zero forces results from quadratic terms of the velocity vectors. The external force vector of an ANCF gradient deficient beam element proposed in this work can be obtained from (Shabana 2008) ( )FxSQ p T e = (8) where F is the force vector applied at point that has the position xp. An example of the external force is a uniformly distributed pressure on the surface that contains the two vectors rx and ry; in this case the external force can be written as ( )\u222b \u222b \u2212 = 2/ 2/ 0 , b b l T e dydxyxp nSQ (9) where p is the pressure and n is the unit vector normal to rx and ry i.e. yxyx rrrrn \u00d7\u00d7= ." + ] + }, + { + "image_filename": "designv11_84_0001126_2015-01-2515-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001126_2015-01-2515-Figure8-1.png", + "caption": "Figure 8. Unheaded rivet vertical inject injector cutaway view", + "texts": [ + " The injector purges with the injector positioned in front of the headstone, so purged fasteners contact the headstone before falling harmlessly to the floor. With good injector reliability, purge functions will only be used in extremely rare cases. The position of the fastener in the guide chute in the three injectors discussed already is set by the fastener head. The use of a parallel gripper has also been successfully implemented for use in injecting unheaded fasteners. In this situation, since no head feature is available to index the fastener, the fasteners must be indexed off of their tails. Figure 8 shows a cutaway view of the unheaded rivet injector. The fasteners exit the tube at speed and bounce off the stopper until they settle into the guide chute with their tail pressed against the stopper. There are three different stopper sizes, which index the tails of the fastener in different spots depending on the fastener grip range being fed. Another challenge of feeding different diameters of fastener with the same hardware is pushing the fastener the proper distance into the feed nose. To make sure that the proper diameter fastener gets pushed the proper distance, the tube end fittings double as a hard stop for the pusher\u2019s up position. When using the tube dedicated to a certain diameter, the pusher\u2019s forward stroke is set by that tube\u2019s end fitting, ensuring that the proper hard stop is used for the proper fastener. This can be seen by looking closely at figure 8. Horizontal Inject Version The parallel gripper injector is implemented in vertical axis riveting machines. In Figure 10 you can see that the designer chose a rotary version of the parallel gripper to make the system more compact. You can see where a replaceable urethane catcher's mitt is incorporated into the track. This feature is easily replaced. Multiple grips and diameters can be fed down the same injector. Due to the synchronized motion the fastener is guided to the center of the fingers. By reversing the air as shown in Figure 10 the guide chutes are opened wide to allow any fasteners or FOD to fall free of the machine and be captured by a reject removal component" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001757_cjme.2015.0119.050-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001757_cjme.2015.0119.050-Figure1-1.png", + "caption": "Fig. 1. Points with the extreme normal fitting errors", + "texts": [ + " If a straight line is taken to fit R(i) P , the normal distance from a point in R(i) P to the fitting line is a measure of the fitting error. A straight line, determined by minimizing the maximum fitting error, is defined as the saddle line. The corresponding minimal maximum fitting error is defined as the saddle line error. The saddle line can be sought from the min-max principle or the saddle point programming. A point in R(i) P with the maximum fitting error is called as the saddle line fitting-point, or the characteristic point. Points P(1), P(2) and P(3) in Fig. 1 are three characteristic points. Based on the saddle point programming, the saddle line of R(i) P is determined by a finite number of characteristic points. Once the characteristic points are found, the saddle line error can be obtained through the analytic geometry method. The properties of the saddle line error can be investigated further. Point P of a rigid body has the coordinates (xPm, yPm) in the moving Cartesian coordinate system {Om; im, jm}. Let (x(i) P , y(i) P ) be a discrete displacement vector of P" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002717_1.4882556-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002717_1.4882556-Figure6-1.png", + "caption": "FIGURE 6. (a) The fundamental unit of the crease pattern in FIGURE 5(b), and (b) the fundamental unit of the folded pattern in FIGURE 5(d)", + "texts": [ + " Note that { This amounts to assigning colors to three flat foldability types which result in the mv-assignment shown in FIGURE5(b). The assignment of flat foldability types to the colors is given in FIGURE 5(c). The subgroup that fixes the colors in the given coloring is . This group is also the symmetry group of the crease pattern with mv-assignment shown in FIGURE 5(b). The mv-assignment will give rise to the folded pattern given in FIGURE 5(d). The fundamental unit ofthe crease pattern with mv-assignment shown in FIGURE 5(b)and the corresponding fundamental unit of the folded pattern shown in FIGURE 5(d) is given in FIGURE 6(a) and (b) respectively. In this example, the group that keeps the folded pattern invariant is a plane crystallographic group of type and is a subgroup of the symmetry group of the folded pattern with fundamental unit . Considering another correspondence of colors and flat foldability types (FIGURE 7), which is different from the correspondence in FIGURE 5(c), leads to another crystallographic flat origami (FIGURE 3(c)). In this paper, we considered an unassigned crease pattern with symmetry group (plane crystallographic group) derived from an Archimedean tiling by the hinged tiling method" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003721_detc2014-34297-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003721_detc2014-34297-Figure11-1.png", + "caption": "FIGURE 11. Parallel platform where the mobile platform generates a subspace, xu\u03021 \u2295 ru\u03022 , with respect to the fixed platform.", + "texts": [ + " In a similar way, the direction associated to the second subalgebras or subspaces of each leg must be parallel. With this single condition, Step 3 of the synthesis process is complete. Examples Two examples of the synthesis of parallel platforms where the 11 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use velocity states of the mobile platform with respect to the fixed platform, V m/ f generates a subspace, xu\u03021 \u2295 ru\u03022 are shown here. 1. Consider the parallel platform shown in Figure 11. Leg 1 generates a subspace xu\u03021 \u2295 hu\u03022,p, corresponding to case 2, Table 3. Leg 2 generates the subspace xu\u03021 \u2295 ru\u03022 , corresponding to case 1, Table 3. Leg 3 generates the subspace gu\u03021 \u2295Sgu\u03022 , corresponding to case 3, Table 3. Moreover, the conditions for forming a platform were also used, that is, the directions associated to xu\u03021 , in legs 1 and 2, and gu\u03021 , in leg 3, are parallel, while the directions associated to hu\u03022,p, ru\u03022 and Sgu\u03022 , in legs 1, 2 and 3, respectively, are also parallel" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001567_icma.2015.7237783-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001567_icma.2015.7237783-Figure10-1.png", + "caption": "Fig. 10 Kinematic diagram of eyeball mechanism", + "texts": [ + " 9, coordinates {1}, {2}, {4} are base coordinates which are fixed on head skeleton. The origin of coordinate {1}, {2} are the center of the motor rockers. Its revolve angle are 1 and 4 respectively. The origin of coordinate {4} are the center of eyeball. Coordinate {3}, {3\u2019} are fixed on the eyeball and revolve with it. Coordinates {3} revolves around z axis in reference to coordinate {2}. Its revolve angle is 2. Coordinate {3\u2019} revolves around x axis in reference to coordinate {3}. Its revolve angle is 3. Fig. 10 is the kinematic diagram of eyeball mechanism. It is a RSSR space linkage mechanism. During eyeball\u2019s movement, the length of connecting rod CD and AB are constant. Based on this characteristic, coordinates\u2019 relation between points C and D, points A and B could be deduced. The coordinates of point A in coordinate system {1} and point B in {2} are: [ ]1 1 112.5cos 12.5sin 2.57 T AO \u03b1 \u03b1\u2212 \u2212= \u2212 (1) 2 2 26cos 6sin 4.07 TO B \u03b1 \u03b1= \u2212 \u2212 \u2212 (2) Make coordinate transformation: 1 2 1 2 10.6 12.5 cos 6 cos 78 12" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003051_amm.394.245-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003051_amm.394.245-Figure1-1.png", + "caption": "Fig. 1 Structure model of H1250G spiral bevel gear grinding machine", + "texts": [ + " This paper puts forward on-machine measuring principle of tooth profile errors and the processing method of error datum based on the generating principle of spiral bevel gears and the structure model of H1250G spiral bevel gear grinding machine. Moreover, based on the AutoCAD development platform and using the VBA programming language embedded in AutoCAD, the on-machine measuring system was developed. The measuring process is simulated by Boolean operation between solid model of gear and probe and then the tooth profile errors of actual tooth profile relative to theoretical ones was obtained. Fig.1 is the structure model of H1250G spiral bevel gear grinding machine. It consists of four linear axes and four rotational axes. Axis A is the rotational axis of workpiece and axis C is the rotational axis of grinding wheel. Axis B is the rotational axis for adjusting the angle between workpiece axis and grinding wheel axis. Axis D the rotational axis of diamond dressing roller. Linear axis X, Y and Z forms Cartesian coordinates system. Linear axis W realizes the movement of on-machine measuring mechanism" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002837_dscc2013-3851-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002837_dscc2013-3851-Figure8-1.png", + "caption": "Figure 8: One wheel of the Segway colliding with an obstacle.", + "texts": [ + " This gives an indication of the severity of injury that may be suffered because the energy must be dissipated during the accident. Hazard 6 will be considered as an example. If the machine makes an unexpected motion that drives one of the wheels into a relatively small obstacle, then the device usually pitches forward and turns toward the obstacle. This can launch the rider off the front of the vehicle. Actually, because the machine turns toward the obstacle, but the rider\u2019s momentum carries them forward (in their original reference frame), the rider falls at an angle relative to the machine, as illustrated in Figure 8. Figure 8a shows the Segway traveling towards the obstacle. Figure 8b shows the right wheel striking the obstacle, and the Segway turning towards it, while the rider\u2019s momentum carries him along the original path of the Segway. Figure 8b also shows how the compounding factors of the wheels and handlebar block the legs of the falling rider when this hazard occurs. The objective of this analysis is to estimate the rider\u2019s total energy at the onset of the accident. The total energy of the rider can be defined as: E = T +V, (1) 7 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use where T is the rider\u2019s kinetic energy and V is the rider\u2019s potential energy" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001429_9781118886397.ch14-Figure14.9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001429_9781118886397.ch14-Figure14.9-1.png", + "caption": "Figure 14.9 Derivation of forces on spacers. (a) general view of subspan and (b) midspan cross section of subspan (Source: EPRI (2006)).", + "texts": [ + "29) \ud835\udeffL can be expressed as \ud835\udeffL = 8D2 2 3S \u2212 8D2 1 3S = 8 3S (\ud835\udeffD \u2212 2D1) (14.30) where \ud835\udeffD = D1 \u2212 D2 (14.31) The change in tension can be expressed in terms of the change of vertical dis- placement as follows: \ud835\udeffH = 8aE 3SL1 \ud835\udeffD(\ud835\udeffD \u2212 2D1) (14.32) The electromagnetic forces during a phase-to-phase fault will act to move the conductors apart, placing phase-to-phase spacers in tension. After the fault is cleared, the conductors will swing together, compressing the spacers. These forces can be analyzed using the diagram of Figure 14.9. Using the previously defined terminology, for any subspan swing angle, \ud835\udf03, Fc = F sin \ud835\udf03 + W cos \ud835\udf03 (14.33) Fspacer = 2Fc cos \ud835\udf03 = 2(F sin \ud835\udf03 + W cos \ud835\udf03) sin \ud835\udf03 (14.34) and sin \ud835\udf03 = zt D (14.35) cos \ud835\udf03 = \u221a 1 \u2212 ( zt D )2 (14.36) In the simple case where Fout reduces to zero before maximum swing is reached (i.e., the fault clears), Fspacer = 2SW zt D \u221a 1 \u2212 ( zt D )2 (14.37) Since D = WS2 8H (14.38) the maximum spacer force is Fspacer = 16H zt S \u221a 1 \u2212 ( zt D )2 \u2245 16Hzt S (14.39) This will be the maximum spacer force in both tension and compression for the usual case where the fault has cleared before maximum conductor deflection has occurred" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003423_humanoids.2014.7041380-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003423_humanoids.2014.7041380-Figure2-1.png", + "caption": "Fig. 2. Link configuration of the robot", + "texts": [ + " The position of joint j with respect to the origin of the world coordinate frame can be described by r\u3008j\u3009/0, while its axis orientation by the unit vector aj and its joint variable by \u03b8j . The link j has a mass distribution characterized by the mass mj , the position of its center of mass described by rc\u3008j\u3009/0 and the tensor of inercia with respect to this point (and described in the world frame 0) Ic,j . Additionally, the joint j also drives all the link structure connected to the link j. Let us denote as m\u0303j , r\u0303c\u3008j\u3009/0 and I\u0303c,j the mass, the position of the center of mass and the inertia tensor of all this link structure driven by the joint j. See Fig. 2. Then, the additional linear and angular momenta with respect to the origin of the world coordinate frame for the robot which are yielded by \u03b8\u0307j (Pj,0 and Lj,0 respectively) can be calculated as Pj,0 = m\u0303jaj \u03b8\u0307j \u00d7 ( r\u0303c\u3008j\u3009/0 \u2212 r\u3008j\u3009/0 ) = Mj,0\u03b8\u0307j , (11) Lj,0 = ( r\u0303c\u3008j\u3009/0 \u00d7Mj,0 + I\u0303c,jaj ) \u03b8\u0307j = Hj,0\u03b8\u0307j . (12) Such that, M0 = [ M1,0 M2,0 \u00b7 \u00b7 \u00b7 Mn,0 ] , (13) H0 = [ H1,0 H2,0 \u00b7 \u00b7 \u00b7 Hn,0 ] , (14) Hc = H0 \u2212 r\u0302c/0M0. (15) Finally, let us consider (9) and substitute (15) into it in order to get a final expression for Hp [7]: Hp = r\u0302c/pMc + ( H0 \u2212 r\u0302c/0M0 ) = H0 \u2212 r\u0302p/0M0, (16) given that M0 = Mc as Pj,0 = Pj,c" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003752_icep.2014.6826661-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003752_icep.2014.6826661-Figure8-1.png", + "caption": "Fig. 8 The model of paste solder and board in the analysis", + "texts": [ + " Most of laser energy (about 95%) is energy loss. The factors of energy loss are the reflection of laser light at a solder surface, the convective heat transfer from the solder to air and the thermal conduction to the board. If the losses can be evaluated quantitatively, we have a possibility to obtain the optimal laser condition. Therefore, to evaluate the energy loss, we analyze the temperature distribution of laser soldering. In this paper, as the first step of the analysis, the tendency of the temperature rise of the board is discussed. Figure 8 shows the analysis model. Because it is difficult to simulate laser heating, we set the fixed temperature at 220 \u2103, which is the melting temperature of the solder, at the under face of the solder. Therefore, reflection of laser and the melting phenomenon of the solder are not considered. The shape of the paste solder is same as the disc shape at the experiment. The size of board is 40 mm \u00d7 40 mm \u00d7 0.5 mm. The materials of the board are the same in the experiment (glass epoxy, stainless steel and copper)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001184_1.3663086-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001184_1.3663086-Figure1-1.png", + "caption": "FIGURE 1. (a) Single-mass, crank-and-rod driven shaking conveyer, (b) the coercive force applied on the trough.", + "texts": [ + " These authors analyzed, the physical model of the ideal vibrating system consists of linear spring and sinusoidal excitation (ideal source) [4], the physical model of the vibrating system consists of linear spring and nonideal source [5, 6] and the physical model of the vibrating system consists of a cubic nonlinear spring and nonideal source [7]. In this study, vibrational conveyers are constituted by a trough and elastic stands of equal length connected to trough inclinedly. Forced vibration motion driven by the crank-and-rod mechanism is given to this system \u201cFIGURE 1\u201d. Crank-and-rod mechanism with elastic connecting rod can also be used, an elastic component added to the operation mechanism slowly increases the amplitude of the system from a low International Congress on Advances in Applied Physics and Materials Science AIP Conf. Proc. 1400, 60-65 (2011); doi: 10.1063/1.3663086 \u00a9 2011 American Institute of Physics 978-0-7354-0971-2/$30.00 60 Downloaded 24 Sep 2013 to 146.232.129.75. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003091_amm.86.730-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003091_amm.86.730-Figure2-1.png", + "caption": "Fig. 2 Third-order elliptical gear solid model", + "texts": [ + " The next is completion shape of the first \u201c i \u201d tooth by creating chamfer dR and fillets gR . Because the radius of curvature on the section curve is not equal and the radius of equivalent gear is not equal either, the tooth profile of each tooth is not necessarily the same, as the above steps should be made one by one to the shape of the tooth. But the tooth of the long axis is symmetrical, so it just made the half of one side. The upper part of teeth on the long axis through the command of [mirror] can get another half of the teeth, and then complete all of the tooth shape (Fig.2). This modeling method is also applicable to other non-circular. Wire-cutting Set. Wire-cutting function modules[9]. which provided by Pro/E can be machining simulation of non-circular gear. The process is that work-piece and solid modeling of non-circular gear which created transfer to manufacturing interface. Then selecting two-axis WEDM machine, reasonable arrange the processing operation of work-piece, determine the cutting volume of processing surface, cutting feed rate, location plane and other technological parameters by rough processing and finish processing" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure26-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure26-1.png", + "caption": "Figure 26 Fork compression mode shape (see online version for colours)", + "texts": [ + " The simulation involved light braking after 4.5 seconds and increased to maximum braking after 5 seconds. The braking compressed the non-linear front springs and increased the stresses in the fork legs. The braking increased the natural frequency of mode number 2, which was related to the fork suspension unit. The frequency was initially 3.53 Hz and increased to a maximum of 4.51 Hz because of the spring compression and stress stiffening during braking. The shape of mode number 2 is illustrated in Figure 26. The Fedem Ilmor modelling and correlation has produced a virtual prototype that includes many of the dynamical properties and effects of the real motorbike during racing. However, a simulation model is always a trade-off between efficiency and accuracy, and not all effects could be included. In this section, some of the effects and limitations of the model are discussed. This model is not an exact geometric representation of the real Ilmor bike because many of the bike parameters were not available" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001914_s11740-014-0582-7-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001914_s11740-014-0582-7-Figure4-1.png", + "caption": "Fig. 4 Setting of the experimental test rig", + "texts": [ + " The three phases are in star-connection and the coils of each phase are connected in series. Magnetic flux direction for the two windings of one phase is contrariwise. Due to the central position of the w windings, the reluctance of this phase is lower than the reluctances of the two other phases. So it is expected, that the inductance of the w-phase is higher. The number of windings of the z-axis is 140 and of the x-axis 120. 3.2 Experimental setup For the analysis of the planar drive, an experimental test rig, shown in Fig. 4, was designed. Regarded to a turning application, the axes of the cross table are chosen to z and x. The z-axis has a traveling distance of 220 mm, the x-axis has 22 mm. Conventional profile rail guides are used. The position is measured by absolute glass scales (Heidenhain LC 483). The maximum size of the workpiece is 250 9 250 mm. In order to measure the disturbance- and feed-forces, a six-axes force-moment dynamometer is placed between the cross-table and the primary part. At this position the forces of the drive are directly measured and not biased by the friction of the guides" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002315_j.proeng.2011.05.102-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002315_j.proeng.2011.05.102-Figure2-1.png", + "caption": "Fig. 2. Seam orientation (plan view) [5]", + "texts": [ + " The cylinder was made of PVC material and used some filler to make it structurally rigid. The cylinder was vertically supported on a six components transducer (type JR-3) had a sensitivity of 0.05% over a range of 0 to 200 N as shown in Figure 1. The aerodynamic forces and their moments were measured for a range of Reynolds numbers based on cylinder diameter and varied wind tunnel air speeds (from 10 km/h to 130 km/h with an increment of 10 km/h). Each test was conducted as a function of swimsuit\u2019s seam positions (see Figure 2). As mentioned earlier, the RMIT Industrial Wind Tunnel was used to measure the aerodynamic properties of swimsuit fabrics. The tunnel is a closed return circuit wind tunnel with a turntable to simulate the cross wind effects. The maximum speed of the tunnel is approximately 150 km/h (Re= 3.06\u00d7105). The rectangular test section dimensions are 3 meters wide, 2 meters high and 9 meters long, and the tunnel\u2019s cross sectional area is 6 square meters. The tunnel was calibrated before and after conducting the experiments and air speeds inside the wind tunnel were measured with a modified National Physical Laboratory (NPL) ellipsoidal head Pitot-Static tube (located at the entry of the test section) which was connected through flexible tubing with the Baratron\u00ae pressure sensor made by MKS Instruments, USA" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002384_2011-01-1424-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002384_2011-01-1424-Figure4-1.png", + "caption": "Fig. 4. Roller Shroud Assemblies", + "texts": [ + " The chilled water flow rate was controlled via an electronic PID controlled valve, to maintain the desired traction fluid cooler out supply temperature at 80degC for the durability tests and 90 to 100degC for the traction fluid high temperature life testing. Manually set needle control valves in the hydraulic circuit controlled the flow rate of the traction fluid used to cool the variator. This cooling fluid was feed via the carriage and its piston into shrouds around the roller via two 1.5mm diameter \u2018jets\u2019. Figure 4 shows the roller shroud and carriage assemblies. Each roller was fitted with a rubbing K-type thermocouple just after the disc/roller contact, such that the thermocouple measured the roller crown metal temperature. In order to assess traction fluid durability, a three part test program was developed. The first part investigated the influence of very high variator component temperature on fluid life. The second part investigated the influence of longer-term contact shear combined with high bulk oil temperature on fluid life" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure4.13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure4.13-1.png", + "caption": "Fig. 4.13. CAD model of the transmission system", + "texts": [], + "surrounding_texts": [ + "A second type of anthropomorphic haptic interface [15] is an exoskeleton structure conceived to be located on the dorsal part of the operator\u2019s hand. It consists of 4 parallel exoskeleton structures covering and connected to the 4 fingers and exerting forces to the phalanges (little finger excluded). Each finger exoskeleton, which kinematic scheme is represented in Fig. 4.15, consists of four links connected by revolute joints disposed as the joints of each finger. For each joint of the finger exoskeleton, the joint axis has been designed in order to approximate the instantaneous position of the flexion-extension axis during operation. At the metacarpo-phalangeal joint a passive abduction-adduction movement has been also integrated. The actuation system for each finger exoskeleton is based on three DC servomotors and associated tendon transmission systems. Each tendon is pulling on the middle point of each phalanx of the finger in order to execute the extension movement; at each joint, the flexion movement is obtained by a passive torsion spring integrated in the joint axis. The three motors are located on a cantilever structure 4 Exoskeletons as Man-Machine Interface Systems 73 fixed with the base frame of each finger exoskeleton. Rotation sensors, based on conductive plastics technology, are integrated at each joint, while force sensors, capable of recording the interaction force between the exoskeleton structure and each phalanx, are located directly on the dorsal surface of each phalanx link. A picture of the hand exoskeleton is given in Fig. 4.16. The kinematic structure of the thumb exoskeleton is slightly different to the one of the other fingers. In particular, the cantilever supporting the three motors of the thumb assumes a completely different aspect with respect to the one of the other fingers. One of the critical factors encountered during the design of the system has been that of obtaining a system 74 M. Bergamasco, A. Frisoli, and C.A. Avizzano possessing limited weight and volumes, in such a way to allow a good manoeuvrability of the hand. In terms of the mechanical performances of the hand exoskeleton, a maximum extension force of 0.3 N has been obtained, being the force sensor range of -0.5N \u2013 3.0N. Force resolution is 0.0025N, while the force feedback bandwidth is 0.5 Hz with an angular displacement of 90\u25e6 for all the 3 DOF." + ] + }, + { + "image_filename": "designv11_84_0003575_12.2018296-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003575_12.2018296-Figure6-1.png", + "caption": "Figure 6 airplane", + "texts": [], + "surrounding_texts": [ + "The assumption made by the SMS architecture is that each sensor connected to the system communicates using its own proprietary format for C2. This presents a challenge in providing C2 level access to a common system. The SMS addresses this challenge by abstracting the sensor interfaces away from the core sensor control code. In this way, sensor management is broken up into two pieces \u2013 the SIM and sensor interfaces, which are sensor specific." + ] + }, + { + "image_filename": "designv11_84_0000947_978-3-540-73958-6_5-Figure5.8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000947_978-3-540-73958-6_5-Figure5.8-1.png", + "caption": "Fig. 5.8. Helicopter Coordinate System (HCS)", + "texts": [ + " The model uses two different coordinate systems: \u2022 The base coordinate system (BCS) serves as the fixed Newtonian reference frame for navigation way-points. It is defined by the directions of geographic North (xBCS) and vertical up (zBCS), see Fig. 5.7. The origin of the BCS can be chosen by the user, one obvious choice being the position of the GPS reference antenna. \u2022 The helicopter coordinate system (HCS) is fixed to the helicopter and is defined as xHCS for \u201cforward\u201d and zHCS for \u201crotor axis up\u201d, the origin in the rotor axis and close to the vehicle\u2019s center of mass. See Fig. 5.8. Unless otherwise noted, all coordinates throughout this section will refer to the HCS. The state x of the helicopter describes the location, velocity and all the other values (see Table 5.1) that change over time and which are needed for calculating future states. The update of the state can be calculated using the standard system equation from a function f of the state and the input u to the system: x\u0307 = f(x,u) (5.1) The derivation of f is described in this section. For the simulation of movement, kinematic equations are needed that describe the alteration of position and velocity (called the state of the helicopter) with respect to the current state and applied forces" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure19-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure19-1.png", + "caption": "Figure 19 Engine control (see online version for colours)", + "texts": [ + " The torque pair at the bottom of the headstock (number 3 and 4) were the steering command and its reaction torque, respectively (see Figure 18). The torques corresponded to the steering and frame reaction forces that would be applied to the handlebars by the rider in a real situation. The complete steering control system is shown in Figure 17. The engine performance was also modelled by a control system. The specified engine torque (as a function of engine speed) was used as an input function to the control system. The engine characteristics were based on measurements from the real bike. Figure 19 shows the underlying concept of engine torque control. The system constantly measured the speed of the crankshaft and applied the maximum available torque at this speed, which in turn resulted in a change in the crankshaft speed. The system operated in a constant loop until the maximum crankshaft speed was reached (approx. 2,000 [rad/sec] / 19,100 rpm). The real engine does not deliver an output torque below 2,865 rpm. An initial torque of 30 Nm was therefore applied below this limit to jump start the simulation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002045_dscc2013-4034-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002045_dscc2013-4034-Figure3-1.png", + "caption": "FIGURE 3: PHANTOM TRACK GENERATED THROUGH THE COOPERATION OF 3 UAVS, FOR DIFFERENT INITIAL VALUES OF A UAV.", + "texts": [ + " First, after configuring the simulation, we try to move the UAVs toward and away from the radar to see how the initial configuration of the system affects the consensus of the system, and up to what extent the system tolerates uncertainty of the initial states of the agents. Second, in an odd fashion, we will change the final destination of the phantom 5 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use somewhere along its intended track. As we showed in this paper, it is expected that the UAVs continue to maintain a feasible solution for the phantom track generation. According to Figure 3, evidently, changing the initial position of one of the agents does not affect the phantom track. Although it has not been proven, this could show how robust the phantom track is to such uncertainties. Figure 4 shows the speed and acceleration inputs of the UAVs for the trajectory of Figure 3. It is quite reasonable that after the phantom track straightened, the UAVs\u2019 speed increased and they maintain their direction. FIGURE 4: ACCELERATION AND SPEED PROFILE OF THE UAVS Figure 5 illustrates that when the destination point is changed, the phantom track still remains consistent applying the feasible control, although it could not reach the new destination. This happens since: first, the new destination point is chosen too far, and second, the change happens quite late, so the limitation considered on the actuators of the phantom do not allow it to keep the team goal anymore" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002393_amm.87.123-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002393_amm.87.123-Figure1-1.png", + "caption": "Figure 1. Architecture and physical model of the positioning system", + "texts": [ + "113, Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), Lausanne, Switzerland-22/03/15,22:52:30) In this paper, first, the long-range ultra-precise positioning system and the dynamic model of the vibration are introduced. The second section constructs the active suppression system based on BPNNPID, the effect of which is further validated by simulation and experiment implemented in the Section 3. Finally, results and analysis are conducted in the conclusion. The Physical Model. The long-range ultra-precise positioning system is composed of a coarse stage and a fine stage, as shown in Fig. 1(a). The coarse stage, aiming to realize a long range, is driven by a servo motor along the horizontal sliding guides. The transmission system includes a gear box, a turbine worm and a feed screw nut pair. The fine stage, hanging inside the coarse stage for nano-scale positioning, utilizes a piezoelectric actuator for its superior performance in high-accuracy motion control. The piezoelectric actuator is fixed between the coarse and fine stages by two preloaded springs as shown in Fig. 1(a). Dynamics of Vibration. Fig. 1(b) shows the model of the positioning system and a summary of symbols used are as follows: m1, m2, the masses of the coarse and fine stage; u1(t), u2(t), the coarse and fine inputs; y1(t), y2(t), the outputs of the two stages. Moreover, it is assumed that k1 and b1 are the equivalent stiffness and equivalent damping coefficient of the transmission that connect to the coarse stage, while k2 and b2 are the same for the connections between the two stages. It is also assumed that vibration due to motion of transmission components, collisions and slip-stick effect of friction is part of u1(t), and vibration caused by the parameter uncertainties contributes to both u1(t) and u2(t)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003678_1.4024704-Figure15-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003678_1.4024704-Figure15-1.png", + "caption": "Fig. 15 Total deformation (in meters) of the workpiece. Welding away from fixed face AA0B0B.", + "texts": [ + ", there is no deformation normal to this face. For the first constraint configuration, face AA0B0B is fixed. The other end of the workpiece is free to deflect. Other faces are free to deform. This configuration corresponds to welding away from the fixed face. Figures 12\u201314 highlight the x-, y-, and z-directional deformation of the workpiece. Welding parameters are: welding current 150 A, arc length 3 mm, welding speed 2.5 mm/s, and 100 ppm of oxygen. The total deformation of the workpiece is shown in Fig. 15. For the second constraint configuration, face CC0D0D is fixed. The other end of the workpiece is free to deflect. Other faces are free to deform. This configuration corresponds to welding toward the fixed face. Figures 16\u201318 highlight the x-, y-, and z-directional deformation of the workpiece. Welding parameters are: welding current 150 A, arc length 3 mm, welding speed 2.5 mm/s, and 100 ppm of oxygen. The total deformation of the workpiece is shown in Fig. 19. These results show that the free end deflection can be minimized when welding toward the fixed end" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003482_s207510871501006x-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003482_s207510871501006x-Figure1-1.png", + "caption": "Fig. 1. Formation of the missile control commands in the Cartesian and polar coordinate systems.", + "texts": [ + " The paper proposes a method for mis sile control in the beam in one coordinate, the radius of missile deviation \u03c1 from the beam center [7]. As a result, the control equipment becomes simpler and more reliable. There is no need in a gyro coordinator and there is no \u2018intertwining\u2019 of the coor dinate systems. METHOD DESCRIPTION Consider the method of designing an MCS in one coordinate by an example of a simplest diagram. In a traditional system, the control command consists of the commands in the vertical and horizontal channels (Fig. 1), which is equivalent to two com 2 2( )Y Z\u03c1 = + K YK ZK DOI: 10.1134/S207510871501006X GYROSCOPY AND NAVIGATION Vol. 6 No. 1 2015 MISSILE CONTROL IN THE POLAR COORDINATE SYSTEM 67 mands in the polar coordinate system, the \u2018normal\u2019 and \u2018tangential\u2019 ones. The formation of only one command provides the missile motion towards the beam axis O because its value is almost the same as the size of the total com mand K. The lack of command providing damping of the missile tangential motion under disturbances changes the nature of both the transition and entire guidance processes" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002559_tasc.2014.2388152-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002559_tasc.2014.2388152-Figure1-1.png", + "caption": "Fig. 1. Fabricated HTS coils: (a) Type I coil, (b) Type II coil, (c) Type III coil, and (d) Type IV coil.", + "texts": [ + "12 mm and 0.10 mm, respectively, which is thicker than Kapton tape of 0.04 mm. Type IV is every turn-to-turn insulated coil with Kapton tape. All of the HTS coils were wound 40 turns on a G10 bobbin of 65 mm diameter. Differences of the inductance and length between the four types of the HTS coils were caused by various thickness of the inserted materials. The self-field critical currents measured with 1 \u03bcV/cm criterion were 82 A, 86 A, 80 A, 74 A, respectively in a bath of liquid nitrogen at 77 K. Fig. 1 shows pictures of the fabricated four types of the HTS coils. Fig. 2 indicates an experimental jig and a BSCCO coil to excite the alternating field to the HTS coils. The BSCCO coil was fabricated as double pancake winding structure having a total of 244 turns and a measured self-field critical current at 77 K of 49 A. Table II shows detailed specifications of the virtual armature coil. The gap distance between the armature coil and the HTS field coil was fixed at 30 mm. III. EXPERIMENTAL RESULT A" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003236_csss.2011.5972150-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003236_csss.2011.5972150-Figure2-1.png", + "caption": "Figure 2. The corresponding 5R single loop mechanism", + "texts": [ + " 1, where {xi, zi} (i=1-6) are all unit vectors, ai, li and twist angle i, i+1 are known structural parameters, p and (z6, x6) are the given position and orientation of end effectors respectively. The inverse kinematic problem of the general 5R robot can be reduced to determine rotary angles i (i=1-5). Step 1 Connecting the 1st and the 5th pairs by common perpendicular line (a7x7) of their joint axes z1 and z5, we can convert the general 5R robot to its corresponding 5R single loop mechanism, as shown in Fig. 2. Obviously, the auxiliary structural parameters (lc1, lc5, d1, d5 and a7x7) can be readily determined, where (l5+lc5) is the mutual perpendicular distance (or offset) between successive links a4x4 and a7x7, and lc1 is the offset of a7x7 and a1x1, d1 and d5 represent the right-handrotation angle from x0 to x7 about z1 and the angle from x5 to x7 about z5 respectively. Here, c1 and c5 are rotary variables of the 1st and the 5th pairs. Step 2 Establish loop vector equations, and derive a set of triangular equations" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002155_00368791311292765-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002155_00368791311292765-Figure3-1.png", + "caption": "Figure 3 Noncircular two lobe journal bearing geometry and coordinate axes", + "texts": [ + "eywords Neural nets, Lubrication, Gas bearings, Noncircular lobed bearing, Neural network Paper type Research paper Bmn \u00bc Bmn Pa R2= Cm v\u00f0 \u00de, gas film damping coefficients (Ns/m); m, n \u00bc x, y C\u0304 \u00bc conventional radial clearance (m) Cm \u00bc minor clearance when journal and bearing geometric centers are coincident (m) e\u0304 \u00bc bearing eccentricity ( o oj ; Figure 3) F\u0304 \u00bc F Pa R2, film force on the journal (N) h\u0304 \u00bc h Cm, film thickness (m) Mc \u00bc Mc Pa R2= Cm v\u00f0 \u00de, critical mass parameter (kg) P \u00bc P= Pa, sub-ambient gas pressure (N/m2) Pa \u00bc ambient pressure (N/m2) PL \u00bc PL m R4 v2= Cm, frictional power loss (W ) R\u0304 \u00bc journal radius (m) Smn \u00bc Smn Pa R2= Cm, gas film stiffness coefficients (N/m); m, n \u00bc x, y t\u0304 \u00bc time (s) U\u0304 \u00bc R vU , instantaneous peripheral speed of the journal (m/s) W \u00bc bearing load capacity (N) X, Y \u00bc cartesian axes with origin at bearing geometric center, subscript for components X0,Y0 \u00bc coordinates of journal center in dynamical state 1 \u00bc e= Cm, bearing eccentricity ratio d \u00bc Cm= c, preload in the bearing C \u00bc attitude angle g \u00bc whirl frequency ratio at the threshold of instability m\u0304 \u00bc ambient dynamic viscosity of the lubricant (Ns/m2) u \u00bc angular coordinate measured from x-axis uko \u00bc angle of lobe line of centers u k 1 , u k 2 \u00bc angles at the leading and trailing edge of the lobe uM \u00bc mount angle uT \u00bc tilt angle v\u0304 \u00bc rotational speed of the journal (rad/s) l \u00bc bearing length to diameter ratio, aspect ratio L \u00bc compressibility number k \u00bc subscript and superscript for lobe designation O \u00bc subscript for steady state For many years, developing an intellectual technique that imitate the powerful functioning of the human brain, has been the focal point of interest for many researchers", + " In a symmetric configuration, the line joining the bearing geometric center and the center of each lobe passes centrally through the lobe, while in a tilted configuration, the lines of centers pass eccentrically through the lobes. This change on the location of lobe line centers, is called tilting and the corresponding angle in referred to as tilt angle (uT). The bearing mounting which is referred to as mount angle (uM), is the orientation of a bearing with respect to a fixed load direction. For usual upright configuration, mount angle is zero. The details of ANNs structures as well as the results obtained are presented in the following sections. Figure 3 shows the geometric details of a noncircular bearing configuration. Analysis of gas lubricated noncircular bearings involves solution of the governing equations separately for each individual lobe of bearing, treating each lobe as an independent partial bearing. To generalized the analyses for all noncircular geometries, the film geometry of each lobe is described with reference to bearing fixed Cartesian axes (Figure 3). Thus, the film thickness in the clearance space of the Kth lobe and the pressure governing equation of isothermal flow field in a lobe of bearing, with the journal in a steady state, are expressed, respectively, as (Chandra et al., 1983b): hok \u00bc 1 d 2 \u00f0Xjo\u00decos u2 \u00f0Yjo\u00desin u\u00fe 1 d 2 1 cos u2 u k o \u00f01\u00de \u203a \u203au h3o \u00f0Po \u00fe 1\u00de \u203aPo \u203au \u00fe \u203a \u203az h3o \u00f0Po \u00fe 1\u00de \u203aPo \u203az \u00bc L \u203a \u203au \u00bd\u00f0Po \u00fe 1\u00deho \u00f02\u00de where (Xjo, Yjo) are the steady state journal center coordinates, d is the preload for noncircular bearings and uko is angle of lobe line of centers" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003042_amm.740.69-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003042_amm.740.69-Figure7-1.png", + "caption": "Fig. 7. The contact stress distribution of logarithmic crowned spur gear teeth at (a) single meshing position P, (b) double meshing position B1 and (c) double meshing position B2", + "texts": [ + " Where, z coordinate coincides with the depth direction, c is the depth value for calculation, Ma is the peak value of the von Mises stress , De is the depth of its location from the surface. As depicted in Fig. 5(a), the von Mises stress along the rolling direction reaches its maximum at the depth of 0.1541mm from the contact surface, and that is the origin point of the fatigue cracks. Additionally, along the direction of tooth width (Fig. 5(b)), the origin of fatigue locates at the subsurface of the edges of the contact line, which indicates a tendency of pitting at contact edges, and thereby reduces the contact strength of the gear tooth. In Fig. 7, the stress distributions of logarithmic crowned tooth at meshing position P, 1B and 2B are depicted as exemplary cases. It shows that the contact stress in every Fig. is distributed uniformly along gear width direction, which demonstrates an excellent ability the logarithmic modification has in eliminating the edge effect of tooth surface at different meshing position. Additionally, the contact stress reaches its maximum value at the center of the contact zone and goes down slightly towards the contact edges" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003369_detc2011-48166-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003369_detc2011-48166-Figure4-1.png", + "caption": "Figure 4. TWO HYDRAULIC ACTUATORS (C), PLACED IN AN ORTHOGONAL CONFIGURATION AT 45 DEGREES WITH RESPECT TO THE LOAD CELLS, MOVE THE SHAFT. BOTH ACTUATORS ARE PROVIDED BY HIGH RESOLUTION POSITION AND FORCE TRANSDUCERS.", + "texts": [ + " A in Figure 2) is placed at the non-driven end of the shaft whereas a brand-new standard and equal bearing is placed at the driven end side (pos. B in Figure 2). The bearing housings are designed to hold bearings in the load-on-pad and in the load-between-pad configurations. Each bearing case support is connected to the machine base by two orthogonal 20 kN load cells (pos. H in Figure 3). The two directions are decoupled by means of leaf springs (pos. K in Figure 3). The load is applied in the middle of the shaft by means of two hydraulic actuators (pos. C in Figure 4) placed in an orthogonal configuration at 45 degrees with respect to the load Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/idetc/cie2011/70555/ on 05/23/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use 3 Copyright \u00a9 2011 by ASME cells as shown in Figure 4. The actuators are connected to the shaft by means of two deep groove precision ball bearings (pos. D in Figure 2). The actuators have a nominal force of 25 kN and are able to displace the shaft with amplitude of 0.1 mm with a band of 0-50 Hz and are provided by high resolution position and force transducers. Actuators can be position or force controlled. In this way, it is possible to control the position of the shaft by applying a defined orbit, reproducing what happens in a real application, and to measure both the applied forces by the actuators and the corresponding forces on the supports of the bearings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure24.5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure24.5-1.png", + "caption": "Fig. 24.5. ASIBOT robot design", + "texts": [ + " The body has two links that contain the electronic equipment and the control unit of the arm. In this manner, the robot is self-constrained, being portable with overall weight of 11 Kg. It is important to note that the robot is symmetric, and due to this, it is possible to attach the arm at any of its ends. It is made of aluminium and carbon fiber. The actuators are torque DC motors, and the gears are flat HarmonicDrive. Power supply is taken from the connector that is placed in the centre of the docking station. The range and position of the different joints can be seen in Fig. 24.5. ASIBOT is designed to be modular and capable of fitting into any environment. This means that the robot can move accurately and reliably in between rooms and up or downstairs. It can be transfered from/to a wheelchair [9]. For this purpose the environment is equipped with serial docking stations which make the transition of the robot from one to another possible. This degree of flexibility has significant implications for the care of disabled and elderly people with special needs. Modularity makes the system able to grow as the users degree of disability changes" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001384_iic.2015.7150723-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001384_iic.2015.7150723-Figure3-1.png", + "caption": "Fig. 3 Different positions of ball no. 2 relative to the inner race defect", + "texts": [ + ")]31z sinei = o (8) Equations (7) and (8) are two coupled non-linear ordinary second order differential equations. Here (j. term corresponds to additional deflection for the travel of ball in the defective region of the inner race and is calculated based on (15). The damping in the system is represented by an equivalent viscous damping C. This section presents the discussion on using spline for simulating the defect pulse generated by impact at the defect present on the inner race. Cracks, pits, spalls are included in the class of localized defects. When such a defect gets struck (Fig. 3), a pulse of short duration is produced. The expanse of the pulse is portrayal of the severity, extent and age of the damage. These factors have impact on the amplitude of response. Different pulse forms being used for this situation are rectangular, triangular, half-sine pulse etc. as shown in Fig. 4. and are the approximations for the shape of the real pulse. In real situation, the pulse generated due to impact at the defect may not be of such a regular shape. The real pulse shape lies in between the rectangular and triangular form (two non-realistic extremities)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002586_iccas.2013.6704206-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002586_iccas.2013.6704206-Figure7-1.png", + "caption": "Fig. 7 position of camera and laser pointer (left), Developed gripper (right)", + "texts": [ + " The manipulator system of MIDERS-3 is depicted in Fig. 6 and consists of three modules; a manipulator, a pan-tilt camera module, a gripper. The manipulator and pan-tilt module possess five and one degrees of freedom, respectively. The pan-tilt module attached at the mobile base. MIDERS-3 can be operated for EOD and surveillance missions by equipped a gripper at the end-effector. For the mine detection the MD and GPR sensor module is grasped by the gripper. Equipped a gripper have 2\ufffd5 km/g grasp force and is shown in Fig. 7. For accurate gripping, two laser pointers are attached. The mine detecting sensor for MIDERS-3 was developed upon the consideration regarding how to diminish or even eliminate the gap between GPR and MD, because the gap caused such a small overlapping detected area at one scan, which eventually retarded the whole process. Isung Engineering INC. cooperated in this project for developing all-in-one sensor with significantly minimized interference between GPR and MD as in Fig. 8. In keeping with the purpose of the microscopic scanning strategy, the size of all-in-one mine detecting sensor became much smaller than previous developed sensor" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002344_detc2013-12837-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002344_detc2013-12837-Figure9-1.png", + "caption": "Figure 9. DEFINITION FOR ORIENTATION AND DIMENSION OF A CONTACT ELLIPSE", + "texts": [ + " Furthermore, the orientation and dimension of the contact ellipse can be calculated based on the following information: (1) the principal directions and curvatures of the two mating surfaces at the contact point; (2) the angle formed by the first principal directions of the mating surfaces; and (3) the elastic deformation of the tooth surfaces under load. Friction force is neglected here because the gear set is assumed to be in excellent lubrication. Finally, the bearing contact is formed as a set of contact ellipses on the tooth surface during the meshing process. As Fig.9 depicts, angle is measured counterclockwise from )2( Ie to )1( Ie and can be evaluated by [2, 3]: )(tan )2()1( )2()1( 1 II III ee ee . (57) In addition, the orientation of the contact ellipse is determined by 2cos)]-()-[( )sin2-( 2tan (2)(2)(1)(1) (2)(2) IIIIII III . (58) Angle is measured counterclockwise from )1( Ie to the - axis of the contact ellipse. 8 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/asmep/77583/ on 03/02/2017 Terms of Use: http://www" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002259_amm.461.506-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002259_amm.461.506-Figure1-1.png", + "caption": "Figure 1. Grasping mechanism for perching", + "texts": [ + " No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 149.171.67.164, University of New South Wales, Sydney, Australia-14/07/15,21:30:19) This paper is focused on the control strategy design, but its fundamentals, such as the perching mechanism platform and the perching procedure, need to be introduced briefly. Perching Mechanism. Our newly designed perching mechanism is also based on the grasping concept, exactly as that from birds [17]. Its specifications are illustrated in Figure 1. The design adopts a force amplifier concept to magnify the grasping force generated by the actuation force. The digits and the frame are made of carbon fiber sheet to improve the mechanism strength, while other materials such as Delrin and Nylon are utilized for shafts, transmission arms, supports and spacers. It should be noted that the design takes into account the case of landing to ground in which the gripper will stay in open status instead of gripping. Similar gripping mechanism can be found in [18]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002120_j.euromechsol.2013.11.016-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002120_j.euromechsol.2013.11.016-Figure1-1.png", + "caption": "Fig. 1. A three-point rigid system representing a moving rigid body.", + "texts": [ + " The main innovation presented is that, given the particular nature of our approach, it systematically groups a number of desirable features which allow a simpler and better way of conceiving and understanding the rotational motion of a rigid body, thus resulting in several advantages: (a) it is integral, since it considers the angular velocity concept as a whole: matrix and vector representations together, (b) contrary to most available results, it provides a detailed derivation based in vector quantitiesdwithout a preconceived definition of the angular velocityd, thus leading to a clear understanding of the physical features associated with a representation of the angular velocity matrix of the formU \u00bc _RRT , (c) it leads naturally to simple and particularly useful expressions for the angular velocity vector,u, and, (d) it allows a readily extension of the results to three useful and important representations of the angular velocity vector, which involve the position and velocity of three noncollinear points pertaining to a moving rigid body. Finally, it is important to remark that, although the emphasis has been put on the rotational phenomenon, the multi-rigid-body system under analysis may undergo an arbitrary spatial motion. 2. Kinematic description of a moving rigid body A rigid body that undergoes a spatial displacement may be described by the position vectors, p1, p2 and p3, of three noncollinear points pertaining to the body, see Fig. 1. 3 Both frames, M : UVW and F : XYZ, are right-handed orthonormal coordinate systems. On one hand, since lines l12, l13 and l23 belong to a rigid body, vectors p2 p1, p3 p1 and p3 p2 must remain of the same magnitude throughout a rigid-body motion. On the other hand, angles a, b and g must be preserved throughout the motion as well. Thus, these three points must obey the following constraint equations: \u00f0p3 p1\u00de$\u00f0p3 p1\u00de \u00bc l213 (1) \u00f0p2 p1\u00de$\u00f0p2 p1\u00de \u00bc l212 (2) \u00f0p3 p2\u00de$\u00f0p3 p2\u00de \u00bc l223 (3) \u00f0p2 p1\u00de$\u00f0p3 p1\u00de \u00bc l12l13 cos g (4) \u00f0p1 p2\u00de$\u00f0p3 p2\u00de \u00bc l12l23 cos a (5) \u00f0p1 p3\u00de$\u00f0p2 p3\u00de \u00bc l13l23 cos b (6) which are usually known as stiffness conditions of a rigid body. 2.1. Pose of a moving rigid body The pose of a rigid body requires the knowledge of the location of one point of the body and also, the attitude of the body. The location of a point is usually representedbyapositionvector and, the attitude of the body may be described in terms of the orientation of a frame attached to the body. Such requirementsmaybe satisfied in terms of the position vectors of the three noncollinear points shown in Fig. 1. To this end, we define the following unit vectors: uh p2 p1ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi\u00f0p2 p1\u00de$\u00f0p2 p1\u00de p \u00bc p2 p1 l21 (7) mh p3 p1ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi\u00f0p3 p1\u00de$\u00f0p3 p1\u00de p \u00bc p3 p1 l31 (8) vh m \u00f0m$u\u00deu l ; lh ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi fm \u00f0m$u\u00deug$fm \u00f0m$u\u00deug p (9) whu v (10) In this way, the pose of the bodymay be represented by position vector p1 and the unit vectors u, v andw, which constitute amoving frame UVW (a right-handed orthonormal coordinate system), as it is shown in Fig", + " Representations of the angular velocity matrix The angular velocity matrix U is just a (3 3) matrix that is closely related with the angular velocity vector u of a moving rigid body in such a way that: u p \u00bc Up (20) being p an arbitrary vector and symbol \u201c \u201d stands for the usual cross product between vectors. 3.1. Centroidal representation This representation of the angular velocity matrix was derived by resorting to the position vector c and the velocity vector _c4 of the centroid associated with the triangle 123 shown in Fig. 1. The mathematical formulation is given by Angeles (2007): _P \u00bc UCP (21) where: _P h \u00bd _p1 _c _p2 _c _p3 _c (22) P h \u00bdp1 c p2 c p3 c (23) c h p1 \u00fe p2 \u00fe p3 3 (24) _c h _p1 \u00fe _p2 \u00fe _p3 3 (25) However, the centroidal representation of the angular velocity matrix,UC, can not be computed from equation (21) becausematrix P is always singular, no matter what the relative location between points is. 3.2. Proposed representation This section introduces a derivation of the angular velocity matrix with a lower level of abstraction, thus privileging the physical insight and the geometric interpretation of the results" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003403_amm.419.438-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003403_amm.419.438-Figure3-1.png", + "caption": "Fig. 3 The 3D models of the golf wood head.", + "texts": [ + " Grip (5) has a big end (7) and a small end (8) of the truncated conical shape. The grip (5) includes theouter grip (9) and the inner grip (10). The outer grip is made of thermoplastic elastomer (TPE), the inner grip is made of natural rubber. Compared with existing technology, the advantages are that the assembly process is simple. The physical parameters were shown in Table 1. This study bases on the design rules of golf wood club for four materials. This paper used SOLIDWORK software to draw the 3D models of the golf head. Fig. 3 shows the 3D models of No.1, No.3 and No.5 golf wood head. Fig.3 (a) shows the No.1 golf wood head. The loft angle is 10 o . The volume is 395cc; Fig.3 (b) shows the No.3 golf wood head. The loft angle is 15 o . The volume is 180cc; Fig.3 (c) shows the No.5 golf wood head. The loft angle is 21 o . The volume is 160cc. This paper used ANSYS to analyze different materials of golf head and shaft. The element used SOILD45. The reason is that this element is suit for 3D structure. The advantages are that the element has eight nodes and every nodes has x, y and z DOF. Fig.4 shows the maximum stress of clubface. Fig. 5 draws the variation of time and velocity for club head. This paper adopted the initial velocity of 35m/s for golf head, making transient analysis for Lagopus, Fagus and Pinaceae" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003428_amr.338.94-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003428_amr.338.94-Figure4-1.png", + "caption": "Fig. 4 Fusion of two particles in SLM", + "texts": [ + " Owing that the melting thickness is much small, the variation of total volume can be omitted. Therefore the radius of spheres after fusing can be determined according to conservation of volume. i.e.: \u2212\u2032 \u2212\u2032\u2022=\u2022 2 3 2 2 3 4 2 3 4 2 3 33 x r x rr \u03c0\u03c0\u03c0 (27) Where r\u2032 is radius after fusing, the term on the left side of equation (27) is the volume of two spheres, and the right side represents the volume after their fusion. Equation (27) can be indicated as follows: ( )xrFr ,=\u2032 (28) I.e. r\u2032 can be obtained if x and r are given. Seen from fig.4, two adjacent spheres are put in a cubic solid. The spheres contact the four planes respectively and the other two planes are through the center of two spheres individually. This cubic is considered as a volume unit. Then the space between the surface of two spheres and the cubic planes will shrink after fusing process. The space 1pV before fusing is: ( ) 333 1 81.3 3 4 2 rrrVP =\u2212= \u03c0 (29) The space 2pV after fusing is: ( ) ( ) \u2212\u2032 \u2212\u2032\u2212\u2212\u2032\u2022\u2032= 2 3 2 2 3 4 2 2 22 2 x r x rxrrVP \u03c0\u03c0 (30) Then the final porosity f\u03b5 of SLM parts can be: i p p f V V \u03b5\u03b5 1 2= (31) The equation indicates that f\u03b5 is the function of particle radius r , laser power P and scan speed V , and can be calculated through these given parameters" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure5-1.png", + "caption": "Figure 5 CUTTER AND GEAR ELEMENTS", + "texts": [ + " (2)C C N 1The expression \u201ccrossed hyperboloidal gears\u201d is an extension of \u201ccrossed cylindrical gears\u201d where \u201ccrossed\u201d is used to indicate that the cylindroidal coordinates used to parameterize the toothed bodies in mesh are determined using the two distinct cylindroids ( ; ) and$ $i o ( ; ). For generalized gear pairs, the sum 0.$ $ci co pi po< < \u0153 The moving trihedron of reference R for the input geari element becomes R (3) i pi i pi i\u0153 s s\u201a \u00d4 \u00d7 \u00d5 \u00d8N N\u2019 \u2019< < and the moving reference trihedron R for the cutter element isc R . (4) c pc c pc c\u0153 s s\u201a \u00d4 \u00d7 \u00d5 \u00d8N N\u2019 \u2019< < These two trihedrons are depicted in Figure 5. The two coordinate systems (x , y , z ) and (x , y , z ) are aligned where xi i i c c c i and x , y and y , as well as z and z are coincident. A passivec i c i c transformation is introduced such that the direction of theCc/c cutter's axis of rotation relative to the moving trihedron R isc C Cc/c cc T R . (5)\u0153 The two moving trihedrons R and R are constructed such thati c the spiral tangencies and as well as the pitch surface\u2019 \u2019< \u03c32 > \u03c33 sets in under the condition (15) where \u03d5 and c are the angle of internal friction and cohesion of the soil, respectively. This condition is mandatory for the three-dimensional problem in any coordinate system. Incomplete limiting equilibrium is theoretically defined by (15), since in addition to this condition, it is possible to propose yet another two relationships between principal stresses; this is uncharacteristic when only a horizontal force acts on the pile. The horizontal load on a circular pile can be calculated in a cylindrical coordinate system z, r, u. Figure 5 shows the computational diagram of the forces acting on the pile, and the stress state of the soil medium in the near-pile space. According to this diagram, two slip surfaces, which run at angles +\u03bc = +(\u03c0/4 \u2212 \u03c0/2) in the direction of the larger principal stress \u03c31 exist at each point of the soil medium. Consequently, two families 1 3 1 21 tan , cos 2 2 c \u03c3 \u03c3 \u03c3 \u03c3\u03d5 \u03d5 \u2212 + \u2212 = 3 2 2 1 1 sin ; 1 ; 2 4 1 sin 2 2 0 h H F Q H h h Hh Qh QH \u03d5\u03be \u03d5 \u03b3 \u03be \u03b3 \u03be \u239b \u239e+ \u2212= = +\u239c \u239f\u2212 +\u239d \u23a0 \u2212 + + = 2 (1 ) ( ) 0; 3 KQh Fh a F H h\u2212 \u2212 \u2212 \u2212 = of slip lines exist in the near-pile soil medium" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001126_2015-01-2515-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001126_2015-01-2515-Figure3-1.png", + "caption": "Figure 3. A guide block with fastener.", + "texts": [ + " They attach with one socket head cap screw each to the guide block holders (4), which are in turn attached to the outputs of the parallel gripper (3). When the pusher (2) is extended, the stopper (1) is lined up at the termination of the feed tube. After the fastener arrives and the pusher retracts, the guide blocks form a receiving hole for the incoming fastener. Then, as the pusher comes forward, the fastener is forced down the guide chute into the finger assembly (6). The guide chute has a small lead-in to ensure that fasteners travel down it correctly even if they seat slightly high vertically. This feature is easy to observe in Figure 3, as is the stepped geometry that indexes the head feature and sets the seat position. As discussed in the previous section, when the parallel gripper is actuated closed, the guide blocks form a space for the fastener to seat into, shown clearly in Figure 3. If there is some problem with the injection, such as the wrong fastener being called, a fastener being fed backwards, or some other malfunction, the injector can use its purge function to clear the feed tube and guide chute of jams. Figure 6 shows an injector tool in purge position. The parallel gripper has been opened, thereby opening the guide blocks and creating a clear path for purge. Some consideration must be made to prevent the purged fasteners from rocketing into the aircraft panel or otherwise finding their way into places they should not be" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003298_s1052618813040158-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003298_s1052618813040158-Figure4-1.png", + "caption": "Fig. 4. Determination of the vector e' = e + \u03b4: (a) e' = |e + \u03b4| > e; (b) e' = |e + \u03b4| < e.", + "texts": [ + ", \u03bc = \u03b4tors/(2\u03b4bend) = 0, which is available since \u03bc 1, then, the deflected rotor will behave as the unbalanced one, which eccentricity e' is made of the vector of proper imbalance e with respect to the point O2 where the disk is fixed on the shaft and the vector of initial deflection of the shaft \u03b4 in this point e' = |e + \u03b4| (Fig. 2). Thus, the preliminary bending of the shaft depending on the modulus values and the mutual location of e and \u03b4 can lead to an increase and decrease in the eccentricity e' (Fig. 4), as well as in the acting centrifugal inertial forces and levels of rotor vibrations caused by these forces. Upon taking the torsion deformation at the deflected rotor oscillations \u03bc = \u03b4tors/(2\u03b4bend) \u2260 0 into account, the system of the equations of motion becomes parametric, and the parameter is the rotation angle of the rotor \u03c8 (Fig. 2). The given parametric form causes oscillations of twice the frequency of the working frequency. It was found that the system of equations of motion turns out to be similar to that of the rotor on the shaft having different principal moments of inertia of cross section [1]", + ", n ~ 1), from (14) and using the above assumption of the equality of the torque and loading moments, we obtain the expression for a relative change in the rotational velocity per revolution: (15) According to data [2], the eccentricities for steam turbines do not exceed the largest value, which is sev eral thousandths of the radius of gyration of the disk r. The value of admissible reversible bending in the span center \u03b4 is 20 \u03bcm [2], i.e., the value e' = |e + \u03b4| is commensurable with the value of e (Fig. 4). Therefore, if we take e' ~ 10\u20133r, a ~ 10e' ~10\u20132r according to [2], evaluation (15) gives \u0394\u03c9/\u03c9 ~ 10\u20134, i.e., during one turn the angular velocity of the rotor has changes in value of the order of 0.01%. Consequently, we can neglect the third equation of system (10) given the above assumptions with suf ficient precision, and the rotational velocity can be taken to be constant = \u03c9 = const. Example. As noted above, the influence of the rotor shaft torsion is very small. Therefore, the main index of the rotor vibration activity is the value of its amplitude on the critical velocities at forced oscilla tions with the eccentricity increased due to the initial shaft deflection", + " The values obtained for the vibration oscillations at the mid span of the rotor at resonance are 3\u20134 times smaller than the gaps between the fixed rotor and the seal in the high pressure cylinder, being equal to \u0394 \u2245 1.2\u20131.5 mm. There fore, we can conclude that the selection of the tolerance during the heating test \u03b4tol = 20 \u03bcm is sufficiently justified. When the rotor is run out due to the angular acceleration (slowdown), the real amplitudes during pass ing the resonances will be lower. On the other hand, the values obtained for amplitudes are related to the ideal balanced rotor. In practice, they will be added to the oscillation amplitudes caused by the rotor imbalance (Fig. 4). CONCLUSIONS The system of equations of motion of the deflected rotor coincides with the corresponding system for initially rectilinear rotor, if the following replacements are implemented: The displacements u1 and u2, which are the shaft deflections at the attachment point of the geometrical disk center, are replaced by the displacements of some reduced point of the disk, which is placed on the intersection between the straight line connecting the centers of shaft end sections and the disk plane; The imbalance vector is replaced by the sum of the disk imbalance vector with the vector of disk geo metrical center displacement with respect to the reduction point" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001895_amm.288.208-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001895_amm.288.208-Figure1-1.png", + "caption": "Fig. 1 Simplified model of alternate blade cutter", + "texts": [ + " (ID: 142.103.160.110, University of British Columbia, Kelowna, Canada-10/07/15,11:57:56) Establishment of gear tooth surface equation The tooth surface of spiral bevel gear has generally two different processing methods, molding method and generating method [2]. Different processing methods have different mathematical models with different expressions. An example of the formatted wheel member of spiral bevel gear pairs is represented to analyze the establishment of gear tooth surface equation. As Fig. 1 shows, the cutting trajectory of head-cutter is a conical surface. The cutting cones of the outside blade and the inside blade are shown in Fig. 2. The head-cutter is represented by vector function rt(u, \u03b8) as \u2212 +\u00b1 +\u00b1 = = \u2192 1 sin)]tan 2 ([ cos)]tan 2 ([ 1 ),( ),( ),( ),( 0 0 u u w r u w r ur ur ur ur tz ty tx t \u03b8\u03b1 \u03b8\u03b1 \u03b8 \u03b8 \u03b8 \u03b8 . (1) Here, u and \u03b8 are the surface coordinates; u is the point along the axis of the cutter head on the tool side edge to the nose plane; \u03b8 is the angle turns from the x-axis positive when the head-cutter is cutting; r0 is the cutter nominal radius; w is the cutter point width; \u03b1 is the blade angle" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001126_2015-01-2515-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001126_2015-01-2515-Figure5-1.png", + "caption": "Figure 5. Full vertical inject injector assembly cutaway view.", + "texts": [], + "surrounding_texts": [ + "As discussed in the previous section, when the parallel gripper is actuated closed, the guide blocks form a space for the fastener to seat into, shown clearly in Figure 3. If there is some problem with the injection, such as the wrong fastener being called, a fastener being fed backwards, or some other malfunction, the injector can use its purge function to clear the feed tube and guide chute of jams. Figure 6 shows an injector tool in purge position. The parallel gripper has been opened, thereby opening the guide blocks and creating a clear path for purge. Some consideration must be made to prevent the purged fasteners from rocketing into the aircraft panel or otherwise finding their way into places they should not be. The injector purges with the injector positioned in front of the headstone, so purged fasteners contact the headstone before falling harmlessly to the floor. With good injector reliability, purge functions will only be used in extremely rare cases." + ] + }, + { + "image_filename": "designv11_84_0003149_icpe.2011.5944638-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003149_icpe.2011.5944638-Figure9-1.png", + "caption": "Fig. 9. Hardware of the experimental setup.", + "texts": [ + " The main reason is that under square wave computation, the high-order harmonics of the square- wave current will cause additional torque ripple and power loss. Output efficiency comparisons among different commutation schemes are summarized in Table I. Several experiments are performed to verify the mathematical analysis conducted in previous sections. The specifications of the BLDCM and the PMSM used in the experiment are listed in Table II and Table III, respectively. In addition, the hardware of the experimental setup is illustrated in Fig. 9. Our theoretical analysis indicates that the efficiency of the sinusoidal commutation is better than that of the square wave commutation. In order to justify the theoretical analysis, we have to fixate the root-meansquare value of the current at certain desired value, which is often tedious and complicated in practice. Hence, in the experiment, in order to solve this tricky problem, we first fixed the load torque at 0.588 N-m. Subsequently, we recorded the current dada at the DC source end under a fixed speed for two different commutation schemes" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003948_amm.423-426.1936-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003948_amm.423-426.1936-Figure1-1.png", + "caption": "Fig. 1 Assembly relationship of hydraulic press parts", + "texts": [ + " On the other hand, the weight of upper beam accounts for a relatively large proportion in total weight of hydraulic press, the heavier upper beam is, the more production cost will be consumed. In order to improve structure of upper beam, a method of structural optimization is proposed. Strength and stiffness analysis of hydraulic press Hydraulic press is consisted of upper beam, sliding beam, lower beam, main piston cylinder, assistant piston cylinder, main piston, assistant piston, vertical column, upper plate, lower plate, tension rod, upper die, lower die, nut. Assembly relationship of hydraulic press parts is shown in Fig. 1. A quarter of hydraulic press is used to establish finite element analysis mold for strength and stiffness analysis in view of symmetrical structure[1-2]. Friction coefficient between piston and piston cylinder is set to 0.001 while friction coefficient used for general condition is set to 0.15[3]. All DOFS of lower beam bottom are constrained in order to avoid rigid body motion[4-5]. Two All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002844_iccas.2013.6703865-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002844_iccas.2013.6703865-Figure1-1.png", + "caption": "Fig. 1 Definition of the body-fixed coordinates (x, y, z), angle-of-attack a, side-slip angle (3, total angle-of attack at, and aerodynamic roll (or bank) angle CPa , where Vm is the total velocity, and p, q, T are angular velocities about each axis.", + "texts": [ + " We proposes a simple nonlinear operator to integrate sinusoidal errors effectively in the sense of averaging. By using the operator we introduce a new control law, which is equivalent to conventional proportional and integral control, for the rolling missiles. Let us consider coordinate systems, including body axes, aeroballistic axes, instantaneous manoeuvre axes and manoeuvre axes, which are needed to describe rolling airframe control systems. Body coordinates (Ocg : x, y, z) are fixed on airframe and its origin lies at the cen ter of gravity (c.g.) of the airframe. Figure 1 shows the definition of body axes and relevant angles. As an exam ple we write the angles of attack, lateral accelerations and control variables of rolling airframes with a single plane of control surfaces on the body axes as follows. [ \ufffd ] [ TJy ] T/z [ \ufffd: ] [ (30 sin (Pot + \u00a2o,) ] ao cos (Pot + \u00a2a) , [ TJyo sin (Pot + \u00a2ry) ] rlzo cos (Pot + \u00a2ry) , [ 50 cos (:ot + \u00a28) ] , (I) (2) (3) where (3, a are the side-slip angle and angle-of attack, 'T/y, 'T/z denote the lateral accelerations along each axis, 5y, 5z represent the fin deflections for yaw and pitch con trol, respectively, Po is the roll rate, and \u00a2c\" \u00a2ry, \u00a28 are roll positions where the total angle-of-attack, the maxi mum acceleration, and the maximum fin deflection are achieved, respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003976_2014-01-1670-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003976_2014-01-1670-Figure8-1.png", + "caption": "Figure 8. Analysis area", + "texts": [ + " This transparent oil ring was the same geometry as the ordinary oil ring and the coil expander was assembled to provide almost the same tensional force as such ordinary oil ring because the tensional force of the intended 2 piece oil ring is provided mainly by the coil expander inside of it (Figure 7). The distribution of fluorescing particles was equalized in the ondina oil, the particles with oil was injected into the back clearance of the oil ring groove under motoring condition, and the photo of the particles flow was taken. The direction and velocity of oil were measured by obtaining the moving distance from the particle position of the photograph. The experimental conditions are shown in Table 4 and the measurement area is shown in Figure 8. Measurement results are shown in Figure 9. The oil dilution rate of oil ring without oil vent hole worsened and it fell to a maximum of 0.8 time by the thrust side, it fell to a maximum 0.6 time by the front side, and it had the especially large falling on the front side in comparison with the ring with oil vent holes. This phenomenon was assumed to be caused by the reduction of oil flowing into the oil ring groove, but the oil behavior around the oil ring such as back clearance, oil vent hole and rail were still left unexplained" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003466_amr.724-725.1398-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003466_amr.724-725.1398-Figure2-1.png", + "caption": "Fig. 2 Equivalent circuit of BLDCM", + "texts": [ + " As a compound element, IGBT is simple to be controlled, has great out capability. But it is very expensive and mainly applied in high power electric vehicle. The position sensor of BLDCM usually adopts electromagnetic, photoelectric and magneto-sensitive type. For electric vehicle, Hall sensor is widely used because it is simple, small and cheap. In the control system of BLDCM, the conversional scheme adopts the BLDCM with position sensor, detects the location of the rotor by the sensor to realize commutation, and uses the mode of 3-phase 6-state PWM. Fig. 2 shows the equivalent circuit of BLDCM. BLDCM is usually used with the mode of 3-phase 6-state PWM. At any moment, only two of the phases work and the other phase shutdowns. And the conduction sequence is decided by the position sensor. The conduction time of every phase is 120 degrees of electrical angle, realize commutation every 60 degrees. It must forbid the up and down bridge of the same phase to be conduction at the same time in case the power devices would be shorted. The torque of BLDCM is proportional to the phase current, and speed regulation can be achieved by changing the PWM duty cycle" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001963_ijcnn.2014.6889627-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001963_ijcnn.2014.6889627-Figure2-1.png", + "caption": "Fig. 2. The block diagram of the overall control system.", + "texts": [ + " Let the controller be given by u \ud835\udc3b\ud835\udc34\ud835\udc3c\ud835\udc46\ud835\udc45\ud835\udc41\ud835\udc36\ud835\udc56 = u\ud835\udc52\ud835\udc5e\ud835\udc56 +H \u22121 \ud835\udc56 u \ud835\udc34\ud835\udc3c\ud835\udc46\ud835\udc45\ud835\udc41\ud835\udc36\ud835\udc56 \u2212H \u22121 \ud835\udc56 u\u210e\ud835\udc56 (47) where u\ud835\udc52\ud835\udc5e\ud835\udc56 is defined in (16), u \ud835\udc34\ud835\udc3c\ud835\udc46\ud835\udc45\ud835\udc41\ud835\udc36\ud835\udc56 is in (17), and u\u210e\ud835\udc56 is given by u\u210e\ud835\udc56 = 1 \ud835\udf0c2\ud835\udc56 s\ud835\udc56 + 1 \ud835\udf0c2\ud835\udc56 B\u0304\u22a4P\ud835\udc56e\u0304\ud835\udc56 (48) for which \ud835\udf0c\ud835\udc56 is a positive constant, and the matrix P\ud835\udc56 = P\u22a4 \ud835\udc56 > 0 is the solution of the following Riccati-like equation P\ud835\udc56\u03a6\u0304\ud835\udc56 + \u03a6\u0304 \u22a4 \ud835\udc56 P\ud835\udc56 +Q\ud835\udc56 \u2264 0 (49) Then, system states of the closed-loop system and parameter estimates are bounded and the \ud835\udc3b\u221e tracking performance is achieved within the prescribed attenuation level \ud835\udf0c2\ud835\udc56 . Proof: Please see Appendix II. \u25a0 From the above analysis, the block diagram of the overall control system is shown in Fig. 2. Hence, the design procedures of the adaptive integral sliding recurrent neural control design for nonlinear interconnected systems with \ud835\udc3b\u221e performance see Fig. 3 can be summarized as follows (\ud835\udc56 = 1, 2, . . . , \u2113): Step 1: Select B\u2020, \ud835\udc66\ud835\udc58\ud835\udc56 , \ud835\udf53 \ud835\udc56 (h\u0303\ud835\udc56, ?\u0303?\ud835\udc56, z\u0303\ud835\udc56), \ud835\udf14\ud835\udc56, \ud835\udf011\ud835\udc56 , \ud835\udf01 2\ud835\udc56 , \ud835\udf01 3\ud835\udc56 , \ud835\udf01 4\ud835\udc56 , \ud835\udc59 0\ud835\udc56 , \ud835\udc59 1\ud835\udc56 , \ud835\udc59 2\ud835\udc56 , \ud835\udc59 3\ud835\udc56 , \ud835\udc59 4\ud835\udc56 , \ud835\udc59 5\ud835\udc56 , \ud835\udc59 6\ud835\udc56 , \ud835\udc59 7\ud835\udc56 , \ud835\udc59 8\ud835\udc56 , \ud835\udf10 1\ud835\udc56 , \ud835\udf10 2\ud835\udc56 , \ud835\udf1b 1 , \ud835\udf1b 2 , \ud835\udf10 3\u0304\ud835\udc56 , \ud835\udf10 4\u0304\ud835\udc56 , \ud835\udf1b 3\u0304 , \ud835\udf1b 4\u0304 , \ud835\udc4e\u22121 \ud835\udc561 , \ud835\udc4e\u22121 \ud835\udc562 , \u039b\ud835\udc56, \ud835\udf001\ud835\udc56 , \ud835\udf00 2\ud835\udc56 , \ud835\udf003\ud835\udc56 , \ud835\udf004\ud835\udc56 , \ud835\udf16, \ud835\udc5d(\ud835\udc61), and the initial conditions for parameter vectors w\u0302\ud835\udc50\ud835\udc56 , h\u0302\ud835\udc56, " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002780_amm.121-126.3087-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002780_amm.121-126.3087-Figure7-1.png", + "caption": "Fig. 7 Pressure distribution Fig. 8 Velocity distribution", + "texts": [ + "8mN at the rotational speed of 1.5\u00d710 5 rpm and gear height of 20\u00b5m. As illustrated above, the bearing force of the air journal bearing is 0.72mN at the gear number of 18 and the gear height of 30\u00b5m, which is the closest to the weight of the rotor (G=0.6mN). And the rotational speed of the shaft is 1.18\u00d710 5 rpm, which is relatively low and can be achieved easily. The pressure and velocity distribution of the air journal bearing at the gear number of 18, gear height of 30\u00b5m and rotational speed of 1.18\u00d710 5 rpm are shown in Fig. 7 and 8. It shows that the pressure distribution in air bearing zone on the right side is bigger than that on the left side, the pressure on bottom is bigger than that on top. It means that the center of rotor is located at the lower right corner. This is due to the eccentricity of rotor. The load capacity is equilibrium to the mass of the rotor. The rotor can be lifted by the self-active load force that generated by the compressed air. This can be verified by the simulation data. So, the rotor system is in a stable state" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001946_amm.698.552-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001946_amm.698.552-Figure2-1.png", + "caption": "Fig. 2. Direction vectors", + "texts": [ + " The rotation angle of link FA relative to the fixed link AB (hinge A angle) is the driven link rotation angle. The kinematic analysis of the mechanism has allowed defining the relation between the driving link rotation angle and the driven link rotation angle. A complete kinematic analysis of Brikard\u2019s linkage is given in [4]. Brikard\u2019s linkage consists of six identical links AB, BC, CD, DE, EF, and FA. Despite the identical parameters of links, it is possible to assemble a mobile mechanism only by correct orienting of links during the assembly process. Let us consider link BC (Fig. 2). All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 128.210.126.199, Purdue University Libraries, West Lafayette, USA-07/07/15,12:22:43) This link contains two hinges B and C whose axes interbreed at 90\u00b0. Randomly choose a positive direction of the hinge B axis, direction vector B. The hinge surface corresponding to the direction vector is a working face" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001567_icma.2015.7237783-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001567_icma.2015.7237783-Figure9-1.png", + "caption": "Fig. 9 Coordinates of eyeball mechanism", + "texts": [ + " The two eyeballs of robot\u2019s head have the same size and are placed symmetrically. So only one eyeballs\u2019 kinematics analysis is necessary. Eyeball has two DOFs: pitch and yaw. Its revolving angle are 2, 3 respectively. The eyeball mechanism is driven by two motors. Its revolve angle are 1 and 4 respectively. It is an inverse kinematics problems to deduce 1 and 4 with the expression of 2 and 3. Following are kinematic analysis of left eye mechanism. First, establish coordinates {1}, {2}, {3}, {3\u2019}, {4}. As it is showed in Fig. 9, coordinates {1}, {2}, {4} are base coordinates which are fixed on head skeleton. The origin of coordinate {1}, {2} are the center of the motor rockers. Its revolve angle are 1 and 4 respectively. The origin of coordinate {4} are the center of eyeball. Coordinate {3}, {3\u2019} are fixed on the eyeball and revolve with it. Coordinates {3} revolves around z axis in reference to coordinate {2}. Its revolve angle is 2. Coordinate {3\u2019} revolves around x axis in reference to coordinate {3}. Its revolve angle is 3" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002979_s11249-015-0466-9-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002979_s11249-015-0466-9-Figure2-1.png", + "caption": "Fig. 2 Heat pipe disk: a heat pipe disk after heat treatment. b Rear end of the heat pipe disk. c Front end of heat pipe disk after lapping. d Front end of rotating disk after lapping", + "texts": [ + " Figure 1 shows the design of the heat pipe stationary disk and the conventional rotating disk. Two channels are cut through from rear end of the heat pipe disk to 1 mm away from its front end. The two blocks between channels are for thermocouples. The width of each channel is almost the same size as the rubbing face to ensure that most of the heat generation is absorbed by the working liquid (water is used here). The disk is made of stainless steel (17-4 PH) and is heat-treated to a hardness of 45 Rockwell C. Figure 2a shows the heat pipe disk after heat treatment. Thin layers of wick, made of nylon fiber, are inserted into the channels after polishing the heat-treated disk outer surface. Then, a thin copper cap is soldered to cover the rear end of the channels and sealed by epoxy to prevent water or vapor from leaking, Fig. 2b. After injecting 0.5 ml water into each channel through filling holes, the holes were tightly sealed by screws. The heat pipe disk is then put into a vacuum oven and let the pressure approach absolute zero. During the vacuum process, a small amount of water is pushed out from the sealed hole with air, but water can also seal the tiny gaps between the filling holes and screws to block the air. The filling holes are then sealed with waterproof glue. Figure 2c shows the polished front end (rubbing face), which is lapped to 1\u20132 helium light band. Three thermocouple holes are drilled through the sidewall blocks; two in the front end for measuring the face temperature (one near the inner diameter and another close to the outer diameter) and the third hole for measuring the rear end temperature. The rotating disk is a conventional disk, without a heat pipe, also made of 17-4 PH and heattreated to the same hardness of 45 Rockwell C, see Fig. 2d. As shown in Fig. 3, during the operation, water inside the wick under the rubbing face absorbs friction heating and vaporizes. Then, vapor flows through the near-vacuum space inside the heat pipe housing and condenses at the sidewall and the bottom of the disk. Finally, water flows back to the area under the rubbing face through the wick. 3.2 Rotating Disk with Built-In Heat Pipe The making of heat pipe rotating disk follows the same procedure as that of stationary disk. Figure 4 shows the design of the heat pipe rotating disk, and Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003715_robio.2014.7090476-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003715_robio.2014.7090476-Figure4-1.png", + "caption": "Fig. 4. Distribution of calibration points (over view)", + "texts": [ + " According to above configurations, the corresponding coordinates (x, y, z) of the TCP, in the base coordinate frame of the robot, can be calculated using forward kinematics of the robot, and the mean of the coordinate z, z, can be obtained. If the errors are sufficient small and the ground is level, then the value of z will be consistent. We have performed calibration and compensation to make these errors small. In this case, those points with (z \u2212 z) \u2265 2 mm are discarded. It is well known that three non-collinear points can determine a circle. However if these three points are too closed, the error of the circle calculated by them may be large. We choose the middle point as marked in Fig.4, and the other two symmetrical points whose radii are 30 degrees far away from the middle one. The coordinates of the circle center in the robot base coordinate frame and the radius of the circle can be obtained from them. From the above 48 configurations, there will be 19 sets of radius and center coordinates calculated in the same manner. Observe these results and save those points with |Rc \u2212Rs| \u2264 1 mm, where Rc and Rs stand for the calculated radius of the circle and the radius of the sphere, respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002117_2013-01-2011-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002117_2013-01-2011-Figure10-1.png", + "caption": "Figure 10. Finite Element Cab model used for Inertia Relief Analysis", + "texts": [ + " Figure 5 shows the displacement of the cab rear mount in the vertical direction for the 5 configurations. In addition to the string pots, the cab was instrumented with accelerometers as well. Figures 6, 7, 8 show the locations of the accelerometers. CVM-D (Complete Vehicle Modeling - Diamondback), which is a Volvo in-house developed tool, was used to calculate the Forces entering the cab mounts based on the displacements measured by the string pots. CVM-D model used for the analysis is shown in Figure 9. A fully trimmed cab model as shown in Figure 10 was used for the analysis. Forces extracted from CVM-D were applied at each cab mount and the stresses were extracted while giving support cards using PARAM, INREL,-2. A single Automatic Side Loading operation takes 10 seconds from start to finish. The vertical global displacements with the 75Lb garbage can, measured at the four locations are shown in Figure 11. At 5seconds time slice, the vertical global displacements show a roll mode of the cab. The front left and rear left move up by 6mm and 7" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003298_s1052618813040158-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003298_s1052618813040158-Figure1-1.png", + "caption": "Fig. 1. The model of single disk rotor with the deflected shaft.", + "texts": [ + " The shaft is supported by elastic bearings with stiffnesses k1 and k2 in the horizontal and vertical directions. The disk on the shaft is located symmetrically with respect to the bearings, the length of the half span is l. The initial curvature of the shaft (the arrow of the static deflection) \u03b4 and the disk eccentricity e are considered to be known. The rotor rotates with some angular velocity . The calculation scheme of the construction (a view in the projection on the plane passing through the bearing axis and the disk mass center) is shown in Fig. 1: epr = ecos(\u03d5 \u2013 \u03c8\u00b7 DOI: 10.3103/S1052618813040158 282 JOURNAL OF MACHINERY MANUFACTURE AND RELIABILITY Vol. 42 No. 4 2013 VOLOKHOVSKAYA, BARMINA \u03c80), \u03b4pr = \u03b4cos\u03c80 + \u03b4st, mgpr = mgcos(\u03c8 + \u03c80) are the eccentricity, total shaft flexure, and rotor weight pro jections on the specified plane given a fixed rotor; \u03b4st is the static deflection of the shaft in this plane; the meaning of the angles \u03d5 and \u03c80 is evident from Fig. 2. Figure 2 conditionally shows the scheme of the rotating disk at some arbitrary time (a side projection). In this case, it is assumed that the bent shaft is in its initial state, i.e., not deformed by the forces P1 and P2 of the bearing reaction. With rigid bearings, the point O (Fig. 1) is the projection on the disk plane of the fixed axis of the bearings; the point O1 is the projection on the same plane connecting the ends of the non deformed bent shaft; the point O2 is the geometrical center of the disk (the attachment point of the disk on the shaft); O3 is the mass center of the disk having the vector\u2014eccentricity e; the vector \u03b4 = cor responds to the projection of the cantilever of the bent shaft on the disk plane and is equal to bending vec tor of the nondeformed shaft. The reaction forces of bearings P1 and P2 should be such that the point O1 after deformation of the shaft by these forces moves to the point O", + " Let us determine the projections of displacements u1 and u2 at axis Ox1 and Ox2: By adding the bearing compliances in the direction of axes Ox1(1/k1) and Ox2(1/k2) with use of (1), (2), and (3), we find the displacements in the directions of these axes for the rotor installed on the anisotropic compliant bearings: (5) (6) Note that the correlation \u03b4tors/\u03b4bend ~ (\u03b4/l)2 1 is of a sufficiently small value, so the values \u03b4tors/\u03b41 1, \u03b4tors/\u03b42 1 are small according (6). Taking this into account, we solve system (5) for the reactions P1 and P2 conserving only the terms of the first order with respect to \u03b4tors/\u03b4bend. Considering the directions of P1 and P2 (Fig. 1) against the displacements, we obtain (7) The system of equations of the deflected rotor consists of the equation of forward motion along Ox1 and Ox2 taking the origin at the fixed point O of the space (Fig. 1): , (8) dutors Mtorsdz GItors y\u2013 1 2 Ppa y 2 dz GItors , utors\u2013 1 2 Ppa y 2 zd GItors ; 0 l \u222b\u2013= = = utors 1 2 Ppa\u03b4tors, \u03b4tors\u2013 \u03b4 2 l 2GItors .= = u1 uax \u03c8cos upa \u03c8, u2sin\u2013 uax \u03c8sin upa \u03c8.cos+= = u1 \u2013P1 \u03b41 \u03b4tors \u03c8sin 2 +[ ] P2\u03b4tors \u03c8 \u03c8,cossin+= u2 P1\u03b4tors \u03c8 \u03c8cossin P2 \u03b42 \u03b4tors \u03c8cos 2 +[ ];\u2013= \u03b41 \u03b4bend 1/k1, \u03b42+ \u03b4bend 1/k2.+= = P1/m u1\u03a91 2 4\u03bc\u03c71 u1\u03a91 2 \u03c8sin 2 u2\u03a92 2 \u03c8 \u03c8cossin\u2013[ ],\u2013= P2/m u2\u03a92 2 4\u03bc\u03c72 u2\u03a92 2 \u03c8cos 2 u1\u03a91 2 \u03c8 \u03c8cossin\u2013[ ];\u2013= \u03a91 2 = 1/ m\u03b41( ), \u03a92 2 = 1/ m\u03b42( ), \u03c71 = \u03b4bend/ 2\u03b41( ), \u03c72 = \u03b4bend/ 2\u03b42( ), \u03bc = \u03b4tors/ 2\u03b4bend( )" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure8.28-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure8.28-1.png", + "caption": "Fig. 8.28 Three phases of the gait cycle", + "texts": [], + "surrounding_texts": [ + "individual components or elements in order to know their behavior and then rebuild the original system for its analysis. This represents a general approach method of discretization for continuous problems defined bymathematical expressions. In FEM as in our models, the term known as domain represents a \u201ccontinuous system.\u201d Also, there are the \u201cboundary conditions\u201d or \u201climits\u201d which are ranges or conditions from a function that interacts in the continuous system and will be solved in the problem. The mesh generation is called \u201cdiscretization,\u201d where we generate the \u201celements\u201d and \u201cnodes.\u201d An element is a part of the continues system that is separated from the continuous system, and its typical shapes are triangles, quadrilaterals, or tetrahedron. These elements are united in apoint that is called \u201cnode\u201d andoften represent the vertex of many elements. The numerical analysis was made in the commercial simulation software ANSYS\u00a9 with the following characteristics:\nThe ortheses were printed in Z-Ultrat, a polymer based onABS. Young\u2019sModulus was 1850MPa and Poisson\u2019s ratio of 0.35, and the model was loaded with 450 N that represents more load than the boys apply on the ortheses. A friction coefficient of 0.8 was assigned to obtain a friction force of 55.53 N. For the boundary conditions, we considered the three human gait phases as shown in Figs. 8.28 and in 8.29.\nThe meshes were generated with 287,601 elements, 422,134 nodes, and 107,604 elements, 189,641 nodes for the orthesis of the clubfoot and the orthesis of the healthy foot, respectively, and the results are shown in the next pages. Figure 8.30 shows the generated meshes.", + "Interferometry is an experimental test that uses digital correlation of images that allowed to measure the displacement in 2D and 3D images. The technique consists", + "in taking a series of pictures or video during the test from the initial to the final stage. For the preparation of the model, it is necessary to paint the surface of the model in white as shown in Fig. 8.31. Next, a mottled with black painting is made to generate a contrast that allowed the program run the simulation as shown in Fig. 8.32. If the program does not distinguish the mottled, then it is necessary to repeat the process of painting. In images like in video, the file is divided in virtual subsets called facets, and through an algorithm of correlation, a region or facet of the initial image is searched until the final image is obtained to determine the displacement vector in each set of facets. For this study, the mottled was performed in both ortheses.\nThe experimental study was made in GOM correlate software using human gait phase\u2019s videos in both ortheses, Fig. 8.33. File formats accepted by the software are MP4, MPEG, or AVI. In this study, an AVI file was used.\nThrough a surface component, conditions like quantity of points, area selection were generated. The software identifies these parameters and performs the analysis, see Figs. 8.34, 8.35, and 8.36." + ] + }, + { + "image_filename": "designv11_84_0003538_amm.325-326.870-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003538_amm.325-326.870-Figure6-1.png", + "caption": "Fig. 6. Determining a unique coordinate system", + "texts": [ + " The measurement settings are: \u2022 A = 0 o , B = 0 o (the touching probe orientation); \u2022 The scanning direction: from right to the left; \u2022 Setting the plane in which the touching probe is moving: the OLXLYL plane; \u2022 The start and stop points of the scanning ; \u2022 Determining the approaching direction by setting the angle between the touching probe and the axes of the coordinate system: XL = 180 o , YL = 90 o , ZL = 90 o ; \u2022 The step of the scan: p = 0,3 mm; \u2022 The security distance: ds = 0,5 mm. The experimental data is saved in a text file for post-processing. The processing of the two data sets can be divided in two subprocesses: transforming the data in the same coordinate system and overlapping the suitable surfaces. Determining the unique coordinate system. Figure 6 presents the two sets of data, where it is obvious that the surfaces cannot be compared only by the transformation of the coordinate system. To avoid introducing processing errors, only the system of the theoretic profile is transformed. According to figure 6, a roto-translation is necessary: a rotation around the YL axis and a translation in the direction\u201e of the ZL axis. Therefore the two sets of surfaces/points are located on the same cone. Overlapping the adequate surfaces. To overlap the suitable flanks, the angle between the theoretical and measured surfaces must be determined. These angles are rotation angles around the ZL axis determined on the pitch cone of the bevel gear. The measured surface does not exactly overlap on the theoretical surface, thus there are errors between them" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000926_gt2008-50641-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000926_gt2008-50641-Figure1-1.png", + "caption": "Figure 1 \u2013 Coordinate system of the foil bearing", + "texts": [ + "org/ on 12/20/2017 T limitations, the concept of gas foil bearing appeared in the mid sixties. For such bearings, an elastic and dissipative structure is added between the housing and the rotating shaft. During the last decades, several types of foil bearings (tape type, leaf type, bump type\u2026) have been designed. Among all these designs, the bump-type foil bearing is now considered to be the most efficient [1], [2]. This type of bearing is made of one or several corrugated sheets covered by a smooth top foil (Fig. 1). The corrugated and flat top sheets are welded at one end and free at the other end. Due to rotor speed, the gas pressure is applied on the top foil which pushes radially the corrugated foil. As this foil is welded at a single extremity, the bumps slide along the circumferential direction and dry friction occurs with both the housing and the top foil. When the bearing is submitted to vibrations (unbalance mass, shocks, self excited vibrations,\u2026), the relative displacements within the foil structure contribute to the energy dissipation and to the overall stability of the system", + " presented the experimental minimum film thickness of a first generation foil bearing in function of the applied static load [21]. The dimensions, the geometry, and the working conditions of the foil bearing are given in Table 1. The results obtained with the NDOF model are compared in Figs. 6 and 7 to the experimental data from [21] as well as with the theoretical results from [10] obtained by using Iordanoff\u2019s formula. The coordinate system of the bearing and the angular location of the top foil\u2019s spot weld are shown in Fig. 1. The welding is thus located at 180\u00b0 (upwards), i.e. at the opposite of the static load which is oriented downwards. Two rotating speeds of the shaft are considered: 30 000 rpm (Fig. 6) and 45 000 rpm (Fig. 7). Copyright \u00a9 2008 by ASME rms of Use: http://www.asme.org/about-asme/terms-of-use Dow Figures 6 and 7 depict the minimum film thickness versus the load. The correlation is satisfactory for moderate to heavy loads, but discrepancies are obtained for light loads where the structure plays a less significant role", + " A similar behavior was obtained for a foil bearing constituted of five strips along the circumference, containing five bumps each (Table 1). The results for both foil bearings are summarized in Fig. 11: In this example, few differences are observed between the single strip case and the multiple strips case. Both bearings exhibit an optimal friction coefficient around 0.2 \u2013 0.3. The influence of the top foil\u2019s spot weld location has also been addressed. The coordinate system used to define this location is depicted in Fig. 1. Four spot weld locations have been tested (0\u00b0, 90\u00b0, 180\u00b0, and 270\u00b0) for both bearings (respectively with one and five strips). The friction coefficient is fixed at 1.0=f\u03bc and the mass of the shaft at kg M 8.1= . The results for the bearing constituted of a single strip are represented in Fig. 12: It clearly appears that the spot weld location plays an important role for the bearing stability. The bearing is at the stability limit when the spot weld is located at 180\u00b0 (upwards). A welding point located at 90\u00b0 (on the right hand side) renders the bearing stable whereas a welding point located at 270\u00b0 (on 7 Copyright \u00a9 2008 by ASME erms of Use: http://www" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003641_peds.2013.6527115-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003641_peds.2013.6527115-Figure7-1.png", + "caption": "Figure. 7. Maximum torque characteristics.", + "texts": [ + " The thickness of the magnets is 3 mm in the P-F1 model, 1.5 mm in the P-F2 model, and 1.0 mm in the P-F3 model. In this regard, the PM volume is the same in each layer. The width of the flux path between PMs is 4.5 mm in the P-F2 model and 4.0 mm in the P-F3 model. The increase in the number of layer is expected to cause a decrease in Ld and an increase in Tr The motor parameters and the torque characteristics of the flat structures are shown in Figs. 6 and 7, respectively. The broken line in Fig. 7 represents the torque of the S-F1 model. As the number of layers in the models with flat magnets increases, Ld decreases and Tr increases. As a result, the maximum torque of the flat magnet structures increases, and the torque of the P-F3 model is the highest among these structures (85% of that of the S-F1 model), as shown in Fig. 7. B. V-shaped Magnet Structures The rotor structures with V-shaped magnets are shown in Fig. 8. The P-V1, P-V2, and P-V3 models are single-layer, two-layer, and three-layer structures, respectively, with Vshaped magnets. The thickness of the magnets is 3 mm in the P-V1 model, 1.5 mm in the P-V2 model, and 1.0 mm in the P-V3 model. In this regard, the PM volume is the same in each layer. The width of the flux path between the PMs is 4.5 mm in the P-V2 model and 4.0 mm in the P-V3 model. The increase in the number of layers is expected to cause a decrease in Ld and an increase in Tr" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001216_icit.2015.7125200-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001216_icit.2015.7125200-Figure3-1.png", + "caption": "Fig. 3. Rotor of developed motor and q-axis flux[1]", + "texts": [ + "(1) because of the drop in q-axis inductance caused by the magnetic saturation. Therefore, the locations described in [4] have been deep in the rotor core to make the q-axis magnetic path wide enough. The stator of the developed motor is the distributed winding one which has 24 slots, because the iron loss of the distributed winding stator can be less than one of the concentrated winding stator even if the stator is small and the winding factor of the distributed winding stator is more than one of the concentrated winding stator. Fig.3 shows the developed rotor. Features of the motor with the rotor of Fig.3 are shown as follows. \u2022 The permanent magnet is shallow in the rotor core, even though buried permanent magnets are in a single layer \u2022 The q-axis bridges exist between the magnets \u2022 The grooves exist between the edge of the magnet and the rim of the rotor \u2022 The number of turns of the stator winding is increases There features are explained bellow. When the permanent magnet is deep in the rotor core, the magnet torque becomes small, as the maximum flux linkage \u03c8a in Eq.(1) decreases because the leakage fluxes \u03d5as exist as shown in Fig.1. Therefore, the rotor with permanent magnets buried shallowly is used. B. Existence of the q-axis bridges Even though the permanent magnet buried shallowly, if the q-axis bridges between magnets are made as shown in Fig.3, the decrease of q-axis inductance caused by the magnetic saturation is relieved because q-axis flux flows enough as shown in Fig.3. The effect of making the q-axis bridges is shown by magnetic field analysis using FEM. TableI shows the specification of the motor model for magnetic field analysis. Fig.4 shows the motor model. When the U-phase and V-phase currents flow as shown in Fig.4, in case of the rotor position shown in Fig.4, the q-axis flux is maximum. Fig.5(a),(b),(c) show the results of the magnetic field analysis at 3A when the width of the q-axis brigdes tq is 3mm,5mm,7mm, respectively. Fig.5(a),(b),(c) show that the magnetic field saturation occur at the part [A] of Fig.4 at any tq because the permanent magnet buried shallowly. However, Fig.5(a),(b),(c) show that q-axis flux flows enough through the part [B] of Fig.4 with increasing tq . Therefore, the existence of the q-axis bridges cause that the q-axis flux flows enough as shown in Fig.3 even though the permanent magnet buried shallowly. The sound noise caused by torque ripple increases when the permanent magnet is buried shallowly because of the shortage of the flux due to permanent magnet \u03d5s which flows the teeth of the stator as shown in Fig.6 Therefore, the grooves of the rotor are made as shown in Fig.3. The depth of the grooves and the air gap are defined as L and g, respectively, as shown in Fig.7. Fig.8 shows that the sound noise decreases with increasing L/g (i.e. increasing the depth of the grooves). Therefore, the existence of the grooves cause the suppression the sound noise. Fig.9 shows a example of torque-velocity curve of IPM motors. It is possible to increase flux linkage \u03c8a in Eq.(1) by increasing the number of turns of the stator winding, but the maximum velocity \u03c9b at constant torque region shown in Fig", + " \u2022 Stator outer diameter is 90mm compared with a conventional motor\u2019s stator outer diameter of 105mm. \u2022 The flux linkage \u03c8a, contributed to the magnet torque as shown in Eq.(1), is 0.193Wb compared with a conventional motor\u2019s flux linkage of 0.143Wb even though the stator outer diameter is very small. The reasons are that: \u2013 The permanent magnets are shallow in the core of the developed IPM motor according to Chaper III-A. Fig. 10. Rotor used magnets buried deeply TABLE II SPECIFICATION OF MOTORS developed a conventional IPM motor IPM motor Rotor figure Fig.3 Fig.10 Stator outer diameter 90.0mm 105.0mm Stator inner diameter 51.0mm 56.0mm Number of stator slot 24 24 Magnet Nd-Fe-B Nd-Fe-B Stator stack length 56.0mm 50.0mm Rotor stack length 56.0mm 50.0mm Number of turns of stator winding N 40 27 flux linkage \u03c8a in Eq.(1) 0.193Wb 0.143Wb d-axis inductance Ld(at 3A) 16.6mH 4.35mH q-axis inductance Lq(at 3A) 45.1mH 11.3mH Saliency ratio(Lq /Ld) (at 3A) 2.72 2.60 | Ld-Lq | (at 3A) 28.5mH 6.95mH Fig. 11. Developed motor and a conventional motor The efficiencies of the developed IPM motor and a conventional IPM motor are measured" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003646_amm.86.889-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003646_amm.86.889-Figure2-1.png", + "caption": "Figure 2. CNC machining center configuration", + "texts": [ + " In tool path planning, various aspects should be taken into consideration: (1) the geometric accuracy and surface quality of the machined surface, (2) the time for machining, and (3) the configuration of the machine tool for manufacturing. For this study, the material of the work piece was 40CrMnTi and the carbine tools are used. Machine Tool Configuration. To manufacture the long-cone-distance gears, at least four-axis (one-axis motion for the rotary table and three-axis motion for the cutting tool) controls are required for NC machining by one set-up. All the four axes can be simultaneously controlled, as figure 2 shows. Machining strategy. To obtain the geometric accuracy and surface quality of the machined surface, volume is removed by several processes: (1) rough cut with several flat end mills; (2) semifinish cut with several ball end mills; and (3) finish cut with a ball endmill.To minimize the machining time, a larger tool is desired for the rough cut and semi-finish cut. The process plan is shown in table2. manufactured based on CAD/CAM system. The tool path of machining tooth surface is as figure 3 shows" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001600_ecc.2013.6669841-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001600_ecc.2013.6669841-Figure5-1.png", + "caption": "FIGURE 5. Antagonistically actuated VSA system with a reaction wheel.", + "texts": [ + " We employed two additional testbeds to benchmark the performance of our system: a single-link VSA only system and a single-link system actuated by a reaction wheel only. The dynamics of the latter system is identical to the combined system except the elastic torque \u03c4E,1 = 0. For the single-link VSA system, the 1\u00d7 1 dynamic matrices are M = [ml1(L l c,1) 2 + mbL2c,b + J l 1 + Jb], (20a) C = [0], (20b) B = [bl1], (20c) N = [g(ml1L l c,1 + mbLc,b) cos(\u03c81)]. (20d) B. EXPERIMENTAL SETUP The robot described in the previous section was designed in Solidworks and built as an experimenteal testbed (see Fig. 5). Most of the parts were 3D-printed using UP Plus2 rapid-prototyper. The joint shafts and the link were CNC-machined out of steel and textolite, respectively. Reaction wheel was realized using a bicycle brake disk and connected to the link via low-friction ball bearings. Due to their high torque and speed, two Dynamixel MX-28 servomotors were used to actuate the joint 1 through NEEs. VOLUME 4, 2016 4623 These servos were interconnected in a chain fashion and communicated with the computer via a USB2Dynamixel module" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001973_ijaisc.2011.042716-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001973_ijaisc.2011.042716-Figure4-1.png", + "caption": "Figure 4 Robot\u2019s modelling, plane ( , , )O x y", + "texts": [ + "R O x y z= An omnidirectional robot (Figure 3(a)\u2013(d)) is a robot with two speeds: translational speed according to the axes x and z and the rotational speed around the axis y . Consequently, this robot can move freely in all the directions. This advantage is possible only with Swedish wheels (or offcentred orientable). As shown in Figure 3(d), we can produce an omnidirectional robot by having three Swedish wheels on the tops of an equilateral triangle (Campion et al., 1996). The robot itself is associated to a reference mark ( , , , ).R O x y z\u2032 \u2032 \u2032 \u2032 \u2032= As shown in Figure 4, this later is identified by a vector (\u03be = (x, z, \u03b8)) called posture or situation of the robot, where (x, z) is the position of the robot in the plane ( , , )O x z (the abscissa and ordered of the point O\u2032) and \u03b8 is the angle formed between the axis x of the environment mark and x\u2032 of the robot mark. The axes y and y\u2032 are equal. For the simulation, we employ the open-source simulator Sinbad (Hugues and Bredeche, 2006). The environment adopted is a field (9 \u00d7 9 m2) referred by the reference mark ( , , , )" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003611_s11204-011-9112-1-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003611_s11204-011-9112-1-Figure4-1.png", + "caption": "Fig. 4. Diagram used for analysis of pile subject to horizontal load in classical Rankine half space (a) and stress state of soil on pile shaft when functioning in accordance with scheme (b).", + "texts": [ + " Moreover, the upper zone of the soil increases, and is displaced downward with increasing load. The pile simultaneously experiences a certain rotation, which, as is indicated below, is one of the important factors that must be considered in developing a method for the analysis of piles. It is also established from the experiments that the dependence of pile displacement on load is nonlinear [3]. It is most convenient to examine laws governing interaction of the \"pile-soil\" system in a classical Rankine half space (Fig. 4) with consideration given to the need established in [1] for evaluation of the resistance of a pile in all steps of its horizontal loading. A pile in a Rankine half space can, in first approximation, be treated as a plane rod to which a horizontal force Q is applied to its head section. The point where this force is applied can be adopted as the origin of coordinates x, z (see Fig. 4). In the general case, Rankine's solution for the stress state of a medium in a half space bounded by a slope \u03b5 assumes the following form: (2) where \u03b3 is the specific weight of the soil medium, \u03b5 is the surface slope of the medium, and q is the load distributed over the surface of the soil. It follows from (2) that for each plane of the slope, it is possible to omit the overlying section of the soil medium, and treat this plane as the boundary surface. When the surface is horizontal, \u03b5 = 0, solution (2) is simplified, and the paths of the principal stresses, as before, coincide with the slip lines that parallel the surface of the soil medium", + " 1 sin ( ) ; 1 sin z x c z z q \u03c9 \u03d5\u03c3 \u03b3 \u03c3 \u03c9 \u03d5 \u2212 = + + + 2 2 2 2 2 2 cos (1 )(1 ); cos cos cos (1 ) tan ; cos cos sin cos sin sin , x z xy z q z q \u03c3 \u03b3 \u03b5 \u03bb \u03bb \u03c3 \u03b5 \u03d5 \u03b3 \u03b5\u03c4 \u03bb \u03b5 \u03b5 \u03d5 \u03bb \u03b5 \u03c9 \u03d5 \u03b5 \u23ab += \u2212 \u00b1\u23ac \u23ad += \u2212 = + \u2212 It is accepted to consider that the coefficient \u03c9z is a switch: \u03c9z = +1 for the active pressure, which corresponds to the minimum value of the fractional term in (3), and \u03c9z = \u22121 for the development of passive soil support, i.e., corresponding to the maximum stress state. In truth, the coefficient \u03c9z should be treated as a plasticity index that varies from 1 to \u22121. Here, it is possible to assume that when a load Q is applied to the head of a pile at point O (see Fig. 4), it rotates about a certain point A removed from the by head by a distance z = h. In this connection, the maximum stress state (\u03c9z=0 = \u22121) should develop within the pile essentially at point O on plane z = 0. This is explained by the fact that the value of h will vary during loading, increasing from h 0 to values corresponding to maintenance of limiting equilibrium. The minimum value of the stresses (\u03c9z=h = 1) will be maintained at point A, since the soil at this point is virtually unconsolidated", + " At point B, which corresponds to z = h/2, therefore, it is possible to set \u03c9z=h/2 = 1 on the assumption of the indicated linearization of the \u03c9z values from z. The existence of cohesion c for the soils can be modeled in the calculations by the application of a load q = c/tan\u03d5 over the surface of the soil medium under consideration [4]. When z < h, therefore, the value of \u03c3c in [3] should be determined with consideration given to this situation (3') The stresses on the shaft of the pile \u03c3xO,, \u03c3xA, and \u03c3xB at points O, A, and B, respectively (see Fig. 4), of the pile shaft can be assumed as: (4) Permitting the possibility of linearization and determination of the variation in stresses between the indicated points, the total resistance F of the soil when a load Q is transmitted onto the pile (5) The law describing the stress state of the soil within the surface of a pile and in the near-pile space also corresponds to experimental data derived from measurement of these stresses; this is apparent from Figs. 3 and 4. According to (4) and (5), we have (6) when c = 0", + " 2 2 2 4 xO xB xB xA xO xB xA h h F \u03c3 \u03c3 \u03c3 \u03c3 \u03c3 \u03c3 \u03c3 + +\u239b \u239e= + = + +\u239c \u239f\u239d \u23a0 1 sin ; ; 1 sinxO c xA ch \u03d5\u03c3 \u03c3 \u03c3 \u03b3 \u03c3 \u03d5 \u2212= = + + 1 sin 1 when 1. tan 1 sin z c z z c \u03c9 \u03d5\u03c3 \u03c9 \u03d5 \u03c9 \u03d5 \u239b \u239e\u2212\u2245 \u2212 \u2245\u239c \u239f+\u239d \u23a0 The exact value of a can be determined, proceeding from the principle of the displacement of gravitating load platforms in accordance with (5). The largest of these values can be adopted for safetymargin analyses. For known F and a, the maximum bending moment Mmax in the pile when loaded horizontally Mmax = ahQ. (7) The displacement U of the pile head under a load Q can be determined with respect to the displacement of point B (see Fig. 4) due to the bending stresses acting on plane z = h/2 as compared with their minimum value (8) which had previously been specified, where \u03c3xB is the stress at point B according to (4), \u03c3x,min is the previously effective stress corresponding to the minimum stress state in plane z = h/2, as determined from (3) when z = h/2 and \u03c9z = 1, b is a dimension of the zone of propagation of the bending stress state in plane z = h/2, and E is the compression modulus of the soil. The dimension b of the zone can be determined from the equilibrium condition of prism OBCD (see Fig. 4) (9) where F0 is the resultant supporting force of the soil against face OB of prism OBCD, Fx,min is the resultant force of the minimal stress state on face CD of prism OBCD. In sandy soils: (10) The displacement U of the pile head in conformity with (8), (9), and (10) (11) in sandy soils. Using (5-7) and (11), it is therefore possible to evaluate both the maximum bending moment in the piles when acted upon by a horizontal load Q, and also the displacements of their head section. For this purpose, it is necessary to know the depth of the point at which the pile is fixed in the soil, and also the support FK of the soil in the lower section of the pile", + " The actual soil support in each specific case will be smaller, however, since it is realized by the soil only when necessary. 22 1 sin 2 1 8 tan 1 sinB h U U E \u03b3 \u03d5 \u03d5 \u03d5 \u239b \u239e\u2212= = \u2212\u239c \u239f+\u239d \u23a0 2 0 ,min 1 1 sin ; . 2 2 2 1 sinxB x h h F F \u03d5\u03c3 \u03b3 \u03d5 \u2212\u239b \u239e= = \u239c \u239f +\u239d \u23a0 0 ,min 0 ,min tan 2 or , tan 2 x x h F b c F F F b h c \u03b3 \u03d5 \u03b3 \u03d5 \u239b \u239e= + +\u239c \u239f\u239d \u23a0 \u2212 = + ,min , 2 xB x BU b E \u03c3 \u03c3\u2212 = For actual piles, soil support will be realized not in the region of the lower end, but over the entire length of section KA of the pile (see Fig. 4). In evaluating the specific-pressure distribution of the support along section KA, it is possible to state that its value will increase with depth from zero at point A. If the increase is assumed to be linear, therefore, the resultant support FK will be situated at a distance (H \u2212 h)/3 from the lower end of the pile (where H is the depth to which the pile is embedded) on the condition that the horizontal stress \u03c3K around the end of the pile will not exceed the maximum value determined from (3)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure21.1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure21.1-1.png", + "caption": "Fig. 21.1. Schematic exposition of the situation in MIS: The instrument is moved around an invariant fulcrum point. In consequence the surgeon can command only four degrees of freedom (\u03b1, \u03b2, \u03b3, l) inside the patient\u2019s body.", + "texts": [ + " In this section the peculiarities of manual minimally invasive surgery (MIS) are described and advantages as well as disadvantages are discussed. Subsequently, a short introduction in minimally invasive robotic surgery (MIRS) is given which illustrates the research needs. Minimally invasive surgery is an operation technique which was established in the 1980s. In contrast to conventional, open surgery there is no direct access to the operating field and the surgeon employs long, slender instruments. These are inserted into the patient through narrow incisions which are typically slightly bigger than the instrument diameter (see Fig. 21.1). The main advantages of MIS, compared to open surgery, are reduced pain and trauma, shorter hospitalisation, shorter rehabilitation time and cosmetic advantages. However, MIS is faced with at least three major disadvantages [1]: (a) As the surgeon does not have direct access to the operating field the tissue cannot be palpated any more. (b) Because of the relatively high friction in the trocar1 and due to the torques which are necessary to rotate the instrument around the entry point, the contact forces between instrument and tissue can hardly be sensed. This is especially true when the trocar is placed in the intercostal space (between the ribs). (c) As the instruments have to be pivoted around an invariant fulcrum point (see Fig. 21.1), intuitive direct hand-eye coordination is lost. Furthermore, due to kinematic restrictions only four degrees of freedom (DoF) remain inside the body of the patient. Therefore, the surgeon cannot reach any point in the work space at arbitrary orientation. This is a main drawback of MIS, which makes complex tasks like knot tying very time consuming and requires intensive training [2, 3]. As a consequence MIS did not prevail as desired by patients as well as by surgeons and while most standard cholecystectomies (gall bladder removal) are performed minimally invasively in the industrialised world, MIS is hardly used in any other procedure to this extent" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003219_amm.87.30-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003219_amm.87.30-Figure2-1.png", + "caption": "Figure 2. ADAMS simulation model", + "texts": [ + " In this vibration mode, the energy which is imposed on the jth generalized coordinate vector is ( ) ( ) ( ) 6 2 1 1 , 2 ij ni i i k e j k M j k\u03c9 \u03d5 \u03d5 = = \u22c5 \u22c5\u2211 (12) So Eij is used to represent system energy distribution as ( ) ( ) ( ) [ ] [ ] [ ] 6 1 , 100% i i ij k ij T i i i j k M j k e E e M \u03d5 \u03d5 \u03d5 \u03d5 = \u22c5 \u22c5 = = \u00d7 \u22c5 \u22c5 \u2211 (13) if the value of Eij is relatively large, it indicates that the system movement under the ith vibration mode is dominated by the jth generalized coordinate vector\u2019s mode type, and it also indicates that the system energy decoupling level is high. 3.1. ADAMS Model Establishing. Simulation model is established in ADAMS/View module in the ADAMS as shown in Fig. 2. In Figure 2, the model coordinate system is consis-tent with powertrain centroid coordinate system, the direction of gravity is set up as the opposite direction of z , and the powertrain is replaced by a body (Part) with mass and rotary inertia. The ground (Ground) is used to represent the rigid carframe, and the mounting part connecting the powertrain and the carframe is replaced by a connector (Connector Bushing) which is provided with three axial stiffness and zero axial damp, zero rotary stiffness and zero rotary damp [11]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002384_2011-01-1424-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002384_2011-01-1424-Figure3-1.png", + "caption": "Fig. 3. Geometry of Test Disc and Rollers", + "texts": [ + " The rotational speed and torque, of both the module mainshaft (containing the disc SAE Int. J. Engines | Volume 4 | Issue 12136 and rollers) and the module lay shaft, were measured. The RMS vibration level of the variator module was measured using an accelerometer. All test rig data including traction fluid pressure, temperature and flow rate were recorded via a Keithley data acquisition system. The geometrical dimensions of the test pieces (disc and rollers) used in the current work were as shown in Figure 3. Disc and rollers were made from KUJ7 material [14] with Ra<0.1 A Shell prototype traction fluid was used for all tests. Rheological properties of this fluid are currently confidential. Traction fluid was used to lubricate, cool and control the variator. A separate hydraulic power pack enabled independent supply of the roller carriage piston reaction pressures, endload and lubrication cooling flows. A water to oil heat exchanger supplied with chilled water controlled the temperature of the traction fluid lubrication cooling flow" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003213_s10711-011-9606-z-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003213_s10711-011-9606-z-Figure5-1.png", + "caption": "Fig. 5 \u03c91", + "texts": [ + " , hg be the 1-handles attached to a 3-ball B3 as shown in Fig. 4. Let d1, d2 be two disks properly embedded in Hg such that d1 \u222a d2 = B3 \u2229 h1, and Hg\u22121 = Hg \\ (h1 \\ (d1 \u222a d2)). Under the parametrization [0, 1] \u00d7 D2 of h1 such that {0} \u00d7 D2 = d1, {1} \u00d7 D2 = d2, we denote {1/2} \u00d7 D2 by m1. The map \u03c1 is the rotation of Hg on itself such that \u03c1 brings h1 to h2, h2 to h3, . . ., and hg to h1. Let c1 be the arc connecting \u2202d1 and \u2202d2 indicated in Fig. 4, and K1 the regular neighborhood of h1 \u2229 c1 in Hg . The map \u03c91 is a half-twist of K1 indicated in Fig. 5. We define \u03c92 for h2 in the same manner as \u03c91. Let be a disk properly embedded in Hg whose boundary is a circle indicated in Fig. 4. This disk divide Hg into a handlebody H2 of genus 2 and a handlebody of genus g \u2212 2. Fig. 6 a1,2 and b1,2 Fig. 7 A lantern among non-separating disks By twisting H2 through \u03c0 in the clockwise direction, we have a homeomorphism \u03c1\u2032 1,2 on Hg such that \u03c1\u2032 1,2(h1) = h2, \u03c1 \u2032 1,2(h2) = h1 and \u03c1\u2032 1,2( ) = . The homeomorphism \u03c11,2 is defined to be the composition \u03c11,2 = \u03c91\u03c92\u03c1 \u2032 1,2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003007_s00542-010-1211-9-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003007_s00542-010-1211-9-Figure2-1.png", + "caption": "Fig. 2 Inertia and rotating coordinates of the journal and thrust bearings", + "texts": [ + " The validity of the proposed method was investigated using the transient motion of a rotor with respect to the inertia coordinates, as well as the rotating coordinates. It also investigated the stability of the disk-spindle system due to the whirl radius, rotating speed, and tilting angle. 2.1 Determination of dynamic coefficients This study extended the method of Jang and Yoon (2003) for FDBs with a rotating groove by including rotational degrees of freedom in the perturbed equations in order to investigate the moment coefficients and tilting effect. The coordinate system of the journal bearing and the thrust bearing with rotating grooves are shown in Fig. 2. The governing equations for the journal and thrust bearings were obtained by transforming the Reynolds equation into the hz and rh planes, respectively. o RoH h3 12l op RoH \u00fe o oz h3 12l op oz \u00bc R _h 2 oh RoH \u00fe oh ot \u00f01\u00de o ror r h3 12l op or \u00fe o roh h3 12l op roh \u00bc r _h 2 oh roh \u00fe oh ot ; \u00f02\u00de where R is the radius of the journal, _h is the rotational speed of the shaft, h is the film thickness, p is the pressure, and l is the viscosity coefficient. The perturbations applied to the rotor at a quasiequilibrium position are shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002114_amr.291-294.1195-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002114_amr.291-294.1195-Figure1-1.png", + "caption": "Fig. 1. Three-dimensional assembling demonstration", + "texts": [ + " Thus, under the condition of that the basic dimension is not changed, by conducting optimization of profile modification and parameters optimization on traditional structure, the axial dimension is increased by width of one cycloid gear, transmitting torque is increased about 50%, and the goal of large capacity with small size is realized. The structure of three cycloid gears is suitable for those cases of large speed ratio. Japan FA adopts the structure of three cycloid gears, of which the driving ratios are i=29, 59, 89, 119 respectively. See Fig. 1. No-clearance meshing of standard gears and analysis of meshing between pin gear and cycloid gear Discussion of Deformation Coordination Conditions Fig. 2 is a sketch of force analysis. Suppose the pin gear is stable, a moment cT is stressed on the cycloid gear; the cycloid gear rotates by an angle of \u03b2 due to the elastic deformation of driving part. If the deformation of cycloid gear, pin gear sleeve (or pin gear) and rotating arm is ignored, the bending of gear pin and total deformation of gear contacting and squeezing are calculated" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003946_icicip.2014.7010281-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003946_icicip.2014.7010281-Figure4-1.png", + "caption": "Fig. 4. Stator voltage sv and rotor current ri vectors in a transient states", + "texts": [ + " In some special cases, the PLL may be unstable, and the \u03c9ir * is oscillating back and forth. To avoid this problem, the reference angle \u03b8vs * is modified as Fig. 3. The \u03b8vs * plus 2\u03c0 if the change of \u03b8vs is greater than \u03c0. On the contrast, the \u03b8vs * mins 2\u03c0 if the change of \u03b8vs is smaller than -\u03c0. Setting the saturation values to \u03b8vs * with \u00b12\u03c0 and e\u03b8 with \u00b1\u03c0 [21]. As a result, the error e\u03b8 is a monotonic variable on the actual value of the voltage vector angle. A vector diagram presenting the dynamic state of the stator voltage and rotor current vector is shown in Fig. 4. For a fixed rotor speed and load, there is a basic phase and length of the rotor current vector rbasei responsible for generating the reference stator voltage * sv . As can be observed in Fig. 4, Changing the mechanical speed or load causes a displacement of the rotor current vector from rbasei to ri , and this would move the stator voltage vector from * sv to sv . To obtain * sv again, the phase and length of the rotor current vector have to be changed. Length of the rotor current vector is adjusted by the outer voltage PI controller. Phase of the rotor current vector is changed by regulating the rotational speed of the reference rotor current vector \u03c9xy *, which can be described as * * xy m irp (7) The reference rotor current angle speed \u03c9ir * is adjusted by PLL until e\u03b8 is zero" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001965_978-3-658-05978-1_7-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001965_978-3-658-05978-1_7-Figure1-1.png", + "caption": "Figure 1: Life-drive module vehicle architecture.", + "texts": [ + " Therefore, it combines the best of two worlds: A silent and emission free electric drive with an efficient and powerful combustion engine ensuring long traveling distance ability [1]. In urban areas access restrictions, which are expected in the future will be passed with an emission and noise free electric drive of 96 kW and a maximum range of 30-35 km. On roads and highways the combustion engine with 164 kW assists the electric drive to operate with a maximum power of 260 kW and a maximum torque of 550 Nm, which conveys a pure sports car characteristic. To obtain efficiency and sportiness at the same time, a new purpose built vehicle architecture according to figure 1 is suggested. To compensate the mass increase, which comes along with electrification of the powertrain, a new so called life-drive structure design is engaged that is characterized by the employment of lightweight materials [2]. The result is a low weight vehicle, a high driving range, a spacious passenger compartment and an agile driving performance. The central element of the life module is the passenger compartment made out of carbon fiber. The life module integrates the high voltage battery by the energy tunnel and is mounted to the drive module made out of aluminum, which will hold all the power train and vehicle dynamic components including the high voltage components" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001197_12.2189458-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001197_12.2189458-Figure7-1.png", + "caption": "Figure 7. Specimen schematic.", + "texts": [ + "aspx Surface deformation monitoring system is shown in the Figure6. The central wavelength shifts of FBG sensors, which were stuck on the cell walls of flexible honeycomb structure, should be recorded by optical demodulation when the specimen was loaded current and deformed. The center wavelengths of FBG1, FBG2,and FBG3 were 1560nm\u3001 1555nm\u30011545nm respectively. The current excitation applied to the specimen was 2A.The honeycomb core from the fixed end to the free end was marked from No.1 to 8. The specimen with FBG sensors is shown in the Figure7. FBG1, FBG2 and FBG3 were stuck respectively on the honeycomb core walls. Considering the linear relation between central Bragg wavelength shifts and the curvature which can be deduced from the central wavelength shifts of FBG1, 2, 3. According to the surface reconstruction algorithm, the curve of the adjacent FBG sensors location can be inferred from the values shift which is the starting point relative to the previous sensor coordinate of the endpoint. Using interpolation functions to calculate the reconstruction of the honeycomb structure surface" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003982_saci.2014.6840063-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003982_saci.2014.6840063-Figure3-1.png", + "caption": "Figure 3. The transportation unit", + "texts": [ + " III. CONSRUTION OF THE PNEUMATIC SYSTEM The pneumatic system of the device can be divided for 4 main units. 3. The batching The first pneumatically unit is performing the batching of the specimens to the conveyor. The linear actuator is simple pushing out the specimen from feeder tube. The 4/2 pneumatic valve is actuated with a solenoid, see Fig.2. 4. The transportation The second pneumatically unit is responsible for the specimen transportation from the feeder unit to the gripper position (see Fig. 3). Here the cylinder moving is insured by a 3/2 and a 2/2 valves, and the direction of the moving is assured by an OR valve. \u2013 214 \u2013 5. Control of the pneumatic manipulator The next (third) pneumatic module is belonging to the gripper manipulator. The manipulator is consist from a linear actuator (cylinder), which is lifting and lowering the manipulator with a 4/2 electro-pneumatically valve, and a rotational \u201cpneumatically motor\u201d, what is turning the gripper in 900 degrees, from the conveyor position to the selected spout position", + " \u2013 215 \u2013 Conditions: has to be minimally 1 specimen in feeder and the cylinders H1, H2, H3 and F have to be in starting positions. (see Fig.2) Effect: H1 cylinder will move to the position 1 with the help of solenoid Y1, and after, H1 immediately returning to the position 0, and Y1 will be inverted. 8. Transportation During the transportation the sensors A\u00c91, A\u00c92, A\u00c93 are determining the materials (steel/copper/plastic), the control mechanism evaluating the sensory data and base on evaluation sending the position command to the slip-way unit. (see Fig.3) 9. Gripper lowing End of transport 10. Gripping It is mean that H3 cylinder is in lower position. The gripper is closed. 11. Gripper lifting Gripper is closed (see Fig. 4) 12. Manipulator rotating Cylinder H3 is in the upper position. The gripper is above slipway. 13. Gripping It is mean that H3 cylinder is in lower position. The gripper is closed. 14. Specimen releasing The gripper is above the slip-way. The gripper is releasing the specimen and the specimen is slipping into the correspondent storing unit" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003671_detc2011-48794-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003671_detc2011-48794-Figure5-1.png", + "caption": "Figure 5. Main dimensions of the toroidal CVT drive shown on the top half of the cross-section (The global Z-axis is perpendicular to the plane of the paper).", + "texts": [ + " Go to the beginning of step 2. The model of a typical toroidal CVT drive used in automotive applications that was presented in [6] is modeled using the present method. Figure 4 shows the toroidal drive model. It consists of an input toroidal disk, an output toroidal disk, two rollers, and a two roller angle controllers. The input shaft, output shaft, and roller angle controllers are connected to ground using revolute joints. The rollers are mounted on the roller angle controllers using revolute joints. Figure 5 shows a sketch of the system along with the major system dimensions. Table 2 shows the values of the dimensions, mass/inertia and discretization parameters of the toroidal drive model. In this system the input/output speed ratio can be adjusted from 0.5 to 2.0 by changing the roller angle \u03c6. In this simulation the speed ratio is kept constant and set to 1, i.e. \u03c6 = 0. There is a very small penetration of the rollers into the toroidal disks. This penetration results in an elliptical contact patch between the roller and the toroidal disk (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002525_j.phpro.2013.03.119-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002525_j.phpro.2013.03.119-Figure3-1.png", + "caption": "Fig. 3. Drawing of the proposed half-hohlraum", + "texts": [ + " This tube has to be implemented with high level requirements: the tube / hohlraum junction must be gastight (with a leakage flow rate better than 3 x 10-5 mbar.l.s-1 - measured under a helium pressure of 1 bar), this tube has to be oriented into a narrow virtual cone (under an acceptance angle of \u00b1 2 degrees). Figure 2 shows targets as they are manufactured today (i.e. with a glued filling tube). Hohlraums are made of two distinct materials: In order to simplify the overall manufacturing process, we are developing a potential version excluding this resin. In that goal we have proposed an adapted design for the hohlraum (given figure 3), with a gold thickness increased up to 100 \u03bcm. This thickness allows to drill a counterbored hole, in order to insert the A laser welding solution has been studied to bond the filling tube onto the gold hohlraum and to fulfil specifications quoted above (gas tightness and spatial orientation). In that frame, we use an infrared pulsed Nd:YAG laser usually dedicated to aluminum welding [9]. Apart from the reflectivity and high thermal conductivity of gold, one additional difficulty is induced by the target geometry", + "K-1 for gold), that compensates for the small mass of filling tube compared to the hohlraum. Thermal load evacuation was then possible through the filling tube, which was not altered. Tantalum, stainless steel, nickel and platinum have a higher boiling point but have a thermal conductivity at least three times lower than gold (maximum 0.9 W.cm-1.K-1 for nickel). So, the heat provided by the laser pulse was not dissipated and vaporized the tubes. In the second stage, nominal gold hohlraums were used (as given figure 3). First, the pulse energy has been decreased down to 1.45 J (peak power of 400 W and pulse length of 5 ms) because the molten pool was too deep with parameters determined above (first signs of drilling have been observed). It can be explained by the thermal mass of gold, far lower for final workpieces (hohlraum) than for test mandrels. The pulse shape has also been modified to improve the join (smooth welded seam, without any cracking). The other parameters (overlaps, focal length, beam diameter, pulse shape) have not been modified" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002235_humanoids.2011.6100821-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002235_humanoids.2011.6100821-Figure10-1.png", + "caption": "Fig. 10. Object and object convex polygon (OCP)", + "texts": [ + " Pdi is defined as the vector having the direction of approach to the object and is an outer unit normal vector of a GRC surface. We define maximum and minimum mass, mmax,i and mmin,i of the grasped object using the i-th grasp type. Given the object shape to be grasped, our planner calculates the object convex polygon (OCP) including the grasped object in object coordinate system. In this paper, we consider the rectangular box as the OCP. For complex object shape, we split the object into several regions and calculate the OCP for each region. As shown in Fig. 10, Our planner splits a vase into three regions. Our planner calculates the eigenvectors of the co-variance matrix of the point set on object surface as OCP axis. Position/orientation of the i-th OCP is defined as p \u00b7/R . Ot 01- (i = 1, . . . ,m), the edge length vector is defined as eoi. To select grasp types and determine the nominal posi tion/orientation of the palm, we introduce some heuristic rules using geometrical relationship between the GRC and the OCP as below. Let sort( a) be the function sorting the elements of the vector a in decreasing order" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003013_j.snb.2011.07.064-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003013_j.snb.2011.07.064-Figure3-1.png", + "caption": "Fig. 3. Schematic representation of the multi-layered structure of the fabricated electrode in this research. Whole areas of electrode and enzyme membrane were covered with the enzyme membrane and enteric coat, respectively.", + "texts": [ + " A Au or Pt wire 1 mm in diameter and 40 mm in length was overed in a heat shrinkage tube, and the edge was polished with 00-, 4000-, and 15,000-grade emery paper and with an agglomrated gamma alumina suspension on a polishing cloth. Lastly, the ires were washed by ultrasonication. The DEP chip was used withut surface cleaning. One microliter of enzyme mixture was diluted en-fold with water and was dropped on the edge of a Au/Pt wire r on a working electrode of a DEP chip and was immobilized in lutaraldehyde vapor at room temperature for 2 h, and the enzyme lectrode was obtained (Fig. 3). The enzyme electrode was dipped into the enteric coat solution nd was naturally dried for approximately 20 min. For the Macrool, the mixed solution of 50 mL of Polyquid, 48.5 mL of water and .5 mL of Macrogol 400 was used for dipping. Artificial gastric and intestinal juice treatment followed the ethods of the Japanese pharmacopeia. Two grams of sodium chloide and 7.0 mL of hydrogen chloride were mixed and diluted to 000 mL to obtain the artificial gastric juice. For artificial intestinal uices, 250 mL of potassium dihydrogenphosphate at a concentraion of 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure28.8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure28.8-1.png", + "caption": "Fig. 28.8. Slave side: a) objects for haptic exploration b) screw and screwdriver", + "texts": [ + " Such a setup provides the operator with a realistic visual information about the location of the objects, the environment, and the telemanipulator. Here, the anthropomorphic construction of the telemanipulator plays an important role: the operator can drive it as if it were his/ her own arm. The visual information is useful not only for motion generation but also for handling the contact and minimizing effects of the impact. The experiment consists of three tasks: \u2022 tracking of free space motion \u2022 haptic exploration of different materials (soft and stiff), see Fig. 28.8a \u2022 driving a screw with an aluminium tool, see Fig. 28.8b. This last experiment consists of three phases: contact with extreme stiff materials, a classic peg-in-hole operation and manipulation in a constrained environment. Fig. 28.9 and Fig. 28.10 show the position and force tracking performance during haptic exploration of different materials (see Fig. 28.8a). The shaded areas indicate the several contact phases. One can see that during free space motion, the position tracking of the slave arm works very well while in the contact situation, as a consequence of the implemented impedance controller, the slave position differs from the master position. Please note that, as the force tracking is very good, this position displacement influences the displayed and felt environmental impedance in such that hard objects are perceived softer then they are. As the master controller is of admittance type, which reacts on the human force input, non zero forces (forces depend on the minimal master dynamics) are necessary during free space motion to change the actual end-effector position" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003814_iciea.2013.6566511-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003814_iciea.2013.6566511-Figure3-1.png", + "caption": "Fig. 3. DH coordinates of the 7-DOF space manipulator", + "texts": [ + " The curves of the optimal solutions and the average population values are obtained using the MATLAB GA toolbox. And then the global optimal solution and the corresponding optimal parameters can be obtained. On this basis, the joint angle, joint angular velocity and joint acceleration expressed are solved, and the optimal path is obtained. IV. Units Simulation Results and Discussions The system is consisted of 7-DOF manipulator and spacecraft base, as illustrated in Fig. 2. The DH coordinates of the system is shown in Fig. 3, and DH parameters are list The corresponding polynomial parameters are as follows: Table . B. Simulation results During numerical simulation in point-to-point task, the corresponding polynomial parameters are as follows: The initial joint angle is set as: Taking the polynomial coefficient as controlling parameter and combing SOA dynamic equations, GA is used to optimize the joint space trajectory to achieve maximum load carrying capacity. The relevant parameters about load and GA are as follows: loadm : 400Kg" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001312_icuas.2015.7152309-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001312_icuas.2015.7152309-Figure3-1.png", + "caption": "Figure 3: Vane pitching to induce a yawing moment on the device.", + "texts": [ + " Where the origin is coincident with the center of gravity of the device and the axes of motion in which the behavior of the device is to be analyzed. The drag on the device is increased by deflecting the vanes by the collective deflection angle, \u03b2c, shown in figure 2. In this way, the device can adjust its descent velocity. Maximum deflection of the vanes is 67\u00b0. To achieve the nominal descent velocity of 54ms-1, the device is designed that the vanes need only to deflect 20\u00b0. The yaw axis is controlled by pitching the vanes about their longitudinal axes, \u03b3 (shown in figure 3). This produces a net moment causing the device to rotate. Range of vane rotation is \u00b110\u00b0. Induction of this yawing moment allows the device to alter its heading. A Pitching moment is induced by the retraction or extension of the two rear vanes, and the opposite movement of the forward vane (shown in figure 4). Thus the differential of lift and drag forces acting on the rear and fore of the device increases, causing the device to keel in the desired direction. To induce a moment in the roll axis, the vanes are adjusted by an amount \u03b2d\u03d5 which is mapped so that the device will roll in the direction of the retracted vane, and away from the extended vane, as is shown in figure 5" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003959_s105261881101002x-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003959_s105261881101002x-Figure2-1.png", + "caption": "Fig. 2. Scheme of the measuring bench.", + "texts": [ + " Nevertheless, the proportionate relationship between the unbalance vector d and position vector r established by the method (1) allows one to refine each of the multipliers M and \u03b5 if specific changes are made to the unbalance vector d, with the respective radius r being simultaneously recorded. Here, \u03bb is a dynamic proportionality coef ficient determined from the bench properties. DESCRIPTION OF THE MEASURING BENCH AND THE PROPOSED METHOD The core of the bench is the rotating platform 1, whose motion is communicated from the motor through the driving shaft 2 with the double cardan joint 3 (Fig. 2). Platform 1, together with base 4, three supports 5, and flexible joints 6, form an oscillating system similar to an astatic pendulum. To position and retain sample 7, and shift it by an arbitrary yet known length, the platform has carriage 8, which can run in guide bearings. In the initial position, the inherent center of mass of the carriage S1 falls on the rotation axis of the platform, and the mass of the carriage m is known and plays the role of a trial mass. The dis placement of the carriage from axis 2 and the coordinate origin can be determined by a measurement tool (not given in Fig. 2). Under the carriage, the balancing weight 9 with the mass m (the same as that of the carriage) travels in similar guide bearings. The weight is intended to equilibrate the centrifugal force from the carriage motion when it shifts from the point of origin W. The weight is transported in the opposite direction for the same distance as the carriage with a double rack and pinion gear whose elements are the carriage, balancing weight, and two gear wheels 10 located on the same axis. They can be blocked and dis connected through mutual motion along the rotation axis, remaining meshed with the racks", + " In the central position, the centers of mass of the carriage S1 and the weight S2 are in the point of origin W and balanced. In this position, the whole system can be out of balance only because of a displacement of the center of mass of the body under study. r L \u03c81/2cos 2 \u03c81/2 \u03c82+( )sin 2 + , \u03b2 \u03c81/2 \u03c82+( )sin \u03c81/2( )cos \u239d \u23a0 \u239b \u239e ,arctan= = \u03bb di ri ,= 104 JOURNAL OF MACHINERY MANUFACTURE AND RELIABILITY Vol. 40 No. 2 2011 ALESHIN Let the unknown position vector of the displacement be R1 in the coordinates W\u03be1\u03b71, and the unknown mass of the body be M. The measurement system for the time intervals (Fig. 2) is formed by photoelectric detector 11 with optical axis 12 and the coordinate axes \u03be1 and \u03b71 modulating the optical path; these axes are plotted on an optically transparent disk 13. The disk is rigidly bound to the platform. Rotation of the platform with the frequency \u03a9 causes a centrifugal force F1: F1 = \u03a92MR1. The measurement process consists of three con secutive steps. Step 1. Subject to the force F1, the platform with the base (Fig. 2) are displaced by the position vector r1. The phase angle \u03b3 of the lag of the vector r1 from R1 is always constant, provided that \u03a9 and M do not change. The value and position of the radius r1 with the polar angle \u03b21 are determined from three time intervals in the frame W\u03be1\u03b71 by the above algorithm. JOURNAL OF MACHINERY MANUFACTURE AND RELIABILITY Vol. 40 No. 2 2011 A METHOD TO DETERMINE THE MASS AND COORDINATES 105 Step 2. Carriage 8 with fixed body 7 shifts for an arbitrary yet known position vector k whose direction is set by the platform\u2019s guide bearings and is the same as that of the axis \u03b71" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure4-1.png", + "caption": "Figure 4 FACE CUTTING", + "texts": [ + " The theoretical or ideal shape of these crossed axis gears is the \u201chour-glass\u201d or hyperboloidal shape shown. Current design and manufacturing techniques approximate a small portion of the hour-glass shape by a conical segment as shown. This approximation results in the following restrictions: face width\u00f1 minimum number of teeth\u00f1 spiral angle,\u00f1 pressure angle\u00f1 and hence restrictions on candidate gear designs. Moreover, these restriction are compounded with a cradle mounted face cutter as depicted in Figure 4. Face cutting further places restrictions on the above limitations together with the gear ratio. One goal of this paper is to present an alternative method for the fabrication of hyperboloidal gears that overcome the limitation of existing face cutting technology with the following features: fabrication applicable to spur and helical gears,\u00ec addendum & dedendum proportional to tooth pitch,\u00ec inter-changeability of hypoid gear pairs,\u00ec correlation between practice and kinematic theory.\u00ec This face cutter cannot readily be used to produce spur type gears" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.17-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.17-1.png", + "caption": "Fig. 2.17. Connection of a single link in the chain with two links", + "texts": [ + " The above direct kinematics method based on the DH convention exploits the inherently recursive feature of an open-chain manipulator. Nevertheless, the method can be extended to the case of manipulators containing closed kinematic chains according to the technique illustrated below. Consider a closed-chain manipulator constituted by n + 1 links. Because of the presence of a loop, the number of joints l must be greater than n; in particular, it can be understood that the number of closed loops is equal to l \u2212 n. With reference to Fig. 2.17, Links 0 through i are connected successively through the first i joints as in an open kinematic chain. Then, Joint i + 1\u2032 connects Link i with Link i + 1\u2032 while Joint i + 1\u2032\u2032 connects Link i with Link i + 1\u2032\u2032; the axes of Joints i + 1\u2032 and i + 1\u2032\u2032 are assumed to be aligned. Although not represented in the figure, Links i + 1\u2032 and i + 1\u2032\u2032 are members of the closed kinematic chain. In particular, Link i + 1\u2032 is further connected to Link i+ 2\u2032 via Joint i+ 2\u2032 and so forth, until Link j via Joint j" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003259_gt2013-94105-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003259_gt2013-94105-Figure9-1.png", + "caption": "Figure 9: Gear tooth face inspection", + "texts": [ + " In the current designs, the idler inspection wheel is not designed to transmit power. When used for gear wheel inspection it would need to be carried on an idling shaft not linked to the power train. 5 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/asmep/76989/ on 07/08/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use Figure 8: Gear tooth root inspection In a similar way to the root inspection idler sensor, an eddy current sensor is embedded in the idler gear (see Figure 9). However, in this case, the sensor (shown in yellow) points out normal to the mid-face point, and is designed to be a tooth face inspection sensor, inspecting one face only of the power gear. This configuration provides very close proximity scanning across the face of the inspected tooth as each tooth traverses the shank of the mating tooth. Various configurations of the sensor concepts illustrated above were considered in terms of their applicability to the fault types identified. Some of the configurations were deemed to be difficult to implement, therefore were not included in further analysis" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002731_icma.2013.6618005-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002731_icma.2013.6618005-Figure2-1.png", + "caption": "Fig. 2 The thi link.", + "texts": [ + " The goal of our research is to develop a method for the coordination of multiple actuators to appropriately distribute energy to each body of a robot through the kinetic chain according to a desired motion. For this goal, it is important to clarify how to coordinate multiple actuators in order to generate proper energy flow. In this paper, in terms of energy flow in the kinetic chain, we evaluated robot motion generated by the coordination of multiple actuators using simulation. In this paper, we deal with a multi-link robot comprised of n rigid bodies (see Fig. 1 and 2 and Table I). The motion of the thi link is described by the Newton- Euler equations (see Fig. 2 (a)): gFv iGiim (the Newton equations), (1) iii T\u03c9I (the Euler equations). (2) where g is the gravitational acceleration vector. The total mechanical energy of the thi link iE is given as follows: Gi T iii T iGi T Giii mmE pg\u03c9I\u03c9vv 2 1 2 1 . (3) where Gip is the position of the center-of-gravity of the thi link. In a musculoskeletal robot, each artificial muscle actuator is attached around some certain joints, and the joint torque is generated by a group of artificial muscle actuators attached around the joint", + " The mechanical power of the thj artificial muscle actuator jP is obtained by using (5): \u03b8\u03b8GuLuP )(TT , (6) where P is the muscle power vector in which the thj element is jP . Energy flow caused by the thj artificial muscle actuator is as follows [23]: 0jP : energy generation and transfer, 0jP : energy absorption and transfer, 0jP : energy transfer. Energy change rate within a robot link shows energy transfer between robot links and via actuators. The energy change rate within the thi link iE is given as follows: Gi T iii T iGi T Giii mmE vg\u03c9I\u03c9vv . (7) The energy change within a robot link is caused via actuators and joints. Fig. 2(b) shows the force and the moment of force acting on the thi link, and the translational and angular velocity of the thi link. The thi joint connects the th1i link and the thi link. If the th1i link exerts the force if and the moment of force i\u03c4 on the thi link, on the basis of Newton's law of action and reaction, the thi link exerts the force if and the moment of force i\u03c4 on the th1i link. In Fig. 2(b), the force if and 1 if are exerted on the thi link by the th1i link and the th1i link respectively. And furthermore, the thi link moves at the angular velocity i\u03c9 , and the thi and the th1i joint move at the velocity iv and 1iv respectively. In such these cases, the energy change within the thi link is as follows: The power i T i vf is supplied from the th1i link into the thi link (in the case of positive value), or discharged out of the thi link into the th1i link (in the case of negative value), The power 11 i T i vf is supplied from the th1i link into the thi link (in the case of positive value), or discharged out of the thi link into the th1i link (in the case of negative value), The power i T i \u03c9\u03c4 is supplied from a group of artificial muscle actuators attached around the thi joint into the thi link (in the case of positive value), or discharged out of the thi link into a group of artificial muscle actuators attached around the thi joint (in the case of negative value), The power i T i \u03c9\u03c4 1 is supplied from a group of artificial muscle actuators attached around the th1i joint into the thi link (in the case of positive value), or discharged out of the thi link into a group of artificial muscle actuators attached around the th1i joint (in the case of negative value)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001113_gt2015-43561-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001113_gt2015-43561-Figure5-1.png", + "caption": "Figure 5: Side view of model domain showing locations of data comparison through the tooth region and at the shroud restriction and outlet.", + "texts": [ + " As the slot is narrow, ro is close to ri and so the swirl number was simplified to: \ud835\udc46 = \ud835\udc64 ?\u0305? Eq 2 Table 1 shows the cases run in the present study and the associated Swirl Numbers. It is not currently known how these values compare to inside the bearing chamber of an aeroengine. Using the computed flow fields associated with [13] a swirl number of around 1.4 is obtained at shroud inlet and so the data for Case 5 might at this stage be considered most representative. Velocity profiles were compared at several positions in the domain, locations as indicated in Figure 5. The progression of the azimuthal velocity profiles through the shroud inlet (Position 1), the bottom of the teeth (Position 2), top of the teeth (Position 6) and at the shroud outlet (Position 8) are shown in Figure 6. Azimuthal velocity is non-dimensionalised by r (local radius and shaft angular velocity). 4 Copyright \u00a9 2015 by Rolls-Royce plc Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 12/24/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use There is clear evidence that initial large differences in azimuthal velocity become significantly smaller as the fluid is swirled by the gear; with very little difference in azimuthal velocity profile evident by the top of the toothed section (Figure 6c)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003420_j.mspro.2014.06.203-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003420_j.mspro.2014.06.203-Figure5-1.png", + "caption": "Fig. 5. Calculation chart of the problem (\u0430) and paths of an edge crack propgation depending on the friction coefficient f and relative length of the stick region c/a (b).", + "texts": [ + " The body in contact weakened by cracks is modeled by an elastic half plane with cracks (see Fig. 1) and the action of the counterbody is modeled by forces distributed over the contact region. As the normal component of the contact load p(x, , t) = p(x, ), the Hertz pressure (relation (4)) is used. Tangent component in case of complete sliding of the fretting couple is modeled by forces (2). To model the action of the counterbody in the case when a stick region is in the contact zone of the fretting couple elements (Fig. 5a), the distribution of contact pressure is used established independently by Cattaneo and Mindlin (see Datsyshyn (2005)): ,, ;,, 0 2 0 22 0 2 0 00 2 0 2 0 cxxxxcxxaapf axxccxxaxxaapf xq y = 0, (6) where the relative length of the region of stick is specified by the ratio c/a; Q and \u0420 are the tangential and the normal components of the external load vector. It is assumed, that the crack is located outside the contact region ( 1) and its edges are a unloaded. The SIF for an initially rectilinear crack is calculated. It is shown that enlargement of the region of stick leads to the substantial decrease in the SIF KI, KII and their maximum values. The analysis of crack growth paths by mode I fracture (Fig. 5b) shows that the crack growth path deviates from the contact zone as the length of the region of stick increases; the same happens as the friction coefficient f decreases in the case where the region of stick is absent. This conclusion agree with two basic known approaches to the structural and technological enhancement of fretting fatigue strength, namely, the characteristics of strength can be improved by either preventing the displacements of one surface over the other (increasing the stick region) or, vice versa, by minimizing the friction forces and, hence, facilitating the indicated displacements" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003379_amm.630.240-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003379_amm.630.240-Figure4-1.png", + "caption": "Fig. 4. A single-mass rotor model", + "texts": [ + " Analytical expression for stiffness of elastic force, depending on the displacement of the shaft is determined by formula )()0( )( )( \u03b5\u03b2 \u03b5 \u03b5 \u22c5== c k de dF c k , where n nn \u03b5\u03b1\u03b5\u03b1\u03b5\u03b1\u03b5\u03b2 )1(...321)( 2 21 +++++= - dimensionless coefficient of nonlinear stiffness. Dependence of coefficient )(\u03b5\u03b2 is shown in Figure 3. It\u2019s evident that stiffness of elastic force decreases with increasing displacement of the shaft, i.e. this system has a soft characteristic of stiffness. As far as we know, this fact deteriorates the vibration characteristics of the rotor. To study the effect of nonlinear force on the dynamic characteristics of the rotor we consider a single-mass rotor model (Figure 4) with the parameters of the shaft: length 520=l mm and diameter 25=d mm; rotor mass 18=m kg; annular seal geometry: length 48=l mm and radius 25=r mm; average radial clearance 0.30 =h mm, and pressure drop across the gap 1.25\u2206p = MPa. The structure of hydrodynamic forces, arising in the gap sealing, can be assumed as \u22c5+\u22c5\u2212\u22c5\u2212= \u22c5\u2212\u22c5\u2212\u22c5\u2212= ,)()0( ;)()0( xqyr c kyb y F yqxr c kxb x F \u03b1 \u03b1 & & where b - damping coefficient; \u03c9bq 5,0= - circulation ratio; yx, - displacement coordinates of the shaft center in the fixed coordinate system; 22 yxr += - radius of the shaft movement orbit" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003191_amr.216.539-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003191_amr.216.539-Figure6-1.png", + "caption": "Fig. 6 Home elevator prototype", + "texts": [ + "25 time of overload magnification, the best operating force of the unit driving motor with 16-pole 15-slot, namely the rated force, is 1000N. In Fig. 5, the cogging force of the unit motor is composed of the end component and slot component.The end component of the cogging torque is main with the period of one pole-pitch. The slot component is the harmonic content. Fig. 2 High-speed elevator In order to verify the feasibility of direct driving high-speed elevator, a small-sized home elevator prototype drove by PMLSM is built, shown in Fig. 6. Fig.6 (a) and (b) are the primary and secondary of the unit motor with 16-pole 15-slot, respectively. Fig.6 (c) is the cubic effect picture of the home elevator prototype drawn by Pro/Engineer software. Fig.6 (d) is the prototype of the home elevator prototype. Based on the experimental platform, the load experiment and power fail interrupts protection experiment were done. The experimental results are shown in the following figure from 7 to 10. As the experimental results show that the prototype operates safely and stably, and can be falling down at a constant low speed in the case of power failure. In this paper, a design scheme of direct driving high-speed elevator is proposed, and then a small-sized home elevator prototype drove by PMLSM is built" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003143_amc.2014.6823298-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003143_amc.2014.6823298-Figure2-1.png", + "caption": "Fig. 2. Spring-Link mechanism.", + "texts": [ + " This method allows to compensate the gravity torque of robot thanks to the counter weight that is shown in Fig. 1. However, it is difficult to improve acceleration of the system because the total weight and inertia become large. Spring-link mechanisms, which are used in some room lamp architecture and drafting instruments are also well known as examples of mechanical gravity compensation methods [7]. In these methods, springs are utilized to reduce gravity force of robots thanks to its tension stiffness. The kinematic model is shown in Fig. 2. However, in these mechanisms, accuracy of the compensation force depends on the posture of robots. Indeed, it is difficult to generate precise compensatory gravity torque in each posture by using only geometrical constraints. To solve these problems, Rahman proposed the wire-pulley mechanism [8]. This method uses springs, wire, and pulley to compensate the gravity force. However, this mechanism con tains a critical point on the pulley. These mechanisms deviates from theory due to spring displacement errors by difference in the length of the wire wrapped around the pulley, and the geometric error caused by changes in the wire contact position on the pulley" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001398_icphm.2015.7245064-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001398_icphm.2015.7245064-Figure4-1.png", + "caption": "Figure 4. Gearbox test rig and its internal structure", + "texts": [ + " For signals of different degradation states, these three parameters exhibit different shapes. In other words, different shape denotes a different value. Hence, these parameters can be used to classify the fault states and to track the degradation path. For \u03b1, in fixed X region, the shape of PDF varies with different fault. For \u03b3, both the region of PDF vary with different fault. For \u03b4, the location of the PDF varies with different fault. However, how these parameters can track the degradation need to be further verified. III. VERIFICATION USING LABORATORY Figure 4 shows the testing bed used run-to-failure data. It consists of a gearbox, a 4 kW three phase asynchronous motor for driving the gearbox, and a magnetic powder brake as the loading. The motor speed can be adjusted to different levels. The load can be adjusted varying the current of magnetic powder brake. The NI data X and the shape of well the trends of \u03b1 values (\u03b2=0, \u03b3=1, \u03b4=0) \u03b3 values (\u03b1=2, \u03b2=0, \u03b4=0) \u03b4 values (\u03b1=2, \u03b2=0, \u03b3=1) GEARBOX for collecting the - acquisition system consists of four acceleration transducers, PXI-1031 mainframe, PXIand LabVIEW software" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003708_gt2014-26673-Figure25-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003708_gt2014-26673-Figure25-1.png", + "caption": "Fig. 25 A finite element model of a bladed disc: a) a whole bladed disc; b) a bladed disc sector", + "texts": [ + "0 0.2 0.4 d /d d /df 1 d /dk d /dg 8 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/17/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use Analogous plots for a case when the cubic spring stiffness is varied from 10 8 to 10 6 N/mm are shown in Fig. 22, Fig. 23 and Fig. 24. One can see that sensitivity of LCO frequency to the spring stiffness is equal to 0 for all values of the flutter intensity and spring stiffness A bladed shown in Fig. 25 is chosen for a test case study of a realistic gas-turbine structure. The bladed disc is assumed to be tuned, which allows using its sector model even for a case of essentially nonlinear vibrations. The use of the sector model does not introduce any simplifications in the nonlinear analysis as it was shown in Ref.[24] provided that the travelling wave vibrations of a bladed disc are analysed. The sector FE model of the bladed disc used in the analysis is shown in Fig. 25. 9 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 08/17/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use The total number of DOFs in the sector model is about 54000. Friction damper are applied between every neighbouring blades at blade tips, and therefore, the damper slip-stick is defined by relative motion of adjacent blades. The friction contact model allows for not only friction forces but also the unilateral interaction along direction normal to the contact plane" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003725_2011-01-1512-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003725_2011-01-1512-Figure5-1.png", + "caption": "Figure 5. Vehicle Dynamics: Additional Signals in the Vehicle and Wheel Co-ordinate System", + "texts": [], + "surrounding_texts": [ + "2.2.1. Transfer rattling noise Front wheel driven based 4WD powertrain (Figure 6 and Figure 15) is composed of engine, transmission, transfer case (power take out), propeller shaft, rear axle and drive shafts. These plenty of inertia parts normally have their own resonance frequencies within driving ranges. These resonances are coming from excessive torsional vibration level of some parts and generating annoying noise. Detailed description of 4WD powertrain under test is shown at Table 3. Torque for rear wheels are electrically controlled by 4WD coupling with its own logic based on driving conditions. To investigate the torsional motions of each driveline components, torsional vibration behaviors of them need to be measured. First of all, angular displacement signals detected by magnetic pickup sensor (induced probe) have been recorded with LMS Test Lab [5] and its frontend system. These basic information signals are measured simultaneously at flywheel (engine), one of input gear of manual transmission, pinion gear of transfer case and pinion gear of rear axle module while targeted driving modes. The magnetic pickup sensors are positioned and fixed by making adequate holes at each position respectively. Figure 7 is showing the example of manual transmission and the sensor. The raw signals were recorded 20 kHz sampling frequency at each position. Angular accelerations of each part can be acquired by post process of differentiating the measured angular displacement signal and their unit also converted to rad/sec from rpm. Consequently, the torsional vibration excited by engine firing can be extracted by order tracking method from the angular accelerations. These sequences of data processing have been done both on the road with vehicle configuration and on the test bed with powertrain set up. Figure 8 is showing the torsional vibration profile while engine running up at 5th gear position from 1000 rpm through 3500 rpm with wide open throttle (WOT). Measured patterns of torsional vibration of the powertrain on test bed are very close to those of the vehicle on the road. For preventing rattling noise on the transfer case, transferred torque to the rear wheels is limited around 1800rpm and, as a result, the trosional vibration is lowered to the acceptable level. That improvement is successfully reproduced with powertrain configuration when comparing the data from vehicle on the road. In addition to that, Figure 9 is the comparing graph of torsional vibrations of the vehicle on the test bed driven by the human driver and those of the vehicle on the road at 6th gear full accelerating condition. From this investigation, close replay can be done successfully with both powertrain and vehicle configuration on the test bed. 2.2.2. Tip in/out shock of the powertrain Shock behavior of the vehicle body while tip in or tip out of the throttle pedal is highly taken into account to the vehicle comfort both NVH and drivability point of view. Backlash minimization, inertia optimization, engine torque sensitivity manipulation and mount system optimization are well known to be the main solution to this annoying phenomenon. Most efficient and effective action is checking this behavior in earlier development stage on the real road load powertrain test bed having no proto type vehicle and only powertrain configuration available. In earlier stage, all kind of actions mentioned above can be considered without restrictions. Figure 10 is the engine revolution changes of powertrain, which can be acquired by simple manipulation (low frequency re-sampling) of angular displacement signal as referred in 2.2.1, on the test bed and vehicle on the road when throttle pedal was tipped in and tipped out. Inertia of the wheels is properly simulated by the dynamometer and consequently, the amplitudes and frequencies of engine RPM changes due to sudden tip in and out are very close to the behavior of real vehicle. Using this experimental environment, inertia tuning, engine control parameter optimization or other works can be done very easily and in earlier stage of development. 2.2.3. Tip in shock of the vehicle As another example, tip in shock of the vehicle is reproduced on the test bed with vehicle configuration. The vehicle under test is described in detail at Table 4. Also the configurations of accelerometers and measurement setup are as shown at Table 5. The shock is appearing in the form of 15Hz vibration on the seat rail x-direction (longitudinal) at 3rd gear position as shown in Figure 11. For the purpose of investigating how closely reproduce the shock on the road at on the test bed, vibrations of the engine, engine mounting (both powertrain side and body side), lower arm of suspension and seat rail are measured as Figure 12, Figure 13 and Figure 14. The level differences of test bed data with road test data come from the absence of rigid body motion of vehicle and contribution of additional vibration passing through other transfer paths. But, their general patterns and characteristics at each point are representing real road tests sufficiently. For identifying the transfer path of shock vibration from the source to the seat rail, additional measurement was done and it was found that rear roll mounting is another main transfer path as shown in Figure 15. Of course, this kind of test, in which the suspension plays an important role, vehicle configuration is better than powertrain setup." + ] + }, + { + "image_filename": "designv11_84_0002924_06104.0139ecst-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002924_06104.0139ecst-Figure1-1.png", + "caption": "Figure 1. (a) Device configuration of recessed AlGaN/GaN biosensor. A Pt electrode is immersed in solution and act as a control gate. The source-drain current in the two-dimensional electron gas (2DEG) is highly sensitive to proximal charges. Positively charged biomolecules, which draw electrons out, increase Ids, while the negatively charged biomolecules act in the opposite way. (b) Id-Vg characteristics of recessed-gate, under-recessed and non-recessed device at VDS = 0.5 V.", + "texts": [ + " We have successfully detected monokine induced by interferon gamma (MIG), an indicator of transplant rejection in patients when detected at a concentration level above nano Molar, in clinical samples, and the electrical responses are consistent with the patients\u2019 clinical behavior. 140 ) unless CC License in place (see abstract).\u00a0 ecsdl.org/site/terms_use address. Redistribution subject to ECS terms of use (see 138.251.14.35Downloaded on 2015-03-23 to IP Device fabrication and threshold voltage control Figure 1(a) shows the device structure and testing configuration of a recessed AlGaN/GaN bioHFET. Ti/Al/Ti/Au multilayers for Ohmic contacts were deposited by e-beam evaporation, patterned by a lift-off process, and annealed at 850 \u00b0C for 30 seconds. The recess of AlGaN barrier was implemented with ICP etching by using spin-on-glass (SOG) as the etching mask. A two-step ICP etching process using BCl3 and Cl2/N2/O2 plasma with atomic or sub-nm resolution control was employed for threshold voltage control (15). Finally, a micro-reservoir was fabricated by patterning UV-sensitive silicone (Dow Corning WL-5150), with an active gate area of 2 \u00d7 2 mm2. In order to evaluate the sensitivity of recessed AlGaN/GaN bioHFETs, a non-recessed bioHFET and an under-recessed bioHFET were also fabricated for comparison. Figure 1(b) shows the Id-Vg characteristics of the three devices. The non-recessed device has a VT of -2.7 V, the under-recessed device has a VT of - 0.9 V and the recessed devi ce has a VT of 0.6 V. Surface functionalization The surface functionalization process for SA-biotin binding, including oxidation, silanization, and biotinylation has been described elsewhere (7,12). The main steps of surface treatment for clinical MIG detection include oxidation, silanization, and anti-MIG immobilization (8). 141 ) unless CC License in place (see abstract)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003236_csss.2011.5972150-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003236_csss.2011.5972150-Figure1-1.png", + "caption": "Figure 1. The general 5R serial robot", + "texts": [ + "3) Polynomials to remaining: RS3: Rem( 3/BS3)= 4) We can obtain the DTS: =+\u2212 =+\u2212+\u2212\u2212 =\u2212+\u2212+\u2212 02 0324416 0163284 2 13 1 3 1 5 1 7 12 2 1 4 1 6 1 8 1 xx xxxxx xxxx Here, we transform the original PS to the form of DTS, and can easily determine eight solutions to original PS because the leading terms of DTS haven\u2019t parametric variables. III. APPLICATION IN KINEMATIC ANALYSIS OF 5R ROBOT A. Inverse kinematic analysis of 5R robot The inverse kinematics problem of the general 5R robot has been discussed [7~8]. The configuration of 5R serial robot is shown in Fig. 1, where {xi, zi} (i=1-6) are all unit vectors, ai, li and twist angle i, i+1 are known structural parameters, p and (z6, x6) are the given position and orientation of end effectors respectively. The inverse kinematic problem of the general 5R robot can be reduced to determine rotary angles i (i=1-5). Step 1 Connecting the 1st and the 5th pairs by common perpendicular line (a7x7) of their joint axes z1 and z5, we can convert the general 5R robot to its corresponding 5R single loop mechanism, as shown in Fig", + " Eliminate y by substituting (2.b) into (2.a), we can obtain: 0)(3 =xp (3) Where p3(x) is sixth-degree polynomial. Therefore, we can transform (2) to the following form. = = = 0)( 0)( 0)( 2 1 3 xp xp xp (4) We can easily obtain x by solving the following equation. 0))(),(),(( 123 =xpxpxprem (5) It is noticed that the leading terms still have parametric variables after simplified (1) by Wu elimination algorithm, which will do harm to determine existing number of solutions. Example 1 As shown in Fig. 1, the position and orientation of the end effectors are given as follows. ]61176172.0,178713200,770590200[6 \u2212= . -.-z ]571495670,773388420,274341130[6 . -. -.\u2212=x ]06980892.0 ,09801240.1- ,34293401.2[=p Structural parameters are shown as below: 01 =l 56.02 =l 61.03 =l 32.04 =l 43.05 =l 76.01 =a 36.12 =a 81.03 =a 31.14 =a 43.15 =a 2.112 =\u03b1 5.023 =\u03b1 4.134 =\u03b1 445 \u03c0\u03b1 \u2212= 356 \u03c0\u03b1 \u2212= Where angular units are radian, other units are meter. With the help of the software (Matlab7.0), all the real solutions to (1) are obtained as below according to the decoupled leading terms elimination algorithm. \u2212== \u2212== 27177764.13)2tan( 30384776.0)2tan( 5 1 c c y x \u03b8 \u03b8 (6) Further, we can determine the rotary angles i (i=1-5) of the general 5R robot as shown in Tab. 1. The solution to (1) is non-unique when the 5R robot has special structural parameters, which is discussed as below. Example 2 As shown in Fig. 1, the position and orientation of the end effectors are given as follows. ]991071.0,085558.0,102262.0[6 \u2212= z ]1333330,6359590,7601170[6 . -. -.\u2212=x ]0 ,0 ,440175.1[\u2212=p Structural parameters are shown as below: 01 =l 2.02 =l 03 =l 04 =l 05 =l 11 =a 02 =a 5.13 =a 04 =a 05 =a 012 =\u03b1 223 \u03c0\u03b1 = 034 =\u03b1 245 \u03c0\u03b1 = 056 =\u03b1 Where angular units are radian, other units are meter. With the help of the software (Matlab7.0), all the real solutions to (1) are obtained as below according to the decoupled leading terms elimination algorithm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001540_chicc.2015.7259811-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001540_chicc.2015.7259811-Figure1-1.png", + "caption": "Fig. 1: sat\u03b51(sig(\u00b7)\u03b11) and V\u03b6", + "texts": [ + ", \u03b4 T n ], we get q\u0307 = v v\u0307 = \u2212rv \u2212 [ sat\u03b51 ( sig ( (I \u2297 L)(q + \u03b4) + (I \u2297B)(q + \u03b4 \u2212 qdl) )\u03b11 ) \u2212 sat\u03b52 ( sig ( (I \u2297 L)v + (I \u2297B)(v \u2212 vdl) )\u03b12 )] (36) Let \u03b6 = (I \u2297 (L + B))(q + \u03b4 \u2212 qdl) \u2208 Rmn\u00d71, \u03b7 = (I \u2297 (L+B))(v \u2212 vdl), we get \u03b6\u0307 = \u03b7 \u03b7\u0307 = \u2212r\u039e\u03b7 \u2212 \u039e ( sat\u03b51(sig(\u03b6) \u03b11) + sat\u03b52(sig(\u03b7) \u03b12) ) (37) where \u039e = (I\u2297 (L+B)) is positive definite. Define V\u03b6k as V\u03b6k = \u23a7\u23a8 \u23a9 1 1+\u03b11 |\u03b6k|1+\u03b11 |\u03b6k| \u2264 \u03b5 1/\u03b11 1 1 1+\u03b11 |\u03b6k| 1+\u03b11 \u03b11 + sgn(\u03b6k)\u03b51 ( \u03b6k \u2212 sgn(\u03b6k)\u03b5 1/\u03b11 1 ) |\u03b6k| > \u03b5 1/\u03b11 1 (38) where \u03b6k is the kth element of \u03b6. V\u03b6k is shown in Fig. 1. It is obvious that V\u03b6k is positive semi-definite, unbounded and is C1, and V\u0307\u03b6k = sat\u03b51(sig(\u03b6k) \u03b11)\u03b6\u0307k. Thus consider the following Lyapunov function V1 = mn\u2211 k=1 V\u03b6k + 1 2 \u03b7T\u039e\u22121\u03b7 (39) whose derivative along (37) is V\u03071 = \u2212r\u03b7T \u03b7 \u2212 \u03b7T sat\u03b52(sig(\u03b7) \u03b12) (40) From (40) it is easy to prove that \u03b7 is globally asymptoti- cally stable. Similarly it can also proved that \u03b6 is globally asymptotically stable. The saturation of (37) will not take effect when |\u03b6(j)| \u2264 \u03b5 1/\u03b11 1 , |\u03b7(j)| \u2264 \u03b5 1/\u03b12 2 , j = 1, ." + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.20-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.20-1.png", + "caption": "Fig. 2.20. Three-link planar arm", + "texts": [ + " With reference to the schematic representation of the kinematic chain, manipulators are usually illustrated in postures where the joint variables, defined according to the DH convention, are different from zero; such values might differ from the null references utilized for robot manipulator programming. Hence, it will be necessary to sum constant contributions (offsets) to the values of the joint variables measured by the robot sensory system, so as to match the references. Consider the three-link planar arm in Fig. 2.20, where the link frames have been illustrated. Since the revolute axes are all parallel, the simplest choice was made for all axes xi along the direction of the relative links (the direction of x0 is arbitrary) and all lying in the plane (x0, y0). In this way, all the parameters di are null and the angles between the axes xi directly provide the joint variables. The DH parameters are specified in Table 2.1. Since all joints are revolute, the homogeneous transformation matrix defined in (2.52) has the same structure for each joint, i", + "82) on the joint variables is not easy to express except for simple cases. In fact, in the most general case of a six-dimensional operational space (m = 6), the computation of the three components of the function \u03c6e(q) cannot be performed in closed form but goes through the computation of the elements of the rotation matrix, i.e., ne(q), se(q), ae(q). The equations that allow the determination of the Euler angles from the triplet of unit vectors ne, se, ae were given in Sect. 2.4. Example 2.5 Consider again the three-link planar arm in Fig. 2.20. The geometry of the structure suggests that the end-effector position is determined by the two coordinates px and py, while its orientation is determined by the angle \u03c6 formed by the end-effector with the axis x0. Expressing these operational variables as a function of the joint variables, the two position coordinates are given by the first two elements of the fourth column of the homogeneous transformation matrix (2.63), while the orientation angle is simply given by the sum of joint variables", + " On the other hand, in all those cases when there are no \u2014 or it is difficult to find \u2014 closed-form solutions, it might be appropriate to resort to numerical solution techniques; these clearly have the advantage of being applicable to any kinematic structure, but in general they do not allow computation of all admissible solutions. In the following chapter, it will be shown how suitable algorithms utilizing the manipulator Jacobian can be employed to solve the inverse kinematics problem. Consider the arm shown in Fig. 2.20 whose direct kinematics was given in (2.63). It is desired to find the joint variables \u03d11, \u03d12, \u03d13 corresponding to a given end-effector position and orientation. uniquely defined. As already pointed out, it is convenient to specify position and orientation in terms of a minimal number of parameters: the two coordinates px, py and the angle \u03c6 with axis x0, in this case. Hence, it is possible to refer to the direct kinematics equation in the form (2.83). A first algebraic solution technique is illustrated below", + " Consider the elementary rotations about coordinate axes given by infinitesimal angles. Show that the rotation resulting from any two elementary rotations does not depend on the order of rotations. [Hint : for an infinitesimal angle d\u03c6, approximate cos (d\u03c6) \u2248 1 and sin (d\u03c6) \u2248 d\u03c6 . . . ]. Further, define R(d\u03c6x, d\u03c6y, d\u03c6z) = Rx(d\u03c6x)Ry(d\u03c6y)Rz(d\u03c6z); show that R(d\u03c6x, d\u03c6y, d\u03c6z)R(d\u03c6\u2032 x, d\u03c6 \u2032 y, d\u03c6 \u2032 z) = R(d\u03c6x + d\u03c6\u2032 x, d\u03c6y + d\u03c6\u2032 y, d\u03c6z + d\u03c6\u2032 z). 2.17. Draw the workspace of the three-link planar arm in Fig. 2.20 with the data: a1 = 0.5 a2 = 0.3 a3 = 0.2 \u2212\u03c0/3 \u2264 q1 \u2264 \u03c0/3 \u2212 2\u03c0/3 \u2264 q2 \u2264 2\u03c0/3 \u2212 \u03c0/2 \u2264 q3 \u2264 \u03c0/2. 2.18. With reference to the inverse kinematics of the anthropomorphic arm in Sect. 2.12.4, discuss the number of solutions in the singular cases of s3 = 0 and pWx = pWy = 0. 2.19. Solve the inverse kinematics for the cylindrical arm in Fig. 2.35. 2.20. Solve the inverse kinematics for the SCARA manipulator in Fig. 2.36." + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001092_j.proeng.2015.07.168-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001092_j.proeng.2015.07.168-Figure3-1.png", + "caption": "Fig. 3 Relationship of each coordinate system on two-link model. Inertial coordinate system [a] constitutes static condition and local joint coordinate system [b] orients the location in keeping with the grip", + "texts": [ + " Defining displacement as u, v, and w for x, y, and z directions, these displacements are formulated by the following equations: [ ] dNwvu T ][==n (1) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )[ ]Tiziyixiiiiziyixiii zyxzyxd 111111 ++++++= \u03b8\u03b8\u03b8\u03b8\u03b8\u03b8 (2) [N] indicates shape function and d indicates node displacement (Eq. 1). In Eq.2, x, y, and z indicate the node displacement of each direction. \u03b8x is the angle of twist (rotation of x axis) and \u03b8y, \u03b8z show the slope of each axis. In Eq. 2, each index shows each axis and each node. 2.2. Motion equation for grip and shaft The motion equation for the grip and shaft is led by a two-link pendulum model (Fig. 3). On this pendulum model, the origin point of inertial coordinate system [a] is placed on the golfer\u2019s shoulder. We then define vector for the direction from the shoulder to the grip end as r and local joint coordinate system [b] which the origin point is the grip end. We also define vector for the direction from the grip end to the origin on the shaft coordinate system as \u03c1. Vector u, which shows the direction from the origin of the inertial coordinate system to the i-th node, is obtained by: The club face Grip Face Toe The i-th element The i+1-th node The i-th node z y xL Fig. 2 Coordinate system on the i-th element. The origin of this coordinate system is put on either end. The length between either end is indicated as L. The negative direction of the y axis is defined as the toe direction and the positive direction of the z axis is the face direction n\u03c1ru ++= (3) [ ] [ ]\u03c1b\u03c1ar == ,r\u0302 (4) r\u0302 indicates the translation component on the inertial coordinate system and \u03c1 is the non-time variable component. Then, the relationship of each coordinate system (Fig. 3) is obtained by: [ ] [ ]Sab = (5) S is the coordinate transform matrix. Using Eq. 5, the motion equation for the grip and shaft is formulated as follows using d\u2019Alembert\u2019s principle: 0=\u22c5un\u03b4 (6) [ ][ ] dN \u03b4\u03b4 bn = (7) \u03b4n is the virtual displacement of deformation and angle bracket is the mass integral. Expanding Eq. 6, Eq. 6 is deformed by: [ ] [ ] [ ] ( )rSNdNN TTT \u02c6~~~ ++\u2212= \u03c1\u03c9\u03c1\u03c9\u03c9 (8) \u03c9~ indicates the antisymmetric tensor of the angle rate. Considering potential energy, we can obtain the motion equation for the grip and shaft of each element" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002776_amm.332.396-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002776_amm.332.396-Figure3-1.png", + "caption": "Fig. 3. Desmopan Membrane", + "texts": [ + " In front of cameras, a calibration plate is moved manually, video cameras recording various board positions that will provide sufficient data to complete the calibration procedure. This successive images allows the software to record key points on calibration plate surface. Focal length and principal point are displayed on the screen and can be controlled by the operator. We are also displayed, the maximum errors. Regarding the determination of extrinsic parameters, calibration plate will be positioned in front of both cameras simultaneously and software (ISRA 4D) is able to compute all, with Q400 system. The studied membrane [8], Fig. 3., is part of the diaphragm pump and is made of desmopan, a thermoplastic polyurethane, so an anisotropic material for which is made in a very wide range of models and types, and can not know exactly longitudinal elasticity module or Young's modulus - E and Poisson's ratio - \u03bd, elements which with geometric characteristics are input into future studies of membrane by Finite Elements Method (FEM). The studied object, in occurrence a specimen band of desmopan membrane (Fig. 4.), must have sufficient space for digital image correlation algorithm to identify field with two cameras. In Fig. 3., we shown the working principle of digital image correlation system Q400. The principle of this deformable facets under the action of external forces. For desmopan specimen band we have followed these steps: - image acquisition deformations to 5 mm and 14 mm; - calibration with selecting intrinsic and extrinsic parameters obtained from calibration; - define assessment area; - set the image to be measured against the reference image; - evaluation (Fig. 5.); - interpretation of results; - Reading of \u03b51 (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003721_detc2014-34297-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003721_detc2014-34297-Figure8-1.png", + "caption": "FIGURE 8. Kinematic chains corresponding to cases from 13 to 18 of Table 1.", + "texts": [ + " Similarly, Figure 6(b) shows specifically the cases 2 and 5. Finally, Figure 6(c) shows specifically the cases 3 and 6. Figure 7 shows the cases from 7 to 12 of Table 1. In each Figure 7(a), 7(b) and 7(c), the first three kinematic pairs generate in a different way the subalgebra yu\u03021,p. Figure 7(a) shows specifically the cases 7 and 10; here the unit vectors u\u03022,1 and u\u03022,2, are parallel. Similarly, Figure 7(b) shows specifically the cases 8 and 11. Finally, Figure 7(c) shows specifically the cases 9 and 12. Figure 8 shows the cases from 13 to 18 of Table 1. In each Figure of 8(a), 8(b) and 8(c), the three first kinematic pairs generate in a different way the subspace S1,u\u03021 < xu\u03021 . Figure 8(a) shows specificaly the cases 13 and 16; here the unit vectors u\u03022,1 and u\u03022,2, are parallel. Similarly, Figure 8(b) shows specifically the cases 1 and 17. Finally, Figure 8(c) shows specifically the cases 15 and 18. Figures 6(a), 7(a) and 8(a), show different ways to generate the subspace S2,u\u03022 , when p4, p5 6= \u221e and u\u03022,1||u\u03022,2. Similarly, Figures 6(b), 7(b) and 8(b), show different ways to generate the subalgebra cu\u03022 , when u\u03022,1||u\u03022,2. Finally, Figures 6(c), 7(c) and 8(c), show different ways to generate the subspace S2,u\u03022 , in Figure 6(c), S2,u\u03022 < xu\u03022 , in Figure 7(c), S2,u\u03022 < gu\u03022 , while in Figure 8(c), S2,u\u03022 < yu\u03022,p. It should be mentioned that, both the Table 1 and Figures from 6 to 8, show only the kinematic chains where the first three kinematic pairs generate the subspace S1u\u03021 , while the last two kinematic pairs generate the subspace S2u\u03022 . However, besides the permissible permutations of the kinematic pairs that form the kinematic chains S1,u\u03021 and S2,u\u03022 proved in the propositions 4 and 5, the following proposition shows that many other permutations of the kinematic pairs are permissible" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002036_978-94-007-2069-5_11-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002036_978-94-007-2069-5_11-Figure3-1.png", + "caption": "Fig. 3 External forces and forces on bearings, gear meshes and within torsion-bar spring", + "texts": [ + " 2) must be satisfied condition for pitch diameters d32 d31 D d22 d12 d11 d21 : (1) Assuming the same module of all the gears, pitch diameters can be replaced by the number of teeth z32 z31 D z22 z12 z11 z21 : (2) The advantage of this type of gearing is high variability in dimensions and therefore large area for optimization of specific applications. For the prototype design and subsequent calculations was chosen case, when d11 D d12 D d31 D d32 D d1 and d21 D d22 D d4 D d2 (Fig. 2). Table 1 lists technical parameters of the gearbox, which are used for subsequent calculations. The gearing on Fig. 1a is a combined six-member mechanism (including frame). On Fig. 3 is release of each member. It should be determine 38 unknown forces and torques, torques Mtk (torsion-bar spring preload) a M2 (load on output shaft). are parameters of equation system. There are 30 equilibrium equations, the remaining eight equations resulting from the geometry of toothing (four meshes, for each one two equations). For the dynamic analysis was created model by Fig. 4. The gear unit is connected by a ball screw with a weight, which is constrained by a ball linear guide. The input of model is known desired kinematics of weight" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001539_chicc.2015.7260615-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001539_chicc.2015.7260615-Figure1-1.png", + "caption": "Fig. 1: The spiral path of Chirp-Z transform in the Z-plane", + "texts": [ + " The spectrum resolution of the N -point FFT is N2 by the definition of resolution. If the resolution of spectrum analysis result is increased by M times, the sampling intervals on the unit circle will accordingly reduce. The corresponding sampling points will increase to NM , thus making the computer time intensive to )ln(MNMN . However, the Chirp-Z transform is capable of representing the signal over a limited range defined by starts and ends frequencies. In addition, one can define a spiral path as shown in Fig. 1. Therefore, the spectrum resolution is improved significantly while the computer time intensive is merely increased a little more. As can be seen from the Fig. 1, the initial spiral radius is defined by 0A . By choosing the parameter 0W appropriately, contours can be selected which spiral either towards the origin or outside the unit circle. Mathematically the Chirp-Z transform can be written as 1 0 ),0()( N n nkn k WAnTxuzU (5) where 0 0 jeAA , 0 0 jeWW and the range of the integers n , k are 1,2,1,0 Mk and 1,2,1,0 Nn . Compared with conventional DFT algorithms, greater frequency resolution can be obtained without any zero padding. Meanwhile, zooming in on poles and zeros of the complex transfer function is possible by adjusting the spiral radius" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002711_amm.110-116.977-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002711_amm.110-116.977-Figure5-1.png", + "caption": "Figure 5 Scheme of asperity contact simulation model", + "texts": [ + " The contact occurs not only at the peak, but also at the oblique surface where the contact surface is not parallel to the mean plan any more. J. ABDO [8] has determined the conditions of contact for a general case, (7) where, z1 and z2 are respectively the heights of two asperities, r is the radial distance between the two vertical central lines, d is defined as the separation to the mean planes of asperities peak of the two surfaces. We can easily find that (8) Two asperities just touch each other when (9) In order to describe a general case of contact, we define 4 parameters illustrated in figure 5: the radius of peak \u03b2 and the semi-width of parabola b=L/2, which determine the shape of an asperity, the radial distance r and the depth of interference h, which determine the contact situation. We use three dimensionless parameters as the Eq. (10): the dimensionless radius of peak \u03b2 0, the dimensionless depth of interference h0, and the dimensionless radial distance r0. 1 2 1 2 1 2 2 ( /2) (no-contact) 2 ( /2) (touch) 2 ( /2) (interference) z z f r d z z f r d z z f r d + \u2212 < + \u2212 = + \u2212 > 2 ( / 2) 2 r f r \u03b2 = 1 2' 2 ( ) 2r z z d h\u03b2 \u03b2= + \u2212 = (10) The discussion of the thermal restriction resistance of such model is divided into two parts", + " Because of the complexity of the model\u2019s geometry, we could not find an analytic expression for the constriction resistance; therefore we performed a series of simulation analysis which are discussed in detail in the next section. The numerical simulation model is based mainly on the contact model described previously, but with a cylinder of radius b and length L which is six times bigger than b. Olsen et al [9] have shown that if the length of the cylinder is four times its radius, the cylinder could be approximated as being semi-infinite. Thus, L is long enough in this model. A diagram of three-dimensional model of the asperity contact is illustrated in the Fig.5. The numerical simulations were performed for different values of\u03b20, h0 and r0. Based on the our surface topography analysis data and also that from E. Ciulli [7], the\u03b20 is varied from 1 to 22, with a approximate increment of 3; h0 is varied from 0.0036 to 0.0545 with a approximate increment of 0.008; r0 is varied from 0 to r\u2019/b with 7 increments. While the values of \u03b2 and h are different, the value of r\u2019/b is different: (13) Steel with thermal conductivity of 16.27 W\u00b7m-1\u00b7K-1 is used as material to perform these analysis The boundary conditions used in the simulations model are as follows: The two three- dimensional contacting bodies have the same shape\u03b20 and thermal conductivity K=16" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure2.8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure2.8-1.png", + "caption": "Fig. 2.8. Critical end-effector motion: rotation around the axis of joint 9 drives the wrist towards a singular configuration producing high accelerations of joint 5", + "texts": [ + " In order to study the capability of the wrist inverse kinematics solution to avoid singular configurations the virtual walls constraining the angular workspace have been removed. These experiments clearly confirmed the benefit of redundant joints in haptic hardware design; it was possible to drive the end-effector to arbitrary orientations while effectively preventing singular configurations. Difficulties with the avoidance of wrist singularities have only been encountered when rotating the end-effector from a vertical orientation (xE \u2016zB) exactly around joint axis 9 into the horizontal orientation as illustrated in Fig. 2.8. In this case \u03b8\u03079 does not lie in the nullspace of J rot resulting in zero selfmotion. As a consequence, the wrist is driven into a singular configuration. In practice, however, the end-effector rotation is rarely exactly parallel to the axis of joint 9. Hence, the singularity avoidance is typically successful, but the selfmotion can induce undesirable high accelerations of joint 5 for fast end-effector motions. This can be avoided by placing a virtual wall keeping \u03b89 within the bounds [ 0\u25e6+\u03b6 180\u25e6\u2212\u03b6] with 0\u25e6 < \u03b6 < 45\u25e6, where \u03b6 is a measure for the distance of the 3R wrist from a singular configuration" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002823_amm.315.884-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002823_amm.315.884-Figure4-1.png", + "caption": "Fig. 4 The inner ring, outer rings and the roller", + "texts": [ + " 15) can contain except sub-structures also other elements, including nonlinearities (e.g. contact). The two-row roller bearing (see Fig. 3) is the subject of non-linear static analysis. The main results are the contact pressure on the rollers. The 8-nodes brick element type is used to create the standard FE model of the bearing [2]. Subsequently, the individual parts of the bearing are defined as substructures - the inner ring, two outer rings and 70 rollers (in two series, 35 rollers both, see Fig. 4). The contact pairs are defined as follows: each roller with both outer and inner ring - total 140 contact pairs. The special macrocommand must be written to define such a number of contact pairs. The macro-command contains the cycle of 35 loops. The nodes on the rings and the roller are selected in every loop. The contact (the roller) and target (the ring) elements are generated on selected nodes. The result of the nonlinear static analysis is the distribution of the contact pressure on the contact areas (see Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001216_icit.2015.7125200-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001216_icit.2015.7125200-Figure4-1.png", + "caption": "Fig. 4. Motor model for magnetic field analysis", + "texts": [ + " Therefore, the rotor with permanent magnets buried shallowly is used. B. Existence of the q-axis bridges Even though the permanent magnet buried shallowly, if the q-axis bridges between magnets are made as shown in Fig.3, the decrease of q-axis inductance caused by the magnetic saturation is relieved because q-axis flux flows enough as shown in Fig.3. The effect of making the q-axis bridges is shown by magnetic field analysis using FEM. TableI shows the specification of the motor model for magnetic field analysis. Fig.4 shows the motor model. When the U-phase and V-phase currents flow as shown in Fig.4, in case of the rotor position shown in Fig.4, the q-axis flux is maximum. Fig.5(a),(b),(c) show the results of the magnetic field analysis at 3A when the width of the q-axis brigdes tq is 3mm,5mm,7mm, respectively. Fig.5(a),(b),(c) show that the magnetic field saturation occur at the part [A] of Fig.4 at any tq because the permanent magnet buried shallowly. However, Fig.5(a),(b),(c) show that q-axis flux flows enough through the part [B] of Fig.4 with increasing tq . Therefore, the existence of the q-axis bridges cause that the q-axis flux flows enough as shown in Fig.3 even though the permanent magnet buried shallowly. The sound noise caused by torque ripple increases when the permanent magnet is buried shallowly because of the shortage of the flux due to permanent magnet \u03d5s which flows the teeth of the stator as shown in Fig.6 Therefore, the grooves of the rotor are made as shown in Fig.3. The depth of the grooves and the air gap are defined as L and g, respectively, as shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003871_amc.2014.6823303-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003871_amc.2014.6823303-Figure7-1.png", + "caption": "Fig. 7. Definition of Linearity.", + "texts": [ + " It is decided according to the leg length. The leg length from 90 to 120 cm was used in this study because the most of Japanese leg length is within this range [6]. We also use an assumption that the upper leg length and lower leg length is same. Figure 5 illustrates the definition of link name. The combination of a, b, c, d, e provides the characteristics of exoskeleton. In this study, the scale is resized 1/10 in cm, that is a=2 means that a=20cm. 7 types of link configuration is chosen as follows; III. DISCUSSION OF MODELS Figure 7 shows the definition of linearity in this study. The parameter \u03b8h is the angle from hip joint to operator\u2019s ankle, \u03b8s is the angle from hip joint to exoskeleton\u2019s ankle, \u03b8 is knee angle, respectively. The value of \u03b8h \u2212\u03b8s shows the linearity. Figure 8-14 shows 3-D plot of linearity. X-axis is operator\u2019s knee angle and Y-axis is operator\u2019s leg length. In red area, exoskeleton has small negative error to the ideal line from hip joint to ankle. In green area, exoskeleton has small positive error. If the red and green area is large, it means that the exoskeleton has good linearity" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.16-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.16-1.png", + "caption": "Fig. 2.16. Denavit\u2013Hartenberg kinematic parameters", + "texts": [ + " In order to compute the direct kinematics equation for an open-chain manipulator according to the recursive expression in (2.50), a systematic, general method is to be derived to define the relative position and orientation of two consecutive links; the problem is that of determining two frames attached to the two links and computing the coordinate transformations between them. In general, the frames can be arbitrarily chosen as long as they are attached to the link they are referred to. Nevertheless, it is convenient to set some rules also for the definition of the link frames. With reference to Fig. 2.16, let Axis i denote the axis of the joint connecting Link i\u2212 1 to Link i; the so-called Denavit\u2013Hartenberg convention (DH) is adopted to define link Frame i: \u2022 Choose axis zi along the axis of Joint i+ 1. \u2022 Locate the origin Oi at the intersection of axis zi with the common normal9 to axes zi\u22121 and zi. Also, locate Oi\u2032 at the intersection of the common normal with axis zi\u22121. \u2022 Choose axis xi along the common normal to axes zi\u22121 and zi with positive direction from Joint i to Joint i+ 1. \u2022 Choose axis yi so as to complete a right-handed frame" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002299_icssem.2011.6081289-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002299_icssem.2011.6081289-Figure6-1.png", + "caption": "Figure 6. Torque transferred by oil film", + "texts": [ + " It can be seen that the temperature at zone with groove is much lower than that at zone without groove. The highest temperature at zone without groove is 42.5 \u00b0C, while the lowest temperature at zone with groove is 30.5 \u00b0C. The temperature at zone with groove gradually reduces along the opposite rotation direction of the friction disk, while at zone without groove the temperature gradually rises from one side to the other. The torque transferred by the oil film is monitored in FLUENT software. The relationship between the torque and iterative times is shown in figure 6, in which the steady value of the torque is 0.335N\u00b7m under the relative speed of lOOOr/min, hence we can deduce the torque transferred by a friction pair is 6.7N\u00b7m. ? \" \ufffd 2 i:7 .... \ufffd 0.35:0 0.3lll 0.2&0 0.= 0.1= 0.1= 0.Cl'ID 3) 40 aJ eo 100 13) Iterative times The variation of the oil film temperature with time is shown in figure 8, in which (a) and (b) represent the temperature of the oil film at zone with groove and without groove respectively. It can be seen that the temperature of the oil film always rises during the measuring time" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003143_amc.2014.6823298-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003143_amc.2014.6823298-Figure1-1.png", + "caption": "Fig. 1. Counter weight mechanism.", + "texts": [ + " Moreover, even if the gravity force can be estimated by dummy robots, the maximum force will be limited because some of the rating is already used to compensate the gravity force. Therefore, it is necessary to compensate the gravity force using external force. Using external force, mechanical gravity compensation method has been also studied to reduce system weight actively. Counter-balance system, which is implemented in industrial robot arm and construction robot is one of the typical examples of the mechanical gravity compensation methods [6]. This method allows to compensate the gravity torque of robot thanks to the counter weight that is shown in Fig. 1. However, it is difficult to improve acceleration of the system because the total weight and inertia become large. Spring-link mechanisms, which are used in some room lamp architecture and drafting instruments are also well known as examples of mechanical gravity compensation methods [7]. In these methods, springs are utilized to reduce gravity force of robots thanks to its tension stiffness. The kinematic model is shown in Fig. 2. However, in these mechanisms, accuracy of the compensation force depends on the posture of robots" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.11-1.png", + "caption": "Fig. 7.11 Frames associated with the MDH parameters of the Gough-Stewart platform. a Frames associated to the base of the Gough-Stewart platform. b Frames associated to the leg i", + "texts": [ + " From a geometric point of view, solving these equations is equivalent to finding the intersection between a line Li defining the displacement of the active prismatic joints and a sphere Si that represents the displacement of the point Ai2 when the platform is fixed and the leg is virtually broken at point Ai2 (Fig. 7.9). The 6\u2013UPS PKM, also called the Gough-Stewart platform, is a robot composed of six legs, each leg being made of a passive U joint fixed on the base, followed by an active P joint and then a passive S joint (Fig. 7.10). The MDH parameters associated to the frames of Fig. 7.11 for one leg are given in Table7.5. For simplifying the computation, the base connecting points Ai1 are considered to all belong to the same plane (O, x0, y0). The parameters corresponding to the S joint are deliberately omitted. The computation of the S joint coordinates is of no interest in that section as they have no effect on the dynamic model if their corresponding friction are neglected (Khalil and Ibrahim 2007). We will deliberately limit the analysis of the IGM of the GoughStewart to the computation of the active joint coordinates only (that can be obtained through the use of the translational part of (7" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002570_ever.2013.6521586-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002570_ever.2013.6521586-Figure1-1.png", + "caption": "Fig. 1. Distribution of current density J at 3 broken rotor bars.", + "texts": [ + " In this article increased control system variables relative sensitivity effect was presented, which can be used in the diagnostic system synthesis. II. BROKEN ROTOR BARS FEM ANALYSIS Electromagnetic field analysis of an asymmetric rotor can be done with the common finite element method application. The study of simulation and experimental machine was done for the SG132-S4 induction motor with 28 bars and two pair of poles. Rotor fault is modeled as 3 broken adjacent rotor bars with symmetrical power supply. Figure 1 shows the current density distribution. There is no current flow in the broken rotor bars and the highest current density is in adjacent bars which causes a local increase of temperature. Presentation of this effect points the most common localization of the rotor bar fault and explains why the adjacent bars are usually damaged. Deformation of the electromagnetic field resulting from the asymmetry is visible in the distribution of the magnetic vector potential (Fig. 2). The penetration of the flux to the damaged bars is clearly visible" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001267_icmtma.2015.228-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001267_icmtma.2015.228-Figure2-1.png", + "caption": "Figure 2. Bogie of middle-low speed maglev train", + "texts": [ + " Refer to the above literature,and on the basis of the dynamical model of single bogie of middle-low speed maglev train established by Liu Yaozong[8], this paper sets up the disturbance input function of non-coplanar magnetic pole surfaces of the F-type tracks. It focuses on the study on the influence of four non-coplanar magnetic pole surfaces on the running of middle-low speed maglev train, and determines the upper limit of admissible value of disturbance amplitude of four non-coplanar magnetic pole surfaces. It provides basis for the detection and maintenance of F-type tracks. Bogie of middle-low speed maglev train is as shown in Figure 2.In accordance with the structure of the bogie and force analysis,the dynamical model of the bogie of middlelow speed maglev train applied in this paper is as follows[8]: 978-1-4673-7143-8/15 $31.00 \u00a9 2015 IEEE DOI 10.1109/ICMTMA.2015.228 9276 The forward direction of the bogie is taken as positive for X axis where vertical direction of its upper side is taken as Z axis and the horizontally perpendicular direction of the track is taken as Y axis. Thus a coordinate system forms and then the dynamical equation set of the single bogie is as follows[8]: \u03b1 \u03b1 \u03b1 \u03b1 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003999_s12283-014-0158-y-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003999_s12283-014-0158-y-Figure1-1.png", + "caption": "Fig. 1 Resulting cue ball locations after a ball-pot for different topspin and sidespin imparted on the cue ball [16]", + "texts": [ + " In this context, the work described in this paper is part of a research effort to develop a robotic manipulator for snooker [15]. Each of these three sets of systems described above has the need to analyze various available shots, for a given table state and select the best next shot available. The ability to impart different spins and velocities to the cue ball in combination with some exquisite collision dynamics present between balls results in a variety of ball trajectories giving the players a high level of flexibility (see Fig. 1). Nowadays, a vast number of online virtual snooker games are available, e.g., Snooker Skool from Yahoo! Games. Although most of the online games simulate snooker in 2D, 3-dimensional versions are also found [17]. In addition, the TV broadcasts make use of ball tracking systems such as HawkEye [18] that also simulate and predict ball behavior. Accurate ball collision simulations will make these virtual games and predictions more realistic. Traditionally, ball collisions have been analyzed without incorporating the effect of friction, and the object ball is supposed to move along the line connecting the ball centers at the instant of impact [19]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003259_gt2013-94105-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003259_gt2013-94105-Figure6-1.png", + "caption": "Figure 6: Tip timing and \u2018birdseye\u2019 view", + "texts": [ + " Figure 5 shows some orientations considered for static eddy current sensors for gear damage detection. These are as follows: A \u2013 Sensor located over gear tooth tips for tip timing and \u2018birdseye\u2019 view 4 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/asmep/76989/ on 07/08/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use B \u2013 Side view, various radial positions out from the gear hub C \u2013 Root inspection D \u2013 Cross-mesh sensor, for monitoring the mesh region (gear teeth under load) In Figure 6, sensors are positioned either radially (as shown) or axially offset, or a combination of both to determine irregularities in velocity transfer between mating gears. It should be expected that the velocity profile will not be constant. However, the sensitive measurement in timing can reveal changes from a baseline profile. \u2018Birdseye\u2019 view Sensors are positioned in the same configuration as for tip timing, as shown in Figure 6 (described above), to monitor the surface of the gear teeth tips as they pass the sensor. Using a high sensor drive frequency should allow surface defects to be detected as they modify the signal, compared to a baseline profile for a healthy gear. In Figure 7, both gear wheels are free to rotate, whilst the sensor (in yellow) is fixed, with its mid-point above the gear side face plane. The sensor can be positioned to monitor the mesh region (\u2018cross mesh\u2019 sensor), or may be traversed towards the shaft of either gear wheel to monitor the root or disc parts of the wheel" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002389_2014-36-0018-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002389_2014-36-0018-Figure2-1.png", + "caption": "Figure 2: Twin-Tube damper example.", + "texts": [ + " Damper tuning process it has been, historically, more a subjective matter than an analytical process. For this reason, the whole process of damper tuning, in some cases, it is strictly focused only on the damper force versus velocity adjusting (Figure 1). For the sake of simplicity it is a valid assumption. However, a lot of information is just not considered when one looks just for this damper curve. Page 2 of 10 A damper, itself, can be considered a very complex system. As can be seen in the Figure 2, a twin tube damper type, its function depends on the interaction of many parts: valves, orifices, hydraulic characteristics, etc [1,2]. The resulting system it is, then, more than a simple functional model that is represented just by force versus velocity function. The damper can show inherent characteristics of any second order mechanical system, such as, delay of response, lag, resonance frequency, etc. When such a second order system exhibits any non-linear behavior, it is obviously the impact over the whole system dynamic behavior" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003747_amr.744.262-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003747_amr.744.262-Figure3-1.png", + "caption": "Figure 3 Test Rig layout", + "texts": [], + "surrounding_texts": [ + "Three sets of test seals were made with newly developed manufacture process. The test seals are 12 inches in diameter, made of three different carbon grades that were down-selected from a dozen candidates having potential of enduring high temperature of air and superheated steam. A test rig was set up to test one pair of seals at a time. One of the seals is facing high-temperature steam, while the other one is exposed to hot air. A layout of the test rig is shown below. The test rig is balanced to reach a speed of 10,000 rpm. However, most tests were run under 5,000 rpm to meet the design requirements. Each seal consists of multiple segments. Interlocking joints are implemented to reduce inter-segment leakage. Figure 1 is one complete seal before it is assembled into seal housing. seals are performing equally well in terms of leakage. Below is one example test results on seal leakage. The newly developed carbon seals were shown to be very effective in sealing. The total leakage is equivalent to an annulus of about one thousandth of inch (or 0.000025m). It is about three times better than a brush seal, or ten times better than labyrinth seals. Although the three seals are shown equally effective, but their abilities to withstand high-temperature and steam environment are different. While Grade A and C are shown to have no visible degradation after the tests, Grade B was found with pitting and dusting in post-test inspection, as shown below." + ] + }, + { + "image_filename": "designv11_84_0003211_icfda.2014.6967391-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003211_icfda.2014.6967391-Figure1-1.png", + "caption": "Fig. 1. The block diagram of the CSMC for fractional order plant.", + "texts": [ + " The differential relation between state variables should be derived according to the system dynamics. Let 1 2x ( ) x ( )D t t\u03b1 = defines the dynamic relations of state variables of the system with fractional order derivative. Equation 14 should be rearranged for fractional order plant using the dynamic relations of state variables. Thus, (14) can be rewritten as, [ ] [ ] [ ] 1 11 11 12 1 1 0 1 0 21 21 22 1 1 1 0 1 ( ) x xd b a a x u t c c c c b a a D x K sign c c D \u03b1 \u03b1 \u2212 = \u2212 \u2212 (15) The block diagram of the controller is given in Fig. 1. Describing function analysis is a widely known technique to study frequency response of nonlinear systems. It is an extension of linear frequency response analysis. In nonlinear systems, when a specific class of input signal such as a sinusoidal is applied to a nonlinear element, one can represent the nonlinear element by a function that depend not only on frequency, but also on input amplitude. This function is referred to as a describing function. Describing function method has a wide area of applications from frequency response analysis to prediction of limit cycles" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002019_peac.2014.7038056-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002019_peac.2014.7038056-Figure2-1.png", + "caption": "Fig. 2. The ESM steady vector graph", + "texts": [ + " MODELING OF DTC FOR ESM Assuming a rotor flux reference frame, the equations of ESM can be expressed as: 0 0 0 0 0 0 0 0 0 0 00 0 0 0 00 s r sq r mq sdsd r sd s r md r md sqsq f ff rd rd rq rq R L L iu L R L L iu R iu R i R i \u03c9 \u03c9 \u03c9 \u03c9 \u03c9 \u2212 \u2212\u23a1 \u23a4 \u23a1 \u23a4\u23a1 \u23a4 \u23a2 \u23a5 \u23a2 \u23a5\u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5\u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5\u23a2 \u23a5 = \u23a2 \u23a5 \u23a2 \u23a5\u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5\u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5\u23a2 \u23a5\u23a3 \u23a6 \u23a3 \u23a6 \u23a3 \u23a6 0 0 0 0 0 0 0 0 0 0 0 0 sd md md sd sq mq sq md f md f md md rd rd mq rq rq L L L i L L i dL L L i dt L L L i L L i \u23a1 \u23a4 \u23a1 \u23a4 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5+ \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5\u23a3 \u23a6 \u23a3 \u23a6 (2) where usd, usq, uf - the stator d-axis, q-axis and rotor field voltage, isd, isq, if, ird, irq - the stator d-axis, q-axis, rotor field, and rotor damping d-axis, q-axis current, Rs, Rf, Rrd, Rrq - the stator winding, rotor field winding, and rotor damping cage d-axis, q-axis resistance, Lmd, Lmq - d-axis and q-axis magnetizing inductance, Lsd, Lsq, Lf, Lrd, Lrq - the stator d-axis, q-axis, rotor field winding, and rotor damping cage d-axis, q-axis inductance, \u03c9r - the electrical angular speed. The electromagnetic torque Te can be expressed as: ( ) ( )e p s s s s p sd sq sq sdT n i i n i i\u03b1 \u03b2 \u03b2 \u03b1= \u03a8 \u2212 \u03a8 = \u03a8 \u2212 \u03a8 (3) where np is the number of pole pairs, \u03a8s\u03b1, \u03a8s\u03b2, \u03a8sd, \u03a8sq are \u03b1axis, \u03b2-axis, d-axis and q-axis stator flux linkage. The steady vector graph of the ESM is shown in Fig. 2. In steady stage, we have \u03a8sd and \u03a8sq as follows: cos sin sd md f sd sd s sm sq sq sq s sm L i L i L i \u03b4 \u03b4 \u23a7\u03a8 = + = \u03a8\u23aa \u23a8 \u03a8 = = \u03a8\u23aa\u23a9 (4) From (3) and (4), we have another form of Te: 2sin sin2 2 md f s sd sq e p sm s sm sd sd sq L i L L T n L L L \u03b4 \u03b4 \u239b \u239e\u03a8 \u2212 = + \u03a8\u239c \u239f\u239c \u239f \u239d \u23a0 (5) On the right side of (5), the first part is the main electromagnetic torque of ESM, and the second part is the reluctance torque. Neglecting the stator leakage inductance Lls, the main electromagnetic torque is the cross product of the stator flux linkage vector and the rotor field flux linkage vector" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003641_peds.2013.6527115-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003641_peds.2013.6527115-Figure8-1.png", + "caption": "Figure. 8. V-shaped magnet structures.", + "texts": [ + " The increase in the number of layer is expected to cause a decrease in Ld and an increase in Tr The motor parameters and the torque characteristics of the flat structures are shown in Figs. 6 and 7, respectively. The broken line in Fig. 7 represents the torque of the S-F1 model. As the number of layers in the models with flat magnets increases, Ld decreases and Tr increases. As a result, the maximum torque of the flat magnet structures increases, and the torque of the P-F3 model is the highest among these structures (85% of that of the S-F1 model), as shown in Fig. 7. B. V-shaped Magnet Structures The rotor structures with V-shaped magnets are shown in Fig. 8. The P-V1, P-V2, and P-V3 models are single-layer, two-layer, and three-layer structures, respectively, with Vshaped magnets. The thickness of the magnets is 3 mm in the P-V1 model, 1.5 mm in the P-V2 model, and 1.0 mm in the P-V3 model. In this regard, the PM volume is the same in each layer. The width of the flux path between the PMs is 4.5 mm in the P-V2 model and 4.0 mm in the P-V3 model. The increase in the number of layers is expected to cause a decrease in Ld and an increase in Tr. As a result, the structures with the V-shaped magnets have a large Tm and cause Lq and Tr to increase" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.9-1.png", + "caption": "Fig. 7.9 The two working modes of the Orthoglide leg i", + "texts": [ + "28) where c11 = \u22122z c10 = (x \u2212 d6) 2 + y2 + z2 \u2212 d2 4 c21 = \u22122(x + a) c20 = (x + a)2 + (y \u2212 d6) 2 + (z \u2212 a)2 \u2212 d2 4 c31 = \u22122(y + a) c30 = (x \u2212 d6) 2 + (y + a)2 + (z \u2212 a)2 \u2212 d2 4 from which we can find: qi1 = \u2212ci1 \u00b1 \u221a c2i1 \u2212 4ci0 2 . (7.29) Finally, the passive variables can be found from (7.23) by: q12 = \u2212q15 = atan2 (q11 \u2212 z, x \u2212 d6) (7.30) q22 = \u2212q25 = atan2 (q21 \u2212 x \u2212 a, y \u2212 d6) (7.31) q32 = \u2212q35 = atan2 (q31 \u2212 y \u2212 a, x \u2212 d6) (7.32) q13 = \u2212q14 = atan2 (\u2212y, (x \u2212 d6)/ cos q12) (7.33) q23 = \u2212q24 = atan2 (\u2212z, (y \u2212 d6)/ cos q22) (7.34) q33 = \u2212q34 = atan2 (z, (x \u2212 d6)/ cos q32) (7.35) In (7.29), the sign \u201c\u00b1\u201d corresponds to the two differentworkingmodes of the robot (Fig. 7.9). From a geometric point of view, solving these equations is equivalent to finding the intersection between a line Li defining the displacement of the active prismatic joints and a sphere Si that represents the displacement of the point Ai2 when the platform is fixed and the leg is virtually broken at point Ai2 (Fig. 7.9). The 6\u2013UPS PKM, also called the Gough-Stewart platform, is a robot composed of six legs, each leg being made of a passive U joint fixed on the base, followed by an active P joint and then a passive S joint (Fig. 7.10). The MDH parameters associated to the frames of Fig. 7.11 for one leg are given in Table7.5. For simplifying the computation, the base connecting points Ai1 are considered to all belong to the same plane (O, x0, y0). The parameters corresponding to the S joint are deliberately omitted" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001468_1.4922926-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001468_1.4922926-Figure6-1.png", + "caption": "FIG. 6. Schematic representation of the motor under an external force (dashed arrow, \u2212F x\u0302) which acts on the MN bead (gray). The solid line is the center of the FB bead at x = 0 and indicates the position of a reflecting barrier for motor propulsion, and dotted lines are the center and the interaction boundary of the MR bead at x = LM and x = L, respectively. The solid arrow indicates the direction in which the motor moves.", + "texts": [ + " A simple stochastic model is able to capture many of the essential features of the motor dynamics observed in the simulations. While a similar model with only irreversible reactions was previously studied,27 in this section we consider a general stochastic model with irreversible reactions on the filament in the presence of an external load. In Appendix B, the generalization of the model to include reverse reactions is described. First, we consider an irreversible motor reaction with p\u22121 = 0, and a reversible reaction in the bulk characterized by various values of Q. As shown in Fig. 6, the motor diffuses along the x-axis starting at t = 0 with the MC bead at x = 0 and the MR bead at x = LM, where LM = 4 is the motor length given by the distance between the centers of the MC and MR beads. At time t = 0, a catalytic reaction occurs producing a product bead at the position of MC bead. The distance between the surface of MR bead at MNMR interface and the FB bead is L \u2248 3. We assume that the MR bead begins to interact with the FB bead when the motor moves a distance L to the left" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000806_0734242x20983895-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000806_0734242x20983895-Figure4-1.png", + "caption": "Figure 4. Sketch of the microbial fuel cells reactors used in the experiment.", + "texts": [ + " The single-chamber configuration is mostly selected when oxygen is used as an electron acceptor (Du et al., 2007). The MFC reactors were provided with electrodes: 40 cm pyrolyzed Arundo donax L. hollow stick was used as cathode and granular graphite (\u00d8 3\u20138 mm) was used as anode; they were placed inside a plastic net which was in turn disposed of within a high-density polyethylene slotted vertical pipe (\u00d8 75 mm), in the middle of the column. A felt layer separated the A. donax L. stick from the granular graphite ensuring the electrical insulation of the two electrodes (Figure 4). A carbon cloth rectangle (8 \u00d7 15 cm) was put inside the graphite mass and it was electrically connected to a plastic-insulated copper wire. The electrical connection was then insulated by four layers of a bi-component epoxy resin (UHU\u00ae Plus 5 minutes) and covered by pastafimo for ensuring rigidity to the connection. Electrical connection was tested for internal resistance and fluid contact/leakage by exposure to distilled water. The A. donax L. stick used as cathode is a biogenic material, with a hollow cylindrical shape and porous texture" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003762_s1068798x1410013x-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003762_s1068798x1410013x-Figure1-1.png", + "caption": "Fig. 1. Machining configuration.", + "texts": [ + " (1) and (2), which are implicit multiparametric functions of the number of teeth, their inclination, and the parameters of the initial generating contour. Two position hobbing and an elementary initial generating contour were assumed in [1]. Therefore, in the first hobbing position, the profile on one side of the teeth was formed. The necessary tooth thickness was ensured by adjustment of the tangential or angular dis placement of the initial generating contour and the blank in the second position, where the opposite side of the tooth was shaped. In Fig. 1, we show the machining configuration. Points a, d, l, and f of the tooth profile are specified in a polar coordinate system relative to the tooth\u2019s sym metry axis. In Fig. 2, we show the meshing of a pair of elementary initial generating contours, where the points of the initial generating contour are specified by the inclination to the initial straight line at the arcs corresponding to the head and base of the tooth. qH1,2 Nov , \u03b3H Nov , \u03bdH Nov , qF1,2 Nov , \u03b3F1,2 Nov , \u03bdF1,2 Nov Keywords: gears, Novikov gear, two position hobbing, load capacity DOI: 10", + " For assembly of the gear after hobbing in the second posi tion, the sum of the tangential displacements of the driving and driven gears must satisfy the condition (6) \u03b4s* \u03c0 4\u03c1f* \u03b1l.cos\u2013= \u03b4s* \u03b4s* y1* y2*.+= 610 RUSSIAN ENGINEERING RESEARCH Vol. 34 No. 10 2014 PETROVSKIY 4. EQUAL FLEXURAL STRENGTH OF THE TEETH IN THE DRIVING AND DRIVEN GEARS This conditions ensures that the target functions in Eq. (2) are equal for the driving and driven gears, so that (7) As the first approximation in solving Eq. (7), we use the condition of equal tooth thickness along the chords of the boundary circles (Fig. 1) (8) In the present case, the boundary circles are the initial circles of machine tool engagement, while the points l1,2 are defined by the following polar coordi nates (9) (10) (11) On the basis of Eqs. (9)\u2013(11), Eq. (8) reduces to an equation for the equivalent tangential displacements y1,2 (12) Taking account of Eqs. (5) and (6), we may solve Eq. (12). More precise solutions of Eq. (7) may be obtained on the basis of elasticity theory, by varying the tooth thickness ratio of the driving and driven gears, with redistribution of the coefficient between and For a pair of identical gears (z1 = z2) made of the same material by the same technology, Eq", + " The radii of the tooth tip circles for the specified radial gap coefficient \u03b4* at engagement may be expressed as follows (17) In the elementary initial generating contour, the center of the arc at the tooth base lies on the initial straight line. Therefore, the arc of the base is copied in the head of the tooth being shaped, while point a of the profile corresponds to the intersection of the tooth tip circumference and the arc of the head. The angular parameter of point a is obtained from the triangle OPa (Fig. 1) (18) (19) 6. MINIMUM GAP BETWEEN THE TRANSITION SURFACES The gap between the transition surfaces of the meshing teeth must prevent their contact in all operat ing conditions and compensate the technological errors. The minimum gap may be determined from the condition for the formation of a hydrodynamic oil layer between the transition surfaces. To assess the possible gap, we use Targ\u2019s approxi mate solution for the rolling of viscous liquid between rf1,2 z1 2, 2\u03c1f* 1 \u03b1lsin\u2013( )\u2013 z1 \u03b1kcos ;= \u03c7f1,2 \u03c0 z1 2, \u03d5P1,2;\u2013= \u03d5P1,2 2y0* y2 1,*\u2013( ) z1 2, ;= y0* 0", + " Hence, the pair of elementary initial generating contours corresponding to a Novikov gear with two engagement lines must have two point con tact and a gap 2\u0394e (Fig. 2). From the gear geometry, we obtain a formula for the angular parameter of point d in the initial generating contour 9. ACCOMMODATION OF THE CONTACT AREA With decrease in the number of teeth, the arc ad in the head of the tooth is curtailed by the transitional section of the initial generating contours but must remain sufficient to accommodate the contact area. It follows from Fig. 1 that, if the minor semiaxis b of the elliptical contact area is to be accommodated, we require that \u2013 \u2265 \u2013 \u2265 Formulas for the equivalent semiaxes a* and b* were presented in [1]. The coordinates of points d1,2 are determined from Fig. 1 as the point of intersection of circles containing the arc in the head of the tooth and the transitional involutes. \u03b1d = \u03b1l \u03c1f* \u03c1f* \u03c1a*\u2013( ) \u03b1k \u03b1l\u2013( )cos\u2013 2\u0394e*\u2013 \u03c1a* .arccos+ \u03b1k \u03b1d1,2 b*/\u03c1a*; \u03b1a1,2 \u03b1k b*/\u03c1a*. In polar coordinates, the equations of the circles for points d1,2 take the form (21) (22) (23) We know that point d belongs both to a circle and a transitional involute (24) (25) Hence, we may write the following formula on the basis of Eqs. (21)\u2013(25) A solution may be obtained by iteration with respect to \u03b1d1,2. For \u03b1a1,2, it follows from Fig. 1 that 10. EQUAL LOAD CAPACITY IN CONTACT AND FLEXURE This condition follows from the possible equality of the functions in Eqs. (1) and (2) which indicates that the flexural and contact stress may be redistributed on account of variation in the inclination \u03b2. rd1 2 2rd1 \u03c7d1 \u03d5P1\u2013( )cos \u03b1kcos \u2013 1 \u03b1k 2 cos + = \u03c1a* 2 z1 \u03b1kcos \u239d \u23a0 \u239b \u239e 2 ; rd2 2 2rd2 z2 \u03c7d2 \u03d5P2\u2013( )cos z1 \u03b1kcos \u2013 z2 2 z1 2 \u03b1k 2cos + = \u03c1a* 2 \u03b1kcos \u239d \u23a0 \u239b \u239e 2 ; \u03c7d1,2 inv\u03d11 2, inv\u03b1d1,2.\u2013= rd1 \u03b1lcos \u03b1kcos \u03b1d1cos ;= rd2 z2 \u03b1lcos z1 \u03b1kcos \u03b1d2cos " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003026_amm.401-403.254-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003026_amm.401-403.254-Figure1-1.png", + "caption": "Figure 1 The outer ring has a single point pitting corrosion condition", + "texts": [ + "199, Purdue University Libraries, West Lafayette, USA-29/05/15,05:17:32) Usually assumes that the outer ring and the rack is a rigid support. The inner ring is rigidly fixed on the rotating shaft. Balls pure rolling between the inner-outer raceway with the same distance distribution. We only discuss the bearing's vibration in the radial plane, without consider about the axial movement and the tilt of ferrule. According to the different circumstances of the radial clearance of the bearing, bearing radial load density distribution is different. As shown in Figure 1, normally the bearing has positive clearance. Assume the sensor mounted in the maximum density position of radial load, pitting corrosion in a position of loading area Q(\u03c8). In Figure1, \u03c8\u2014Angle of any location in load area and maximum load, or azimuth angle. In 1896, Hertz presented the classical solution about two elastic bodies' local stresses and deformations in a contact point. Hertz presented the following hypothesis in the analysis: All of the deformation in elastic range, no more than the material's proportional limit. Load vertical to the surface, neglecting the effect of surface tangential stress. Compared with the radius of curvature of load object, the size of contact area is small" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001550_chicc.2015.7259633-Figure13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001550_chicc.2015.7259633-Figure13-1.png", + "caption": "Fig. 13: Thrust state of pitching movement", + "texts": [], + "surrounding_texts": [ + "We simplify the six freedom model of submarine with VVP, so the vertical three freedom model can be shown in follows. State variable is given by: [ ]Tx z= (10) [ ]Tu w q=v (11) The thrust and moment of three degree of freedom are given by: T T x z yF F Q= (12) B E cos sin 0 R sin cos 0 0 0 1 = 11 22 55 0 0 M 0 0 0 0 m m m = 33 11 33 11 0 0 C(v) 0 0 0 = m v m u m v m u (13) Where B ER , M and C(v) denote space coordinate translation matrix, matrix and matrix, respectively. Damping matrix of three degree of freedom is given by: 11 22 55 0 0 D(v) 0 0 0 0 u v q = (14) The thrust and moment of three degree of freedom can be rewritten as the following state equation: T 1 0 0 1 0 x x z z y F F F F L Q = = (15) We suppose the gravity center and buoyant center are superposition and the gravity equal to the buoyancy, and the effect of ocean current is ignored, the g( ) and E are zero, so the g( ) and E can be ignored, and the vertical three freedom model of submarine with VVP can be shown in follows: = O ER v = + +-1 -1 Tv -M (D(v) C(v))v M If the dynamic characteristics of accelerated velocity and the effect to pitching angle velocity q are ignored, we can get the (16) and (17). 2 55 55 55 11 33 ' ( ) ( , ) q l z m d d m m uw T L n u w + + + = (16) sin cos sin( )z u w v= + = (17) In (16) and (17), 11 um m X= , 33 wm m Z= , y qI M , 1tan w u = , 2 2v u w= + , L is the distance from thrust action point to gravity center, ' ( , )n u w is interfere of non-modeling, 55m is the summation of moment of inertia circling Z axis and the additional moment of inertia produced by additional mass, 55 0m > ." + ] + }, + { + "image_filename": "designv11_84_0003482_s207510871501006x-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003482_s207510871501006x-Figure2-1.png", + "caption": "Fig. 2. The diagram of forces acting on point M in the polar MCS.", + "texts": [ + " 1 2015 MISSILE CONTROL IN THE POLAR COORDINATE SYSTEM 67 mands in the polar coordinate system, the \u2018normal\u2019 and \u2018tangential\u2019 ones. The formation of only one command provides the missile motion towards the beam axis O because its value is almost the same as the size of the total com mand K. The lack of command providing damping of the missile tangential motion under disturbances changes the nature of both the transition and entire guidance processes. To answer the question of what kind of changes will take place, consider another diagram (Fig. 2). Material point M experiences control force and gravity force Assume that m is the mass of point M; z, y are the Cartesian coordinates of point M; t is the time counted from t = 0. The initial conditions at t = 0 are the following: (1) Here, the dot over the symbol denotes differentia tion with respect to time; are the velocities along the axes z and y; and z0, y0, are the preset constants. acts in the direction opposite to the axis of ordinates and force , along the radius towards the center. Since is constant and depends on the distance from the origin of coordinates, it is convenient to use, as the generalized coordinates, polar coordinates \u03c1, \u03d5 related to the Cartesian coordinates by the known for mulas: (2) Let us find (3) K\u03c1 K\u03d5 K\u03c1 K\u03d5 F " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003259_gt2013-94105-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003259_gt2013-94105-Figure4-1.png", + "caption": "Figure 4: Schematic showing the principle of operation of an eddy current sensor", + "texts": [ + " An eddy current sensor uses either a permanent magnet to generate a constant magnetic field (passive sensor), or an electric coil, passing an oscillating current through the coil to generate an oscillating magnetic field (active sensor). When a conducting material passes the sensor, the magnetic field interacts with the material, and eddy currents are generated in the material. These in turn generate a magnetic field, which acts to oppose the primary field generated by the sensor, leading to a change in voltage in a sensing coil used to measure the effect of the eddy currents [11, 12]. Figure 4 illustrates the principle of operation of the eddy current sensor. The eddy currents are strongest near the surface of the material, and decrease in strength exponentially with increasing distance from the sensor. This is known as the skin effect, and can be used to measure proximity of the target material to the sensor. Eddy current sensors offer high accuracy measurements, and a particular advantage of this technology is its high robustness and immunity to contamination, making it suitable for use in harsh environments [11, 12]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001136_978-3-319-19788-3_1-Figure1.13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001136_978-3-319-19788-3_1-Figure1.13-1.png", + "caption": "Fig. 1.13 Examples of robots with 2 DOF able to position a point in a plane. a The Dextar: a planar five-bar mechanism (RRRRR planar architecture) designed at ETS Montr\u00e8al (Campos et al. 2010). b The ParaPlacer (PRRRP planar architecture) from the IFW (Hesselbach et al. 2002)", + "texts": [ + " It should be mentioned that the number of independent DOF of the platform of a PKM can be found by analyzing the rank of the parallel kinematic Jacobian matrix A defined in Sect. 7.3.1, when the robot is not in a singular configuration. Methods to compute the mobility of mechanisms are given in Appendix A. Many PKM have been designed in order to be able to move their platform in a plane. We call them the Planar Parallel Manipulators (PPM). We can classify them into three main groups: 1. robots with 2 DOF able to position a point in a plane (Fig. 1.13), 2. robots with 2 DOF able to position a device with constant orientation in a plane (two translational DOF in the plane and one constrained (constant) platform orientation around the axis normal to the plane\u2014Fig. 1.14), 3. robots with 3 DOF able to position a device in a plane (two translational DOF in the plane and one rotationalDOF around the axis normal to the plane\u2014Fig.1.15). There obviously exist other types of possible mobilities (1T1R), but they are not common. Most of the robots of this category are planar, i" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001567_icma.2015.7237783-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001567_icma.2015.7237783-Figure6-1.png", + "caption": "Fig. 6 Jaw mechanism", + "texts": [ + "5mm*13.5mm size to get visual information. Eyelid mechanism is showed in Fig. 4. The mechanism, including upper eyelid and lower eyelid, has 4 DOFs to open and close eyelids. The motors and eyelids also are connected by RSSR mechanism. As Fig. 5 shows, lid mechanism has 4 DOFs to change mouth shape. There are 2 towing points on the upper lid and 1 towing points on each corner of the mouth. RSSR mechanism is also adopted. Jaw mechanism has 3 DOFs. It is a three tier structure, which is showed in Fig. 6. The top tier is upper jaw, connected with the head skeleton. The bottom tier is lower jaw, connected with the robot's chin. Each adjacent layer is connected with linear guide mechanism. The top tier and middle tier could make relative slide forward and backward. The middle tier and bottom tier could make relative slide leftward and rightward. Robot\u2019s chin and upper jaw are connected by shaft. So the upper jaw and lower jaw could open and close. Combining all the movement above, jaw mechanism could make masticatory movement" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001429_9781118886397.ch14-Figure14.7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001429_9781118886397.ch14-Figure14.7-1.png", + "caption": "Figure 14.7 Vertical displacement during fault. (a) At rest and (b) during fault (Source: EPRI (2006)).", + "texts": [ + " As a simplify- ing approximation, it is assumed that the forces to which each conductor is subjected will cause an increase or reduction in sag, but that the conductor will retain a catenary shape. This assumption is supported by experimental results for low currents applied for long durations. The assumption is even more accurate for high fault current levels and short durations, where most of the kinetic energy is imparted to the conductor before the conductor can move appreciably. The terminology used in analyzing the vertical case is the same as for the horizontal case. The configuration used as a basis for calculations is illustrated in Figure 14.7. From the rest position, D0 = D1 = D2, that is, all sags are equal. For any other position, assuming both conductors are a catenary, the average separation distance can be expressed: davg = D0 + 2 3 (D2 \u2212 D1) (14.20) so that the average electromagnetic force is F = \ud835\udf070 2\ud835\udf0b I2\ud835\udcc1 davg (14.21) Note that the electromagnetic force has the effect of changing the effective con- ductor weight. The net accelerating force on each span of the bottom conductor is: Fnet = 2H2 sin \ud835\udf032 \u2212 S(W + F) (14.22) and for the top conductor, Fnet = 2H1 sin \ud835\udf031 \u2212 S(W \u2212 F) (14" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001639_acc.2015.7171009-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001639_acc.2015.7171009-Figure1-1.png", + "caption": "Fig. 1. Target LVLH frame (left), Chaser LVLH frame (right)", + "texts": [ + " The frame origin is the center of mass of one of the satellites and is moving with it. This frame is used to describe motions with respect to the moving position and direction toward the centre of Earth of an orbiting body. The X axis is in the direction of orbital velocity vector and it is called Vbar. The Y axis is in the opposite direction of the angular momentum vector of the orbit and is called Hbar. The Z axis is radial from the spacecraft centre of gravity (CoG) to the Earth centre of mass and is called Rbar (see Fig. 1). The geometry and inertial data for the orbiting modules are set considering as reference the work of [6]. The Hill\u2019s equations [5] are implemented to describe the relative motion of the two bodies in neighboring orbits x\u0308c = Fx mc + 2\u03c9z\u0307c y\u0308c = Fy mc \u2212 \u03c9yc z\u0308c = Fz mc \u2212 2\u03c9x\u0307c + 3\u03c92zc, (1) where x = [xc, yc, zc] T \u2208 R3 is the position vector, mc \u2208 R is the Chaser mass (known and varying with time), \u03c9 \u2208 R is the angular frequency of the circular Target (known and constant) and F = [Fx, Fy, Fz] T \u2208 R3 is the total force vector, which includes the forces due to the thrusters switching on and off, the effects of the errors in shoot and magnitude of the thrusters, and the forces due to the action of the external environment disturbances affecting the system" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002923_gt2013-95585-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002923_gt2013-95585-Figure2-1.png", + "caption": "Figure 2. AIR-RIDING LEAF TIP GEOMETRY", + "texts": [ + " Next the resulting tip force is combined with the linearised leaf seal model previously developed [9, 10] and used to predict the air-riding performance of a real seal. The key outputs of the prediction are air-riding gap height and hydrodynamic torque. The approach is validated by comparing the predictions to corresponding experimental data. This section presents an analytical and numerical investigation on how the leaf tip geometry alters during leaf seal operation and what hydrodynamic forces act on the leaves. For this investigation the geometry is assumed to be temporally frozen, allowing steady-state flow analysis. Figure 2 shows an idealised schematic of air-riding leaf tips in proximity to a rotor. The leaves shown in this sketch have experienced wear in the past, resulting in a leaf tip angle \u03b2wear. Depending on the difference between \u03b2wear and the leaf lay angle at the tip \u03b8, a converging or diverging gap with a cone angle \u03b1 is created between the leaf tip surface and the rotor. \u03b1 = \u03b8\u2212\u03b2wear (1) 2 Copyright \u00a9 2013 by ASME and Alstom Technology Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 02/08/2016 Terms of Use: http://www" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003762_s1068798x1410013x-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003762_s1068798x1410013x-Figure3-1.png", + "caption": "Fig. 3. Novikov gear system: O1, O2, axes of individual gears; \u03a01, \u03a02, initial planes of generating racks; \u03a03, plane of engagement lines; \u03a04\u2013\u03a06, front, middle, and rear end planes; \u03a07, plane of gear axes; k, contact points at engagement lines; P0\u2013P2, pole lines of working and machining engagements; lk, distance between engagement lines.", + "texts": [ + " MATCHING CONDITION: BASIC WILLIS ENGAGEMENT THEOREM The profiles transmitting rotation between parallel axes with specified angular velocity ratio at the con tact points have a common normal, which passes through the engagement pole [2]. Hence, if the gear ratio is to be constant in a Novikov gear with two engagement lines, the common normal to the contact points must lie in a plane passing through the engage ment lines. In that case, the projections of the com mon normals of the machining and working engage ments at the transverse plane will lie on a single straight line. From Fig. 2 and the form of the working engage ment (Fig. 3), we obtain an expression for the engage ment angle (3) where \u03b1k is the profile angle at the contact point of the pair of initial generating contours, while s* = \u2013 is the equivalent distance between the initial straight lines of the pair of initial generating contours and the splitting cylinders of the gears in the working engagement; and are the equivalent arc radii for the head and base of the tooth, respec tively. The equivalent interaxial distance may be written in the form and the equivalent distance between the engagement lines in the form (4) When s* = 0, the initial cylinders of the machining and working engagements coincide" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002354_carpi.2014.7030038-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002354_carpi.2014.7030038-Figure8-1.png", + "caption": "Figure 8. Sensors and actuators used in the Elevators 1Il and IV", + "texts": [ + " The safeness check routine avoids various situations that bring risks of damages to the Elevator or injuries to the operator. The most important routine is one that avoids the fall of the entire elevator together with the vehicle. This verification is done based on the elevation, the inclination and the rotation of the column. At moment, the inclination of the vehicle is not monitored. In future works, also this parameter will be measured using an electronic level. Thus a more precise safeness check against falling will be possible. Fig.8 shows the approximate position of sensors. Figs. 9 and 10 give more details about the sensors for detecting, respectively, the column inclination and rotation. The Table II describes the function of each sensor. Although details are not shown, the control system of the Elevator IV also avoids vehicle moving while the Elevator is at working position. The vehicle can be moved only if the column is set and locked in horizontal position. III. TESTS Several tests were conducted to ensure the structural safety and correct operation of the Elevator IV" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003814_iciea.2013.6566511-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003814_iciea.2013.6566511-Figure2-1.png", + "caption": "Fig. 2. 7-DOF Space manipulator system", + "texts": [ + " Thereafter, the corresponding joint angles and the maximum payload path are determined. The curves of the optimal solutions and the average population values are obtained using the MATLAB GA toolbox. And then the global optimal solution and the corresponding optimal parameters can be obtained. On this basis, the joint angle, joint angular velocity and joint acceleration expressed are solved, and the optimal path is obtained. IV. Units Simulation Results and Discussions The system is consisted of 7-DOF manipulator and spacecraft base, as illustrated in Fig. 2. The DH coordinates of the system is shown in Fig. 3, and DH parameters are list The corresponding polynomial parameters are as follows: Table . B. Simulation results During numerical simulation in point-to-point task, the corresponding polynomial parameters are as follows: The initial joint angle is set as: Taking the polynomial coefficient as controlling parameter and combing SOA dynamic equations, GA is used to optimize the joint space trajectory to achieve maximum load carrying capacity. The relevant parameters about load and GA are as follows: loadm : 400Kg" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002023_amr.314-316.1935-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002023_amr.314-316.1935-Figure1-1.png", + "caption": "Fig. 1 Schematic of the coaxial powder nozzle \uff08a\uff093D model \uff08b\uff09Grid and boundary condition Fig.2 3D model of coaxial powder nozzle Choice of calculation model. The volume percentage of particle phase is far less than 10% in the nozzle\u2019s flow field. Particles can be regarded as isolated from each other because of the great average space between them. The effects of fluid and particles on particles are ignored. A discrete model following Euler-Lagrange equation is adopted, and the gas is treated as a continuous phase, and the powder particles are treated as discrete phase [11].", + "texts": [ + " Taking into account the size and distribution of powder, the cone angles (\u03b1) of powder Advanced Materials Research Vols 314-316 (2011) pp 1935-1943 Online: 2011-08-16 \u00a9 (2011) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AMR.314-316.1935 All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 130.49.59.123, University of Pittsburgh, Pittsburgh, USA-04/04/15,00:27:02) cavity were 60\u00b0, 65\u00b0, 70\u00b0, 75\u00b0, and the gas cavity clearance (\u03b5) was 1.5 mm. Fig.1 is a diagram of coaxial powder nozzle. Because the coaxial powder feeding nozzle is axisymmetric structure, the calculations modeling was taken just one half. Fig.2(a) is the three-dimensional model of the nozzle. Fig.2 (b) shows the meshes and boundary conditions of model, and the boundary conditions which are not marked are wall boundary conditions. Gas-solid two phase flow model. In the discrete phase model, compressed gas is regarded as ideal gas, and the basic turbulence control equation includes continuity equation and momentum equation if the thermal effect on particles by laser is neglected" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001946_amm.698.552-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001946_amm.698.552-Figure6-1.png", + "caption": "Fig. 6. Mechanism (-90-80-70-100-120-80) and its directionvectors", + "texts": [ + " The modification of Brikard\u2019s linkage (\u03b11\u2022\u03b12\u2022\u03b13\u2022\u03b14\u2022\u03b15\u2022\u03b16) is assemblable, if direction vectors of the first and last hinges in the chain are collinear. The mechanism (-90-80-70-100-120-150) and the projection of its direction vectors is shown in Fig 5. Figure 5 shows that the condition of collinearity of the first and last direction vectors is not satisfied, so the mechanism cannot be assembled. Let us replace link FA (-150) by link FA (80) to provide the mechanism assemblability. The result is a new mechanism (-90-80-70-100-120-80) whose direction vectors are shown in Fig. 6. In this case, the condition of collinearity ofthe first and last direction vectors is satisfied, so the mechanism is assemblable. This assemblability condition of 6R mechanisms can be used for the synthesis of new mechanisms. Let us create a new mechanism choosing parameters of links \u03b11 \u03b12 \u03b13 \u03b14 \u03b15 \u03b16 randomly so that the mechanism satisfies the 6Rmechanism sassemblabilitycondition. In the mechanism shown in Fig. 7:\u03b11= -73\u00b0 \u03b12= -65\u00b0 \u03b13= 127\u00b0 \u03b14= -136\u00b0 \u03b15= 67\u00b0 \u03b16= 80\u00b0. Let us simulate the mechanism in the CAD system and ensure its assemblability", + " Brikard's linkage modification (\u03b11 \u03b12 \u03b13 \u03b14 \u03b15 \u03b16) is mobile if all links turn the direction vector in one direction (clockwise or counterclockwise), and each of the three link pairs rotates the direction vector through 180\u00b0. So, Brikard's linkage modification is mobile if: , where k=\u00b11 (1) Let us verify the satisfaction of condition (1) of the mechanisms discussed above. For Bricard\u2019s linkage (Fig. 1): \u03b11= -90\u00b0 \u03b12= -90\u00b0 \u03b13= -90\u00b0 \u03b14= -90\u00b0 \u03b15= -90\u00b0 \u03b16= -90\u00b0 So: , where k=\u00b11 (2) System (2) is correct for k=-1 so Bricard\u2019s linkage is mobile. For the mechanism shown in Fig. 6: \u03b11= -90\u00b0 \u03b12= -80\u00b0 \u03b13= -70\u00b0 \u03b14= -100\u00b0 \u03b15= -120\u00b0 \u03b16= -80\u00b0 So: , where k=\u00b11 (3) System (3) is not correct for k=\u00b11 so the mechanism is stationary. Let us change link parameters to provide mobility of the mechanism: \u03b11= -100\u00b0 \u03b12= -80\u00b0 \u03b13= - 70\u00b0 \u03b14= -110\u00b0 \u03b15= -120\u00b0 \u03b16= -60\u00b0 The assembly of mechanism links in a different order can not affect the assemblability, but it may affect the mobility. Changing the order of links of the mechanism shown in Fig. 9 to (-100-70- 60-80-110-120) will not affect its assemblability, but will lead to the loss of mobility as condition 1 will be violated" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001136_978-3-319-19788-3_1-Figure1.18-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001136_978-3-319-19788-3_1-Figure1.18-1.png", + "caption": "Fig. 1.18 Examples of spatial robots with 3 exotic DOF. a The Tricept. b The Exechon", + "texts": [ + " \u2022 robots with three rotational DOF (also called spherical PKM): most of them allow the platform to rotate around one given fixed point (Bonev and Gosselin 2006). The most known is probably the Agile Eye (Gosselin et al. 1996) (Fig. 1.17d), \u2022 robotswith three exoticDOF: such types of robots have usually someDOF of rotation which are constrained with the DOF of translation [(see e.g. (Bonev 2008)]. Some of them have been designed with an additional wrist which compensates for the undesirable rotations and have found some industrial applications, especially for milling (Fig. 1.18) \u2022 robots with three translational DOF and one rotational DOF around one given axis (also called Sch\u00f6nflies motion generators): they are usually used for pickand-place operations, most often at high-speed. The most functional robot of this type is probably the Adept Quattro (Fig. 1.4) \u2022 robots with six DOF: such as the Hexapod (also known as the Gough-Stewart platform\u2014Fig. 1.19a) and the Hexa (Pierrot et al. 1990) (Fig. 1.19b). 1.3.3 Redundant PKM Redundancy occurs when the number of active joints, na , is greater than the number ndof of independent variables required to define the platform configuration" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003489_amm.630.277-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003489_amm.630.277-Figure2-1.png", + "caption": "Fig. 2. Geometry of seal rings: a) distribution of the pressure on the face on the ring, b) forces acting on the flexibly mounted ring", + "texts": [ + "224, University of Michigan Library, Media Union Library, Ann Arbor, USA-11/07/15,16:33:30) During the seal operation, quick passages of the medium from the gland through the ducts into the cavities are observed. A single injection lasts for t1 = \u03b1/\u03c9, where \u03b1 is a circle sector equal to the length of a cavity. The flow is possible only when a duct is immediately over a cavity. After each injection, there is a sudden increase in the cavity pressure, p2, reaching a value equal to that of p1. Then, however, the pressure falls because of some leakage of the medium through the inner face, A3 (Fig. 2a). a) b) The drop in the medium pressure for a single cavity during the period between injections, so-called pressure impulses, depends on the height of the face gap. The greater the gap height is, the smaller the pressure drop and, accordingly, the smaller the average value of the pressure 2p during the interval between the injections of the medium, T=2\u03c0/i\u03c9 , where i is the number of the inlet ducts (7). The force generated in the clearance, Fs, responsible for separating the two rings being in face contact, is dependent on the pressure 2p . Thus, if the gap height increases, the force becomes smaller and does not balance the external force, Fe, independent of the gap height. The unbalance of the forces acting on the flexibly mounted ring, Fe - Fs >0, causes a decrease in the gap height followed by the forces balance, Fe = Fs. Thus, the negative feedback we observe (Fig.2b) assures the self-regulation of the gap height x. The average pressure in a cavity, 2p , and the opening force, Fs, depend on the cavity volume and the number of injections. A change in the value of the cavity volume an effect on the parameters of the seal (system). Hence, this seal is called semi-active or seal with controlled parameters. Therefore, the seal can be regarded as a self-control system (Fig.3), in which the gap height, x, is the controlled quantity, and the force Fs(x) is the controlling quantity, while the pressure of the sealed medium, p1, the pressure of the barrier medium, pe, the shaft rotational velocity, \u03c9, and the force responsible for contacting the rings, Fe, constitute external excitement quantities", + " ),,( 2 \u03c9\u2206 ppFkxxcxm s=++ , (1) pV QQQQ ++= 31 , (2) where: )( /ln62 1 21 3 3 1 2 pp rr xz Q c \u2212 \u2212= \u03b7 \u03b1 \u03c0 \u03b1 , )( /ln6 32 1 3 3 1 pp rr x Q c \u2212= \u03b7 \u03b1 , (3) xSQ nV = , 2p E V Qp = , (4) x \u2013 flexibly mounted ring axial displacement; m \u2013 mass of the seal ring; c, k \u2013 damping and stiffness coefficients of the flexible support; Fs \u2013 axial pressure force acting on the ring; \u03b1 , z, 1cr , 2cr , V \u2013 geometry of the cavities; Sn \u2013 face area of the ring; E, \u03b7 \u2013 stiffness modulus and viscosity of the sealed fluid; 2p \u2013 average pressure in a cavity; p1, p3 \u2013 pressures at seal outer and inner boundaries (see Fig. 2b); Q1, Q2 \u2013 internal and external flow-rates; QV, Qp \u2013 extrusion and compression flow rates. In the general case, the equation of balance of flow rate is non-linear. Its linearization was performed for a desired gap height corresponding to the condition of the static balance of the forces acting on the ring. From the equation of equilibrium (2) we calculate the average value of the pressure in the clearance as well as the force acting on the ring, which constitute the right side of Eq. (1). Finally, we obtain a linear equation of motion (1), where the quantities causing excitation are medium pressures at the clearance inlet and outlet and the rotational velocity of the shaft" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003087_amm.630.341-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003087_amm.630.341-Figure1-1.png", + "caption": "Fig. 1. Scheme of pump", + "texts": [ + "ntroduction High-speed pumps are used in aerospace industry, chemical industry, power engineering and other areas of technology. Such machines have small mass and size. Due to high angular velocity of the rotor these pumps are driven by gas turbines or high-speed electric motors without reduction gears. Therefore the whole pump unit may be relatively compact and energy-conserving. The cantilevered single-stage pump [1], in which end seal and annular seal have the function of radial-axial bearing, has been examined in the paper. The scheme of the single-stage pump is presented in Fig. 1. The impeller 4 is connected to the flexible rotor 13 through the spherical spline connection 14. The diameter of the rotor has been estimated from the condition of providing the rotor strength for the given torque moment. Front 1 and back 9 seals have the role of radial hydrostatic bearings. The automatic balancing system, consisted of the annular throttle 9, the face throttle 11 and the chamber 10, has the role of selfregulating radial-axial hydrostatic bearing. The radial vanes 7 of the casing 5 decelerate the circumferential flow in the back pocket 6" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003259_gt2013-94105-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003259_gt2013-94105-Figure2-1.png", + "caption": "Figure 2: Bevel gear fractured into three. All teeth have been lost from the parts labeled B and C, whereas teeth on section A have remained intact [7]", + "texts": [ + "org/about-asme/terms-of-use damage to a localised area of the gear tooth experiencing the greatest strain. Fatigue failure occurs in three stages; crack initiation, propagation and finally complete fracture. Gears developed for aerospace have a reduced cross-sectional area in order to save weight. Consequently, cracks are not limited to a tooth cracking off, as in the automotive industry, but propagate through the center of the gear causing it to break, leading to catastrophic failure of the gearbox [7, 8], see Figure 2. In addition to wear and fatigue, excessive forces can cause plastic deformation and/or catastrophic fracture of gearbox components, for example the shearing of gear teeth. Failure of gearbox components can result from a variety of factors. These include misalignment, bearing failure, vibration, unexpected loads or overload, poor lubrication, contamination with water or debris, shaft imbalance, adhesion/abrasion, temperature effects, and material or manufacturing defects. Misalignment is a frequent source of gearbox failure" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003599_1.c031306-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003599_1.c031306-Figure4-1.png", + "caption": "Fig. 4 Kinematics of rolling over a stone at the center of the tire footprint.", + "texts": [ + " For planes, x\u2013y is the ground plane, y\u2013z is the plane that bisects the stone and contains the sideward stone trajectory, and x\u2013z is the plane bisecting the stone and perpendicular to the axis of the tire. The initial velocity at which the tire surface approaches the stone can be calculated as a function of the geometry and speed of the tire and the geometry of the stone. Two distinct cases were considered: 1) The entire stone was within the tire footprint width. 2) The stone was partially inside the footprint width. A. Stone at the Center of the Footprint For complete overrolling of a stone, the vertical and resultant stone\u2013tire contact velocities are shown in Fig. 4. The vertical component of the initial tire\u2013stone contact velocity can be expressed as follows [1] using the derivation shown in the Appendix: vz V 2h=R h=R 2 p (1) The total velocity of initial contact between the tire and the stone can be found by q2 x2 h2 (2) v !q V R 2Rh p (3) v V 2h=R p (4) The angle relative to the vertical at which initial contact occurs is cos 1 vz=v (5) cos 1 1 h=2R p (6) For typical stone and tire parameters, h 20 mm andR 0:2 m, we obtain 13 . B. Stone at the Edge of the Footprint The initial contact speeds are slightly different when the stone is at the edge of the tire footprint (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002476_13-07195-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002476_13-07195-Figure2-1.png", + "caption": "FIG. 2. Schematic of experimental set-up (not to scale). (a) Overall view of the chip holder with attachments. The Clark microelectrode was originally placed over the microscope objective but was later moved. (b) Light path for the excitation light (solid lines) and emission light (dashed lines). Note that only one 3 3 3 array may be viewed at a time by the 103 microscope objective.", + "texts": [ + " Two luminophores were used for the custom sensors. One was platinum(II) meso-tetra(pentafluorophenyl)porphine (PtTFPP) (Pt975; Frontier Scientific, Logan, UT) and the other was a modified (in-house) version of that luminophore. Figure 1 shows the modified PtTFPP structure. The sensor material was picoinjected (PLI-100; Harvard Apparatus, Holliston, MA) via flexible capillary tips made in-house into microwells etched into fused silica glass chips prior to their placement into the chip holder; see Fig. 2. These glass chips (1 3 1 cm) were etched inhouse with 81 microwells ( 300 pL volume per microwell, arranged as a nine sets of 33 3 arrays) and served as the sample holder for the sensor material. The glass chips were centered over the quartz window, with measurements occurring on one 3 3 3 array per experimental trial. The commercial sensor material was also heat treated at 160 8C for 30 min in a tube furnace after picoinjection, whereas the in-house sensor materials were not heat treated. Experiments were conducted on an inverted epifluorescence microscope (Axiovert 200; Zeiss, Go\u0308ttingen, Germany), with prepared chips held in a customfabricated chip holder. All sensor characterization experiments were conducted in deionized water at 37 6 0.1 8C. Oxygen concentration was monitored via a Clark microelectrode (MI-730; Microelectrodes, Inc., Bedford, NH), and various gas concentrations were maintained via the sparging and purging tubes depicted in Fig. 2 (variable oxygen; 5% carbon dioxide [CO2]; the balance, nitrogen gas [N2]). The excitation source for pulsed experiments was a 405 nm light-emiting diode (LED) (208 LumiBright 1.8W; Innovations in Optics, Inc., Woburn, MA) that was collimated and filtered (FF01-405/10-25; Semrock, Rochester, NY). Figure 3 shows the output versus wavelength for this LED, without any filters in place. A secondary excitation source for some pulsed experiments was a 530 nm LED (LXHL-LM5C; Lumileds Lighting, LLC, San Jose, CA) that was collimated and filtered (FF01-543/22-25; Semrock, Rochester, NY)", + " The only other problem was a specular reflection from the Clark microelectrode. The microelectrode contained a polypropylene membrane that was quite shiny and reflected substantial portions of excitation light back through the epifluorescence optical train. That was not an issue for the ORLD method, but it was for the 320 Volume 68, Number 3, 2014 LPL method. The solution was to simply relocate the Clark microelectrode so it was out of the field of view of the microscope objective (as shown in Fig. 2); then the problem was eliminated. The original location of the Clark microelectrode was directly over the array under examination. With the RLEO signal diminished to a negligible level, self-illumination sources corrected, and the LPL method offering enhanced SNRs, the experimental plan was then modified. The LPL method was used for ratiometric and direct intensity data. However, there was some question as to the best pulse width to use for the LPL method. Some Monte Carlo simulations were undertaken in LabVIEW 2010" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001900_imece2013-62877-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001900_imece2013-62877-Figure3-1.png", + "caption": "FIGURE 3. CHORD ERROR BY SAMPLING AT A CORNER WITH ZERO VELOCITY AT THE VERTEX.", + "texts": [ + " The main result of this paper is in Section 4. Before the main result is established, two sections are arranged for comparison and easy understanding of the main result. Section 2 is on geometric error control of the motion passing through each vertex with zero velocity, and Section 3 is on geometric error control of the single-parabola blending motion but with the classical chord error model (2). Consider the corner turning from point A on ray \u2212e1 to point B on ray e2 with zero velocity at vertex o (the origin), as shown in Figure 3. The curve composed of the two edges can be approximated by a sequence of smooth curves whose curvatures at the vertex tend to infinity but elsewhere are almost zero. As a consequence, the chord error control (2) demands zero velocity at the vertex but elsewhere no velocity upper bound. Below we show that this conclusion is incorrect. Let the motion on the two edges be the following: first decelerate with maximal acceleration value a1 along e1 to zero velocity at vertex o, then accelerate with maximal acceleration value a2 along e2", + " Then each ri is rescaled by factor 1/\u03bb > 1, so that the right side of (12) is increased by factor 1/\u03bb to meet the demand |a|\u03c42/(2\u03b5) = r1 + r2 + 2 \u221a r1r2. (12) shows that the maximal acceleration allowed by the machining tool on each edge of a corner may not be reached. Bang-bang control is theoretically not allowed in chord error control. A single-parabola corner-turning is composed of three motions: a rectilinear motion on the incoming edge, a constantacceleration motion whose trajectory is a parabola blending the two edges, and another rectilinear motion on the outgoing edge. As shown in Figure 3, let the corner have incoming ray e1, outgoing ray e2, and vertex o (the origin in the plane). The corner turning starts at a point s1 on edge \u2212e1 and terminates at another point s2 on edge e2. Let a be the constant acceleration vector of the parabola motion. 4 Copyright c\u20dd 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use In 3D Cartesian motion, the plane spanned by vectors e1,e2 at o is called the corner-turning plane" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003317_tai.1960.6371668-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003317_tai.1960.6371668-Figure4-1.png", + "caption": "Fig. 4. Definition of the phase trajectory in co-ordinates.", + "texts": [ + " These conditions, determined from the phase plane by inspection, give rise to a method for determining the existence of a limit cycle for this type of system. For the second-order system a mathematical relationship can be determined and presented in curve form. A detailed derivation of the criterion is given in the Appendix; in outline form the procedure is to convert the equation of the phase trajectory into polar co-ordinates, obtaining \u2014 = = p = D*e \u03bd \u0393 \u0393 ^ 2 (7) and transferring the backlash dividing lines to the new phase plane as shown in Fig. 4. The system is unstable; i.e., the oscillations increase in amplitude, if in Fig. 4, Pi cos \\pi ^ P2 cos \u03c6\u03c4 (8) Combining these equations, the stability criterion becomes f cos \u03c8\u03b9\u03cd\\/\u0390-\u03b62\u03af Vl-f! \u03a7\u03c0-\u03c6\u03c7-\u03c8\u03c4) (9) but this is not useful because \u03c8\u03b9 is not precisely defined. However, near the origin \u03c8\u03b9 approaches \\pm&x and if the inequality exists for \u03c8\u03b9 = ^max then it is certain that a limit cycle exists. Substituting the proper functions for \u03c8\u03b9 : 1^\u2014e Vi-\u03b7\u039b 2fVi-f2 \"2\u0393 -ten\"lvfcf,) (10) X l ^ P l COS \u03c8\u03b9, X2 = C>2 COS \u03c8\u038e, fr\u00b1=;tan-2 f / V l - f 2 T\u00b1=;tangent to dividing line at origin", + " If equations 1 and 2 are combined and nondimensionalized forming a single equation with error, E, as the independent variable, this equation can be manipulated and integrated to define the linear phase trajectory on the phase plane as : \u03952+2\u03b6\u03c9\u03b7\u0395\u0395+\u03c9\u03b7 2\u03952 2\u0393 tan\"\u00bb \u0395+\u03b6\u03c9\u03b7\u0395 Using the linear transformation X*+Y* = D*eV\u00ef-\u00ef2 Letting Y=p sin \u03c8, X = p cos \u03c8, \u03c8 = t an\" 1 Y/X P = De V ^ 2 The backlash dividing lines are located on the X\u2014 Y plane by noting that Y \u0395+\u03ca\u03c9\u03b7\u0395 X \u03c9\u03b7\u0395\\/\u0390-\u03ca2 but the dividing line for separation is the E-axis, for which \u00a3 = 0 , and for this line = tan \u03c8\u03c4 x Vi-P The backlash dividing line for recombination is not readily determined, but the tangent to this curve Xmax on Fig. 3, is easily transferred : \u03a5_\u0395/\u03c9\u03b7 1 f x E Vi-f2 V w 2 and for Xmax E/E \u03c9\u03b7= \u2014 l /2f so Y 2 t 2 - l -. = tan ^ m a x x 2rVi-r2 which may be expressed as an acute angle l - 2 f 2 tan ^max = 2 f V l - f 2 In Fig. 4 the backlash line for recombination is sketched, not calculated, but the basic condition for increasing oscillations is Pl COS \u03c6\\ = P2 COS \u03c8\u03c4 Assuming tha t the angles are measured from the negative real axis in Fig. 4, and tha t pi and Zi are known : W l D = P2 = \u03c62 = -Pie^-t* r*2 --De ^ i - f 2 = = 7\u0393\u2014 \u03c8\u03c4 Substituting, cos <- ^ i ^ c o s \u03c8\u03c4\u03b2 1 - f ( e ?(**- p r \u03bd \u0393 \u25a0\u03be(\u03bd-\u03c6\u03a4- v\u00ef=f* \u03c0 \u2014 \u03c8\u03c4 \u2014 -\u03a8\u03cd ^ 2 -\u03a6\u03cd \u03a8\u03b9) V W Vi- The angle \u03c8\u03b9 may be chosen arbitrarily, and by inspection of Fig. 4 it is apparent tha t if the required inequality exists it must be most pronounced when the point (pu \u03a8\u03b9) is located near the origin on the tangent line, in which case ^i = ^max. Substituting this value for \u03c8\u03b9 : l \u00e2 - e Vf =(\u00b7 -f2\\ A _, i -2r2 IT \u2014 tan * -\"\u2014 2 r ^ i - r 2 - tan\" 1 1 - ^ 2 / Vi-\u03c8' References 1. S T A B I L I T Y C R I T E R I A FOR I N S T R U M E N T S E R V O - MECHANISMS WITH C O U L O M B F R I C T I O N A N D S T I C - TION, M . P . Pas te l , G. J. Thaler. AIEE Transactions, pt " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001377_acc.2015.7172274-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001377_acc.2015.7172274-Figure7-1.png", + "caption": "Fig. 7. Loop-Shaping Iteration 0: CRCBode diagram corresponding to LQG based controller, K0, Eq. (19).", + "texts": [ + " Due to the zeros at the origin, we must have W1(0) < 1, and since we have both CRHP zeros and poles we expect sensitivity peaking near the unstable pole frequencies, s = 1 [rad/s]. Therefore, we allow the sensitivity weighting function to dip here, but control the depth to prevent unwanted oscillations. The final uncertainty weighting function satisfying these requirements is given in Eq. (17). W1(s) = 0.6(s+ 1.2)2 (s\u2212 4.8)(s+ 0.3) (17) W3(s) = 0.01(s+ 1)(s+ 10) s (18) B. Inverted Pendulum: CRCBode Loop-Shaping We initiate the CRCBode design process with the LQG controller obtained in the previous sections. The CRCBode diagram corresponding to K0 = KLQG is presented in Fig. 7. As we can see in Fig. 7, there are intersections with the forbidden regions in the mid-frequency region near 2 [rad/s]. K0 = KLQG = 96.81(s+ 1) (s+ 0.2575)(s\u2212 5.519) (19) Typically in the robust loop-shaping process we cascade low-order compensators, e.g. integrators, lead/lag, and so on, in an attempt to avoid the forbidden regions. Indeed, one of the primary advantages of using the CRCBode plots vs. automated synthesis routines is that the structure and order of the controller can be directly specified. However, in this case the LQG controller is second-order and is sufficient to satisfy the robustness objectives, so we proceed by tuning the parameters without adding additional dynamics", + " Thus we have demonstrated that the CRCBode robust loopshaping procedure has successfully improved upon the linearquadratic Gaussian design, making the system significantly more robust to higher levels of parametric uncertainty. K3 = 115(s+ 1) (s+ 0.2)(s\u2212 5.5) (20) C. Inverted Pendulum: QBode Loop-Shaping When designing the controller directly using the CRCBode approach as in the previous section, even seemingly minor changes to the controller may inadvertently destabilize the system by causing the frequency-response to \u201cjump\u201d over a small forbidden region, such as that seen in the center of the magnitude plot in Fig. 7, in a single step. For this reason, care must be taken to only make incremental changes and check internal stability with each compensator iteration. It is worthwhile in these difficult problems to pursue the Youla based design approach (QBode) presented in this section, since internal stability is guaranteed for stable Q(s) at the expense of possibly high-order controllers. The nominal LQG controller corresponds to setting Q(s) = 0 in the Youla parametrization, Eq. (7). We begin by investigating the QBode diagram corresponding to this choice, shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002823_amm.315.884-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002823_amm.315.884-Figure3-1.png", + "caption": "Fig. 3 The bearing", + "texts": [ + " (13) The first group of equations can be written as ( ) s 1 ssmsmmsm 1 ssmsmm fKKfuKKKK \u22c5\u22c5\u2212=\u22c5\u22c5\u22c5\u2212 \u2212\u2212 (14) or simply: fuK m ~~ =\u22c5 , (15) where sssmsm smssmsmm fKKff KKKKK \u22c5\u22c5\u2212= \u22c5\u22c5\u2212= \u2212 \u2212 1 1 ~ ~ (16) are the reduced stiffness matrix and reduced loading vector. After the evaluation of master DOF (Eq. 15) the slave DOF can be calculated (so called expansion) from Eq. 13. Typical example of sub-structuring can be seen in Fig. 2. The reduced matrix solution of Eq. 16 requires the sub-structures to be internally linear. However the reduced task (Eq. 15) can contain except sub-structures also other elements, including nonlinearities (e.g. contact). The two-row roller bearing (see Fig. 3) is the subject of non-linear static analysis. The main results are the contact pressure on the rollers. The 8-nodes brick element type is used to create the standard FE model of the bearing [2]. Subsequently, the individual parts of the bearing are defined as substructures - the inner ring, two outer rings and 70 rollers (in two series, 35 rollers both, see Fig. 4). The contact pairs are defined as follows: each roller with both outer and inner ring - total 140 contact pairs. The special macrocommand must be written to define such a number of contact pairs" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003641_peds.2013.6527115-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003641_peds.2013.6527115-Figure3-1.png", + "caption": "Figure. 3. Magnetic flux density distribution.", + "texts": [ + " The characteristics of these models are examined using the 2- D finite element method. For the same rotor structure, differences in the characteristics of IPMSMs with sintered magnets and IPMSMs with powder magnets are evaluated. The motor parameters and the magnetic flux density distribution of each model are shown in Figs. 2 and 3. When sintered magnets are replaced to powder magnets, Ld, Lq, and Lq - Ld increase because the magnet fluxes decrease in the rotor and the magnetic saturation in the rotor is reduced, as shown in Fig. 3. The maximum torque of each model is shown in Fig. 4. Here, Tm is the magnet torque, and Tr is the reluctance torque. The maximum torque of the P-F1 model is 75% of that of the S-F1 model. However, Tr of the P-F1 model is higher than that of the S-F1 model because Lq of the P-F1 model is large, and Lq - Ld is larger than that of the S-F1 model. IV. ROTOR STRUCTURES OF IPMSMS USING POWDER MAGNETS FOR HIGH TORQUE Replacing sintered magnets with powder magnets decreases the torque. Therefore, in this section, the influence of the rotor structure using powder magnets on the torque is examined under a constant PM volume using a six-slot stator, as shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002979_s11249-015-0466-9-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002979_s11249-015-0466-9-Figure1-1.png", + "caption": "Fig. 1 Drawing of heat pipe stationary disk (upper) and conventional rotating disk (lower) (units in mm)", + "texts": [ + " Aside from mechanical seals, this technology can be potentially applied to other devices, such in parallel surface thrust bearings. However, frictional characteristics require further investigation. In this paper, a new design and testing procedure is reported to investigate the effect of heat pipe on the tribological performance of parallel surfaces made of stainless steel. 3.1 Stationary Disk with Built-In Heat Pipe The basic design of the heat pipe disk developed for this investigation is very close to that described in [18]. Figure 1 shows the design of the heat pipe stationary disk and the conventional rotating disk. Two channels are cut through from rear end of the heat pipe disk to 1 mm away from its front end. The two blocks between channels are for thermocouples. The width of each channel is almost the same size as the rubbing face to ensure that most of the heat generation is absorbed by the working liquid (water is used here). The disk is made of stainless steel (17-4 PH) and is heat-treated to a hardness of 45 Rockwell C" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003725_2011-01-1512-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003725_2011-01-1512-Figure3-1.png", + "caption": "Figure 3. Vehicle Dynamics Co-Ordinate System", + "texts": [ + " Longitudinal forces are the traction forces coming from engine or brake, aerodynamic drag, rolling resistance and gradient resistance and can be expressed with equation (1). This model is called a single wheel vehicle model and it has only one degree of freedom (1DoF), the longitudinal motion (1) Complex powertrains with interactions between various control units including vehicle stability require a more detailed vehicle model. This must be capable of describing the vehicle's motion in space. The full vehicle simulation is well known as vehicle dynamics model as shown in Figure 3 and Table 2. The full vehicle model describes 6 degrees of freedom (6DoF) and balances forces for linear movements and torques for rotational movements. To ensure the real-time capability of the simulation a simplified multi-body system (MBS) was developed. Simplifications to the vehicle model include: \u2022 The entire vehicle is considered as a rigid body. \u2022 Longitudinal, transverse and yawing motion is taken into account, i.e. the rigid body vehicle moves on one level parallel to the roadway. \u2022 Pitching is calculated by a simplified model (low pass filter)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003796_ijcsm.2013.054683-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003796_ijcsm.2013.054683-Figure2-1.png", + "caption": "Figure 2 The biped walking process of humanoid robot", + "texts": [ + " \u03c31, \u03c32, \u03c33, \u03c34 \u2208 (0, 1) are the weight coefficients, and \u03c31 + \u03c32 + \u03c33 + \u03c34 = 1. When H~R ( ) 1,iS t \u2192 the joint motion trends to be static. When H~R ( ) 0,iS t \u2192 the joint motion trends to be fast transforming. The captured track from human actor cannot be applied on the humanoid robot directly for the limitations joint velocities, link length, size, etc. so we have to analyse the biped walking of humanoid robot and modify the corresponding parameters. The biped walking process of humanoid robot is shown in Figure 2, among which the phase A shows the single-leg standing, B shows the swinging leg touch the floor around the heed and C shows the feet stand on the floor. Assume that \u03c6i is the intersection angle between the ith and (i + 1)th (i = 0, \u2026, 6) links, ri(i = 0, \u2026, 6) is the length of the ith link, qj(j = 1, \u2026, 5) is the intersection angle between the vertical line and the jth link. We set the following constraint of kinematics and ZMP for similar biped walking: \u2022 Phase A (standing with upholding leg while another leg swings): 2 3 4 0 2 1 0,z z z z z z z> > > > > ( ) ( ) ( ) ( ) S S_x S_z A V A TU 1A A A A A U U_x U_z1 1 1 ( ) ( ), ( ) ( ) 0 C {2, 4, 6} ( ) ( ), ( ) ( ) 0 i i i i i i i i t t t t k k i t t t t k k+ + + + \u23a7 \u2202 = = \u23a1 \u23a4\u23aa \u23a3 \u23a6\u2202\u23aa\u2261 \u2208\u23a8 \u2202\u23aa = = \u23a1 \u23a4\u23a3 \u23a6\u23aa \u2202\u23a9 Q q v q q q q Q q v q q q q ( ) { } ( ) { } A R A A c hZMP ZMP ZMPA ZMP A R A A c hZMP ZMP ZMP ( ), ( ), ( ) (u _ foot) C ( ), ( ), ( ) (u _ foot) x x y y \u03c6 t \u03c6 t \u03c6 t A \u03c6 t \u03c6 t \u03c6 t A \u2212 \u2212 \u23a7 \u22c5 \u2208 \u22c5\u23aa\u2261 \u23a8 \u22c5 \u2208 \u22c5\u23aa\u23a9 Q e e Q e e among which, A VC is the velocity constraints on both legs and A ZMPC is the ZMP constraints" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002872_amm.372.486-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002872_amm.372.486-Figure1-1.png", + "caption": "Figure 1: Experimental setup and work piece schematic diagram", + "texts": [], + "surrounding_texts": [ + "Laser head with focus distance and beam diameter of 160mm and 0.480mm was used to convey the laser beam on to the work piece material. The materials with 1.8mm of thickness were cut into 25mm x 5 mm and welded on a clamping device. In the early study, the scanning parameters in Table 2 were determined based to the laser oscillator capacity shown in Table 1. The scanning speed and pulse repetition rate were determined with conditions that the overlap area must be more than 50%. The result of the early study has led to the new set of parameters. These parameters were used to focally analyze the influence of certain processing variables within a smaller range value to obtain the optimized processing parameters. Preliminary study has been done to clarify the range of parameters applicable for welding process of titanium alloy with thickness 1.8mm. Pulse repetition rate fp were set lower than middle range (100Hz and less), to provide average laser power Pavg and pulse width tp with larger modification range. To obtain deep penetration welding beads, the influence of laser beam focus position has been studied. It was found that the penetration at 1.5mm underneath the work piece surface shows better result compared to the beam focusing on the top surface. Further experiments were conducted to analyze the effect of processing parameters on the defect of free deep penetration welding as shown in Table 3. Laser scanning was performed with focus point of 1.5mm under the material top. Argon gas with flow rate 7l/min was delivered using 6.5mm inner diameter nozzle with approximately 45\u00b0 tilted and the distance of 10mm from the scanning point." + ] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure3-1.png", + "caption": "Figure 3 CONICAL SEGMENTS FOR HYPOID GEARS", + "texts": [ + " Gear manufacturing literature is extensive and the majority Preliminary investigations into the \u201cideal\u201d kinematic geometry of spatial gearing have been recognized by Xiao and Yang [10], Figliolini and Angeles [10], as well as Phillips [11]. Hestenes [12] along with Ito and Takahashi [13] introduce a theory for gear design and manufacture. Grill [14] uses an \u201cequation of meshing\u201d to establish a relation between the curvature of one body to that of another body and applies his results in the context of gearing. A portal into certain limitations of existing crossed axis gear technology can be realized by focusing on Figure 3. The theoretical or ideal shape of these crossed axis gears is the \u201chour-glass\u201d or hyperboloidal shape shown. Current design and manufacturing techniques approximate a small portion of the hour-glass shape by a conical segment as shown. This approximation results in the following restrictions: face width\u00f1 minimum number of teeth\u00f1 spiral angle,\u00f1 pressure angle\u00f1 and hence restrictions on candidate gear designs. Moreover, these restriction are compounded with a cradle mounted face cutter as depicted in Figure 4" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003520_detc2014-34759-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003520_detc2014-34759-Figure3-1.png", + "caption": "Figure 3 Flexspline Stress Distribution", + "texts": [ + " The simulation model is a special version of harmonic drive with part number of HDSD-32-120. Its rated load is 200Nm. The tooth number of flexspline is 240, for the circular spline, tooth number is 242. If circular spline is fixed, wave generator is the input, from equation (1) the reduction ratio was 120:1. Geometry Model Wave Generator The wave generator\u2019s profile equation is: \ud835\udc45(\ud835\udf03) = \ud835\udc450 + \ud835\udc640 \u2217 cos (2 \u2217 \ud835\udf03) (9) Where \ud835\udc450 = 79.75\ud835\udc5a\ud835\udc5a, \ud835\udc640 = 0.4\ud835\udc5a\ud835\udc5a. Like a cam-follower mechanism, each tooth of flexspline\u2019s motion will governed by equation (9). Flexspline tooth profile Figure 3 is the existing harmonic drive\u2019s 2D model using contact element. The results can offer several possible optimization directions. 4 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 06/08/2015 Terms of Use: http://asme.org/terms From the figure3, we can see that there are two high stress areas. One is the contact stress at the contact point, the other one is the bending stress at the root fillet. The bending stress can be further decomposed into two components: the bending of the cup caused by wave generator forced deformation and the bending caused by the load. The contact stress can be reduced by a higher contact ratio in order to distribute the load to more tooth. The bending stress concentration can be reduced by a bigger fillet radius" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003026_amm.401-403.254-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003026_amm.401-403.254-Figure6-1.png", + "caption": "Figure 6 The relationship between the mechanical parameters of rotating machinery and \u03c6", + "texts": [ + " 28 we can know the angle range of loading area which determined by radial clearance is + = )(2 arccos)( max \u03d5\u03b4 \u03d5\u03b2 d d L P P (31) Function Q(\u03b2) of load distribution is > \u2264 \u2212\u2212= L LQQ \u03b2\u03b2 \u03b2\u03b2\u03b2 \u03d5\u03b5 \u03d5\u03b2 0 )cos1( )(2 1 1)()( 2 3 max (32) The relationship between the above parameters and the rotation angle of axis was show in Figure 5. From Figure 5 we can know that, because the radial force Fr(\u03c6) of bearing is continuous changing with the rotation of shaft, led the other parameters were changing with the rotation of the shaft. The relationship between parameters of rotating machinery and the rotation angle \u03c6 of shaft was show in Figure 6. Because of radial load F of bearing in ideal state is constant, other parameters are constant, there is obvious difference in two of them. In general, exciting force F far outweigh the shafting gravity G. When the inner ring has pitting corrosion fault and the fault in the load distribution, because of the relative position of pitting and load distribution area is not change, by Figure (26) the amplitude Qmax(\u03c6) of load area is changing along with the rotation of the rotating shaft. But the change of amplitude is small" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003102_gt2014-26275-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003102_gt2014-26275-Figure1-1.png", + "caption": "Figure 1. Multi-Degree of Freedom Rotor-Bearing System", + "texts": [ + " The resulting modal information is then fit with the rational polynomial method to estimate natural frequencies and damping ratios. The candidate identification method is then validated based on experiments with a full scale rotor test rig. The forces used in the simulation are then applied to the test rig, and predicted and experimental system responses using both the MIMO FRF and dFRF techniques are compared. The results of the investigation provide a new technique for improved modal parameter estimation in rotating systems. SYSTEM IDENTIFICATION METHOD Figure 1 shows a schematic of a rotor-bearing system with multiple degrees of freedom. This example includes four discs and a single bearing for illustration purposes. It can be described by the following differential equation: [ ] (t) =cbs msbms fqKKK qGGCqMM )++( )+(+)+( +\u03a9\u2212 &&& (1) In Eq. (1), the external forces f(t) represent unbalance forces, external fluid excitations such as net hydraulic load from each stage, and other effects. The force may be either a constant vector or a function of displacement or velocity of rotor vibration", + " Although the nodal displacement vector has four DOF, it is difficult to measure \u03b1 and \u03b2 in the process of modal identification for multi-degree rotor bearing systems. Therefore, the rotational DOF are neglected. The number of measurement locations for most rotor systems is limited, which results in a limited number of rotor responses that can be used to characterize the system. Equation (1) can be written more compactly as Eq. (2): ( ) )(tfKqqCGqM =++\u03a9+ &&& (2) Here, the displacement response q(t) corresponding to the single measurement plane in Fig. 1 can be described as: ( ){ }Ttytxt )()( =q (3) Similarly, the excitation force applied at a single excitation plane can be described as: 2 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use ( ) ( ){ }Tyx tftft =)(f (4) Through a Fourier transform of Eq. (2), the FRF matrix H(j\u03c9) for the measurement plane and the excitation plane can then be found. \u23ad \u23ac \u23ab \u23a9 \u23a8 \u23a7 \u23a5 \u23a6 \u23a4 \u23a2 \u23a3 \u23a1 = \u23ad \u23ac \u23ab \u23a9 \u23a8 \u23a7 )( )( )( )( j\u03c9F j\u03c9F HH HH j\u03c9Y j\u03c9X y x yyyx xyxx (5) The FRFs then contain information on the system natural frequencies and damping ratios, but this information is not guaranteed to be complete" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003538_amm.325-326.870-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003538_amm.325-326.870-Figure2-1.png", + "caption": "Fig. 2. The coordinate system of the face gear", + "texts": [ + " The errors are identified in Matlab software with graphical illustration too. The generation of the bevel gears are realized by using the face gear, [6]. Therefore, building the bevel gear\u2019s profile starts with the face gear\u2019s profile. The generated teeth is octoid I made by Bilgram-Reinecker gear cutting machine. The cutting tool executes the teeth generation describing a tooth flank of the imaginary face gear that the work piece subjected to processing is to engage. The generation is done by planing. In figure 2 is presented the coordinate system where OX1Y1Z1 is the coordinate system of the face gear and OXFYFZF is the coordinate system fixed at a flank. The parametrical equations of the face gear\u2019s flanks are: = \u22c5\u22c5\u2212\u22c5\u2212= \u22c5\u22c5\u2212\u22c5= \u03a3 vvuz tgvuvuy tgvuvux aa aa st ),( cossin),( sincos),( :)( 1 1 1 \u03c8\u03b1\u03c8 \u03c8\u03b1\u03c8 (1) = \u22c5\u22c5+\u22c5= \u22c5\u22c5\u2212\u22c5= \u03a3 vvuz tgvuvuy tgvuvux aa aa dr ),( cossin),( sincos),( :)( 1 1 1 \u03c8\u03b1\u03c8 \u03c8\u03b1\u03c8 (2) The used parameters are: \u03c8a the suitable angle of the quarter of the pitch, \u03b1 the pressure angle, u and v are liear parameters" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002673_holm.2014.7031067-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002673_holm.2014.7031067-Figure6-1.png", + "caption": "Fig 6. The structure of a typical balance force AEMR", + "texts": [ + " The development and production of this AEMR is limited by this problem. Since 1980s, relay design technology has developed rapidly especially in the USA and Japan. Because of the requirements for low-power relays used in printed circuit boards, polarized magnetic system is widely used. The design of polarized AEMR becomes maturity. The problem in the processing craft of permanent magnet has been partly solved and the adjustment technique problem for polarized relay has been overcome. (4) The balance force magnetic system structure of A EMR is showed in Fig. 6. The contact pressure of the static contact is provided by the return spring in the clapper magnetic system and balance armature structure magnetic system. Generally, it is less than the contact pressure of movable contact which is after closure (the contact pressure is provided by electromagnetic force). So in the closure process of static contact, large bounce will appear on static contact. It means a poor resistance to vibration and a poor impact resistance performance. To solve this problem, engineers in the US invented the balance force relay in 1966", + " The structure of permanent magnet returning type can improve the mechanical environment resistance ability of AEMR appropriately. In the other hand, the permanent magnetic circuit is an open magnetic circuit, so the returning speed of the armature is low which may lead to a serious ablation of the contacts. In order to improve the returning speed of the armature, the open magnetic circuit in balance armature type relay is designed to a closed magnetic circuit. Such AEMR structure is permanent magnet magnetic circuit returning structure. The balance force magnetic system (Fig. 6) uses this kind of returning mode. The AEMR of polarized magnetic system structure uses polarized magnetic circuit returning mode. The returning of armature is determined by the cooperation of the attraction and counterforce characteristics. The attraction characteristic is determined by the polarized magnetic circuit and the counterforce characteristic which is determined by the spring system. The polarized magnetic system structure is a better choice for the overall design for the bistable type of AEMR" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003565_j.mechmachtheory.2011.08.004-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003565_j.mechmachtheory.2011.08.004-Figure2-1.png", + "caption": "Fig. 2. Two output ways of the drive system: (a) stator output, (b) rotor output.", + "texts": [ + " Under condition that the tooth number of the stator is given, as pole pair number of the worm increases, the range of the speed ratio decreases. (3) If the worm and the stator have the lead angles in the same direction, the speed ratio of the drive system for rotor output is equal to the speed ratio for the stator output plus 1. If the worm and the stator have the lead angles in the opposite direction, the speed ratio of the drive system for rotor output is equal to the speed ratio for the stator output minus 1. Two output ways of the drive system are shown in Fig. 2. Hence, the speed ratios of the drive system are near to each other for the two output ways. So, the output torque of the drive system is also near to each other for the two output ways. The main difference between two output ways is: (1) For the stator output, the planetary motion vanishes in the drive system. It is favorable for the manufacture, mounting and adjustment of the drive system. (2) For the rotor output, the moment of inertia of the rotor is relative small. It is favorable for starting and braking the drive system" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003599_1.c031306-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003599_1.c031306-Figure5-1.png", + "caption": "Fig. 5 Illustrations of a) right side view of tire\u2013stone contact at the edge of the tire footprint and b) rearward view of contact between tire and stone.", + "texts": [ + " The vertical component of the initial tire\u2013stone contact velocity can be expressed as follows [1] using the derivation shown in the Appendix: vz V 2h=R h=R 2 p (1) The total velocity of initial contact between the tire and the stone can be found by q2 x2 h2 (2) v !q V R 2Rh p (3) v V 2h=R p (4) The angle relative to the vertical at which initial contact occurs is cos 1 vz=v (5) cos 1 1 h=2R p (6) For typical stone and tire parameters, h 20 mm andR 0:2 m, we obtain 13 . B. Stone at the Edge of the Footprint The initial contact speeds are slightly different when the stone is at the edge of the tire footprint (Fig. 5) and depends on the angle that the tire shoulder makes with the ground. The tire radius is assumed to be large compared with the stone radius so that the initial contact occurs at a position with an x coordinate close to that of the center of the stone. The derivation given in the Appendix leads to the following expression taken from [1]: vz V 2h=R h=R 2 b=R 2tan2 2b=R tan p (7) C. Tire Deflection Equation (1) assumes that the tire has no deformation from vehicle loading and rides tangentially to the runway surface" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001979_amm.284-287.461-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001979_amm.284-287.461-Figure4-1.png", + "caption": "Fig. 4 Gear mesh model of the gear pair", + "texts": [ + " The angular coordinates of the driving and driven gears are defined as didiii (t)t(t) . (1) where i 1 and 2 mean the driving and driven gears, respectively. i is the spin speeds of shaft. The displacement vectors for the mass centers of the gears are defined as iidiiidii ktetWjtetVtr \u02c6)]( sin)([\u02c6)]( cos)([)(\u02c6 ii . (2) where ij\u0302 and ik\u0302 are the unit vectors in iX and iY axes. The distance between centers, d , is changed to d after the motion, and d is defined as 2 12 2 12 )]()([])()([)( tWtWdtVtVtd dddd . (3) In Fig. 4, the gear pair is modeled as the equivalent stiffness mk and damping mc with transmission error )(tet along the pressure line between the teeth. The equivalent stiffness and damping are treated as time-varying coefficients in this paper. The pressure line is defined as the common tangent line of the base circles for the gear set, and the pressure angle p is defined as )](/)[(cos)( 21 1 tdRRtp . (4) where 1R and 2R are the radii of base circles of gears. The angle of the driven gear relative to the driving gear is represented by )]/()[(tan)( 1212 1 dVVWWt dddd " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001659_omn.2015.7288848-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001659_omn.2015.7288848-Figure5-1.png", + "caption": "Fig. 5. Experiment setup.", + "texts": [ + " We then fabricate the concave micromirror within the frame\u2019s hole and apply magnetic beads to the platform to complete this rotatable unit (Fig. 2). Figure 3 shows optical microscope images of the micro lens. With the water-repellent agent coated on the glass substrate, the contact angle of micro lens is 35.8\u00b0, larger than that on a surface without the water-repellent agent. Figure 4 shows the photograph of the 2 x 2 array of the back-side rotatable reflecting units. IV. RESULTS AND DISCUSSION The experiment setup is sketched in Figure 5. The retroreflected light from the device is reflected by the beam splitter, and then detected by the power meter. Testing is first done without tilting the concave mirror. The relations between the front-unit-back-unit gap and retroreflection efficiency at different incident angles are shown in Figure 6. The results show that the gap for maximum efficiency is about 1300 \u03bcm at small incident angles, and that as the incident angle increases, the efficiency decreases. Figure 7 demonstrates the mirror\u2019s tilt angle and retroreflection efficiency versus the magnet\u2019s distance from the substrate at an incident angle of 1\u00b0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003435_978-94-017-7300-3_16-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003435_978-94-017-7300-3_16-Figure1-1.png", + "caption": "Fig. 1 Geometry of a bilayer strip (undeformed). The tetrahedral mesh is created in COMSOL Multiphysics 4.3a", + "texts": [ + " The theory for bending of a film-on-substrate is well established. Stoney first studied the residual stress and the bending radius of a strained bilayer structure [17]. Timoshenko further provided an improved solution for bilayer structures where the film thickness is comparable to the substrate. A recent treatment on subject, which we briefly recap here, was developed by Tsui and Clyne [19]. An undeformed bilayer strip is oriented relative to a right-handed Reprinted from the journal 322 orthonormal basis d i , i = 1,2,3 as shown in Fig. 1, and the strip has a misfit strain \u03b5 along the d1 direction and a zero mis-orientation angle is considered (i.e., \u03c6 = 0 in Fig. 1). In [19], the radius of curvature is predicted as Rt = 1 + 4\u03b1\u03b2 + 6\u03b1\u03b22 + 4\u03b1\u03b23 + \u03b12\u03b24 6\u03b1\u03b2(1 + \u03b2) \u03b5 H1, (1) where \u03b1 = E2/E1 (E1 and E2 are Young\u2019s modulus of the bottom and top layer, respectively), \u03b2 = H2/H1, and \u03b5 is the misfit strain between the two layers. We employ a model based on linear elasticity theory and stationarity principles [22, 28, 33], where both the bending and mid-plane stretching are taken into account, as well as geometric nonlinearity. Without making a priori assumptions about the deformed shape, the total potential energy density per unit area of the ribbon is \u220f = \u222b H1 \u2212H 1 ( 1 2\u03c3 : \u03b3 )dz", + " When the residual stress (\u03c3 0) is present, the principal components are denoted by \u03c30x , \u03c30y , and \u03c30z, along the principal directions e1, e2, and e3 (e3 \u2261 e1 \u00d7 e2), respectively. Thus, in terms of the global Cartesian coordinate system we may write \u03c3 0 = \u03c30ijd i \u2297 dj , (4) where the non-zero components are \u03c30xx = \u03c30x cos2 \u03c6 + \u03c30y sin2 \u03c6, \u03c30xy = (\u03c30y \u2212 \u03c30x) sin\u03c6 cos\u03c6, \u03c30yy = \u03c30x sin2 \u03c6 + \u03c30y cos2 \u03c6, \u03c30zz = \u03c30z. (5) The finite element simulation used in this work is a full three-dimensional model using the structural mechanics module of COMSOL Multiphysics 4.3a [33, 38]. As shown in Fig. 1, the geometry is meshed using \u201cFree Tetrahedral\u201d elements in COMSOL. The boundary condition is set as follows. The point C is fixed in space, the point D has zero displacement in the y direction, while the points A and B have zero displacement both in the x and z directions. Other edges (or points) are free to move. In our model, we mainly consider two scenarios. In the first, there exists a uniform misfit strain between the two layers or the ribbon. In COMSOL Multiphysics, this is modeled by prescribing an initial strain that resides in one (or multiples) of the layers" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002671_0976-8580.86637-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002671_0976-8580.86637-Figure1-1.png", + "caption": "Figure 1: Symmetric hole-entry hybrid journal bearing coordinate system", + "texts": [ + " The available literature concerning the hybrid/hydrodynamic journal bearings indicates that the thermal effects together with the nonNewtonian behavior of lubricant due to additives mixed in the lubricants have been ignored in the analysis so as to obviate the mathematical complexity. The fluidfilm pressure ( )p distributions, fluid-film thickness ( )h profiles and fluid-film temperature ( )Tf distributions of orifice-compensated symmetric hole-entry hybrid journal bearing have been presented considering the combined influence of rise in temperature and nonNewtonian behavior of the lubricant for symmetric bearing configurations shown in Figure 1. The results presented in this paper are expected to be quite useful to the bearing designers. The analysis presented in the following subsection uses finite element method to model the complete journal bearing system operating with non-Newtonian lubricants. The mathematical model, which includes the viscosity variation due to temperature rise and non-Newtonian behavior of the lubricant, involves simultaneous solution of Reynolds equation, Energy equation and Heat conduction equation. The finite element formulation of these governing equations is described below with their boundary conditions" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001058_bfb0109985-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001058_bfb0109985-Figure1-1.png", + "caption": "Figure 1: Non-uniform sampling sliding mode on the manifold xa = 0 The dynamic behavior of this class of systems with two discontinuous functions when the second condition for the relative magnitudes holds (Eqn.(14)) can be summarized as follows:", + "texts": [ + "~2 > 0 , E 2 = U 1 -~- '0,2, E 2 > E 1 > 0 (18) 355 The following theorem describes the sampled data sliding mode charactersitics of the system trajectories. T h e o r e m 1 Suppose Xl( to ) = O, then there exists to < tl < t2 < t3 < . . . < t j . . . such that x l ( t j ) = O , j = 0 , 1 , 2 , . . . (19) = - ( 2 0 ) Furthermore, z~tj zx tj+l tj ( 1 ___~) - - - = + Ix (tj)l ( 2 1 ) If we interpret the time points t l , t2 , . . 9 as the sampling points of the continuous time trajectories x l ( t ) and x2(t), then this theorem shows that a discrete time sliding mode exists on the manifold xl = 0. Figure 1 illustrates the convergence of x2(t j ) at the sampled t ime points to the origin which is given in a Corollary to the above theorem: C o r o l l a r y 1 As tj -* oo, the phase trajectory of the system at the sampled time points t l , t2 , . . . , tj . . . remains on the manifold Xl = 0 and approaches the origin while the time difference between each successive sampled time points also tends to zero: limj_.oo x2(t j) -- O, z l ( t j ) -- 0 (22) limj__.~ At j = 0 (23) - A discrete t ime sliding mode occurs on the manifold xl = 0; sampling is however non-uniform, and the sampling period also tends to zero with increasing t ime - While on the manfiold xl = 0, x2(t) converges to the manifold x2 = 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001137_978-3-319-19743-2_14-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001137_978-3-319-19743-2_14-Figure1-1.png", + "caption": "Fig. 1. Our developed autonomous wheelchair.", + "texts": [ + " Against the case of cars the specific features do not always exist in the environments using the wheelchairs. Autonomous driving wheelchairs have to detect self-position without using the specific landmarks when GPS is unavailable. In this paper, we introduce our developed autonomous driving wheelchair [2] that can detect self-position without using specific landmarks and can decide a route in response to motion of pedestrians. Our autonomous wheelchair can travel in indoor-outdoor environments by the self-positioning and the route planning seamlessly. Figure 1 is our developed autonomous wheelchair \u201cMARCUS\u201d. Our wheelchair equips a 3D laser range finder (3D-LRF) unit for making maps, two wheel encoders for measuring self-velocity, and a 2D laser range finder (2D-LRF) for collision avoidance. The 3D-LRF unit obtains 3D points by rotating a 2D-LRF, which is a LMS151 manufactured by SICK. The rotation axes are a roll axis and a pitch axis. The range of rotation angle is from \u221222.5 degree to +22.5 degree for each axis. The measurement time of obtaining 3D points until rotating the range is 1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002385_cjme.2014.03.537-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002385_cjme.2014.03.537-Figure1-1.png", + "caption": "Fig. 1. Unified coordinate system of journal bearing", + "texts": [ + " Traditional model needs various geometric formulae which leads to heavy workload, complicated and changeable formulae, thus reduces scalability and promotion of the program. A universal computing model unifying geometric description of fix-pad journal bearing can be applied to both standard and non-standard journal bearings, which greatly reducing the programming workload for performance computing program and increasing expansion, and promotion of the program. A unified coordinate system with the bearing center at the origin is established, which takes the positive direction of y-axis as the angle starting line. As shown in Fig. 1, O is the bearing center, Oi is the center of pad i, Oj is the journal center, e is the eccentric distance, \u03b8 is the attitude angle. As shown in Fig. 1, \u03b4i is the eccentric distance of pad i, \u03b2i is the angle between eccentric distance and angle starting line, we named \u03b2i the eccentric distance angle or preload angle in dimensionless. In the unified coordinate system coordinates of Oi is: sin , cos . i i i i i i x y \u03b4 \u03b2 \u03b4 \u03b2 (1) Coordinates of Oj is: sin( ), cos( ). j j x e y e \u03b8 \u03b8 \u03c0 \u03c0 (2) In the unified coordinate system, eccentricity ratio and attitude angle of the journal center relative to pad center can be calculated: 2 2( ) ( ) , arctan " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001914_s11740-014-0582-7-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001914_s11740-014-0582-7-Figure5-1.png", + "caption": "Fig. 5 Coupling of the test rig with a hybrid kinematic machine tool", + "texts": [ + " The maximum size of the workpiece is 250 9 250 mm. In order to measure the disturbance- and feed-forces, a six-axes force-moment dynamometer is placed between the cross-table and the primary part. At this position the forces of the drive are directly measured and not biased by the friction of the guides. For investigations on the planar drive characteristics, the drive must be held in specific positions. In this case, the drive is been moved and held by a five axes hybrid kinematic machine tool, which is shown in Fig. 5. This machine tool was developed in an earlier project at the IFW. Since this project was completed, the kinematic was available for the research and was also used for three axis milling tests. 4.1 Single axis characteristics At first, the single axis characteristics, which are needed for controller design, are analyzed. For the electrical behavior of each axis, the resistance and the inductance must be identified. The acting forces between primary and secondary part are used to calculate the analytical model of the drive and to evaluate disturbing influences", + " The compensated quadrature current icomp, z is not shown in this figure. The feedforward control of the x-axis is built in the same way except for the force constant kM,x. Due to the feed-force saturation, the current depending force constant of the x-axis is stored in a lookup table. After the design of the controller with integrated compensation, the system was tested in an xz-axis milling process. A high-frequency spindle is installed in the hybrid kinematic machine tool to enable appropriate cutting speeds even with a 3 mm tool (see Fig. 5). The machine tool generates motion in the y-axis only. The spindle speed was adjusted to 18.000 min-1 and the feed per tooth was 0.03 mm. The depth of cut was set to 0.2 mm (Fig. 13). The tool path diameter of the milled circle was 10 mm. The result of this first milling experiment without cooling lubricant is shown in Fig. 13. Furthermore, the geometric deviation was measured with a coordinate measuring machine (Leitz PMM 866). The results are plotted in Fig. 14. To evaluate the influence of the hybrid kinematic on the process, also the measured motor position deviation is plotted" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002608_icmic.2014.7020736-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002608_icmic.2014.7020736-Figure1-1.png", + "caption": "Fig. 1. Diagram of the robotic air hockey table system.", + "texts": [ + " Experimental results are presented to verify its performance. The implementation results shown by our system is capable of performing real time calculations to move the linear actuator to its required position accurately. The implementation of this system entails architectures required by industries such as high volume manufacturing lines. Our work was able to prove that it is possible to achieve these forms of rapid mechanical movements by utilising affordable options such as computer vision. The robotic air hockey table we have built is shown in Fig. 1. It consists of a conventional air hockey table, a high speed camera, an electrical linear actuator to drive the mallet, and a standard personal computer for running the controller (not shown in the figure). The high speed camera (Basler A602f Black and White) has been fixed with a fish-eye lens and has been mounted 1200mm above the table with two side support bars such that it can capture the image of the whole table. The camera communicates with a real time linux operating system which is running software based on OpenCV" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002773_aim.2013.6584218-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002773_aim.2013.6584218-Figure6-1.png", + "caption": "Fig. 6. Virtual force using virtual spring-damper systems", + "texts": [ + " In other words, TEE now becomes the robot base frame (fixed to a target frame) and Trobot now becomes the robot end-effector frame. The internal robot configuration i.e. qi=1:3 can be assigned arbitrarily within the robot join limit as illustrated in Figure 5. As is seen, the above work-piece placement problem has been transformed into the problem of bringing P1, P2 and P3 into a common location while staying within the robot joint limit. Step 2: To bring P1,P2 and P3 to a common location, a virtual force (i.e. any two points are connected through a virtual zero-length-spring/with damper) is applied as illustrated in Figure 6. As the operational space control (OSC) framework presented in section 2 is a force-based OSC [11], it is straightforward to use the generated virtual forces (via spring-damper systems) to guide P1, P2 and P3 into a common location. Since the main task here is to bring all robot base frames together, this task will be executed at the highest priority (i.e. Eq. (1)) while other constraints such as singularity and/or collision avoidance can be incorporated into the null space controller (i.e. Eq. (3))" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002496_s10846-013-9971-y-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002496_s10846-013-9971-y-Figure6-1.png", + "caption": "Fig. 6 Three graphs showing the position of the bounded eigevalues of A\u0307m1", + "texts": [ + " 35 is presented in order to observe that the eigenvalues are negative as is shown in Fig. 5. Finally, it is needed to verify if there exists a positive constant \u03b21 such that \u2016A\u0307m1(t)\u2016 \u2264 \u03b21, Thus, one needs to verify whether the corresponding eigenvalues are bounded. The characteristic polynomial of \u2016A\u0307m1(t)\u2016 is given by where kc1(t), kc2(t), kc3(t), kc4(t), kc5(t) and kc6(t) are coefficients which depend of cos(\u03b8d). A numerical method is used again to verify if the eigenvalues are bounded. The Fig. 6 shows that these eigenvalues are bounded, thus there exists a positive constant \u03b21 such that \u2016A\u0307m1(t)\u2016 \u2264 \u03b21 and Am1(t) is uniformly exponentially stable. It is now necessary to find a positive constant \u03b2 such that \u2016F(t)\u2016 \u2264 \u03b2. \u2016F(t)\u2016 is computed by means of [10] \u2016F(t)\u2016 = \u221a \u03bbmax(F(t)T F(t)) where \u03bbmax(F(t)T F(t)) is the maximum eigenvalue of det(\u03bbI \u2212 F(t)T F(t)), this is det(\u03bbI \u2212 F(t)T F(t)) = \u03bb6 \u2212 k f (t)\u03bb5 = \u03bb5 (\u03bb\u2212 k f (t) ) (37) Consequently, six eigenvalues are obtained from Eq. 37. The first five eigenvalues are bounded because they are at the origin and the other eigenvalue can be written as a polynomial function of cos(\u03b8d) as follows k f (t) = bc8 cos(\u03b8d) 8 + bc7 cos(\u03b8d) 7 + bc6 cos(\u03b8d) 6 + bc5 cos(\u03b8d) 5 + bc4 cos(\u03b8d) 4 + bc3 cos(\u03b8d) 3 + bc2 cos(\u03b8d) 2 + bc1 cos(\u03b8d) \u2212 l4 hm4g4 ( k2 3 + k2 4 ) (38) where bc1, bc2, bc3, bc4, bc5, bc6, bc7 and bc8 represent finite values that depend on xd, zd, m, lh, Iyy, g, kb , \u03bbz1 , \u03bbz2 , k1, k2, k3 and k4" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002378_iet-cta.2014.0736-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002378_iet-cta.2014.0736-Figure1-1.png", + "caption": "Fig. 1 Definition of the coordinate reference frames", + "texts": [ + " Define m\u00d7n as the set of m by n real matrices. 03\u00d73 \u2208 3\u00d73 and 13\u00d73 \u2208 3\u00d73 denote the zero and the identify matrices with three orders, respectively. For any matrix A \u2208 n\u00d7n. The symbol \u2016A\u2016 stands for its Euclidean norm or its induced matrix norm. Given a vector x = [x1 x2 \u00b7 \u00b7 \u00b7 xn]T \u2208 n, we also define a sign vector sgn(x) = [sign(x1) sign(x2) \u00b7 \u00b7 \u00b7 sign(xn)]T with sign(\u00b7) \u2208 denoting the standard signal function. Consider a rigid satellite moving in a circular orbit. The coordinate systems used in the attitude control are shown in Fig. 1. The inertial frame Fi(XI , YI , ZI ) with its origin at the centre of the Earth is used to determine the orbital position of the satellite. The orbit reference frame Fo(XO, YO, ZO) rotating about the YO axis with respect to Fi, has its origin located in the mass centre of the satellite. The axes of Fo are chosen such that the roll axis XO is in the flight direction, the pitch axis YO is perpendicular to the orbital plane, and the yaw axis ZO points towards the Earth. The last reference system used is the body-fixed frame Fb(XB, YB, ZB)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003247_amr.960-961.1450-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003247_amr.960-961.1450-Figure2-1.png", + "caption": "Figure 2 Force diagram of the bottom scissor arm BC", + "texts": [ + " Stress analysis in hinge joint of the hydraulic scissor lift platform Considering the analysis is the instantaneous state at the initial position of the hydraulic scissor lift platform, therefore, the problem can be simplified as a issue which solving stress analysis in a system of static equilibrium. So, scissor arm force model can be simplified correspondingly: without considering the effects of inertia force , bias load as well as the weight of the hydraulic cylinder. For this model of a hydraulic cylinder promote three scissor arms, the hinge joint most bottom scissor arm has the maximum force. So the calculation should start from the very next level. If it meets the requirements of the stability and rigidity of other strengths, the other layers are bound to meet .As shown in Figure 2 and Figure 3. conditions, the model chosen in this paper is CDL1MP5: AL- 40\u03c6 -125mm; Material is 45 steel; Piston rod diameter d=70mm;Piston rod length l =1132mm; Hydraulic cylinder length 1089mm; Maximum force F = 13.413KNmax ; Elastic modulus E = 210GPa ; Proportional limit p\u03c3 = 280MPa ; Yield limit s\u03c3 = 350MPa . Calculate flexibility. Piston rod can be simplified into a rod which one end is fixed and the other is hinged, so the length factor \u00b5 = 2 . Because the piston rod cross-section is circular, so I d i = = A 4 (28) Therefore the flexibility \u00b5l \u00b5l \u03bb = = = 129 di 4 (29) Among them, l\u2014\u2014Piston rod length D\u2014\u2014Piston rod diameter \u03bb \u2014\u2014flexibility According to 2 p p \u03c0 E \u03bb = \u03c3 , Critical flexibility can be drawn p\u03bb = 86 p\u03bb \u03bb\u2265 Therefore Piston rod is greater flexibility piston rod, so the Euler formula can be used, Calculation the critical force: ( )2 92 8 2 2 210 10 1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003715_robio.2014.7090476-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003715_robio.2014.7090476-Figure2-1.png", + "caption": "Fig. 2. The workspace of a manipulator", + "texts": [ + " If it is in the dexterous workspace of the robot, the manipulability will be better. We may get the workspace of the robot from the product manuals. By analyzing the robotic Jacobian matrix, and considering the singular configurations, the dexterous configurations should be far from the workspace boundaries [10]. Velocity anisotropy analysis also demonstrates that the dexterous workspace is in the middle of the robotic reachable workspace. Hence we place the workpiece in the middle of the workspace (this process may not be accurate), as shown in Fig. 2. Planning with the offline programming system requires that the coordinates of the spherical workpiece center in the robotic coordinate frame is determined accurately, so that the coordinates of the trajectory points can be converted to the coordinates of TCP. It is impossible to obtain these coordinates accurately by measurement with a ruler, and calibration of the center coordinates is therefore necessary. A center point calibration method is proposed, which needs only one gradienter and a series of calculation [11]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001046_20070903-3-fr-2921.00038-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001046_20070903-3-fr-2921.00038-Figure2-1.png", + "caption": "Fig. 2. Inertial data fusion scheme", + "texts": [ + " The gyro signal \u00b7 \u03b8g is passed through a filter F1(s) while the accelerometer measurement \u03b8a is passed through a filter F2(s). The estimate \u03b8\u0302 is given as \u03b8\u0302 = F1(s) \u00b7 \u03b8g + F2(s)\u03b8a. The complementary filters are such that sF1(s) + F2(s) = 1 where F2(s) is a first order low-pass filter k k+s . For k = 0, F2(s) = 0, and F1(s) becomes a pure integration and \u03b8\u0302 diverges due to gyro drift. On the other hand for k = \u221e, F2(s) = 1, and F1(s) = 0 and the data from the gyro is completely suppressed. The complementary filter scheme is given in figure 2. The filter parameter k has been selected in practice to obtain an estimate \u03b8\u0302 that is as close as possible to the actual value of \u03b8. 5. CAMERA CALIBRATION Camera calibration is the process of determining the optical and internal camera geometric characteristics (intrinsic parameters) and the position and orientation of the camera with respect to a certain world coordinate system (extrinsic parameters) (Hartley et al., 2004). The simple webcam logitech has been used and the camera parameters are estimated by the two planes method which gives us a simple camera characterization (J" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003721_detc2014-34297-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003721_detc2014-34297-Figure7-1.png", + "caption": "FIGURE 7. Kinematic chains corresponding to cases from 7 to 12 of Table 1.", + "texts": [ + " This table shows, if there are, additional conditions to each of the subspaces S1,u\u03021 and S2,u\u03022 , the subspaces resulting after applying the relationships as well as the result of the direct sum of subspaces. Figure 6 shows the cases from 1 to 6 of Table 1. In each Figures 6(a), 6(b) and 6(c), the first three kinematic pairs generate in a different way the subalgebra gu\u03021 . Figure 6(a) shows especifically the cases 1 and 4; here the unit vectors u\u03022,1 and u\u03022,2, are parallel. Similarly, Figure 6(b) shows specifically the cases 2 and 5. Finally, Figure 6(c) shows specifically the cases 3 and 6. Figure 7 shows the cases from 7 to 12 of Table 1. In each Figure 7(a), 7(b) and 7(c), the first three kinematic pairs generate in a different way the subalgebra yu\u03021,p. Figure 7(a) shows specifically the cases 7 and 10; here the unit vectors u\u03022,1 and u\u03022,2, are parallel. Similarly, Figure 7(b) shows specifically the cases 8 and 11. Finally, Figure 7(c) shows specifically the cases 9 and 12. Figure 8 shows the cases from 13 to 18 of Table 1. In each Figure of 8(a), 8(b) and 8(c), the three first kinematic pairs generate in a different way the subspace S1,u\u03021 < xu\u03021 . Figure 8(a) shows specificaly the cases 13 and 16; here the unit vectors u\u03022,1 and u\u03022,2, are parallel. Similarly, Figure 8(b) shows specifically the cases 1 and 17. Finally, Figure 8(c) shows specifically the cases 15 and 18. Figures 6(a), 7(a) and 8(a), show different ways to generate the subspace S2,u\u03022 , when p4, p5 6= \u221e and u\u03022,1||u\u03022,2. Similarly, Figures 6(b), 7(b) and 8(b), show different ways to generate the subalgebra cu\u03022 , when u\u03022,1||u\u03022,2. Finally, Figures 6(c), 7(c) and 8(c), show different ways to generate the subspace S2,u\u03022 , in Figure 6(c), S2,u\u03022 < xu\u03022 , in Figure 7(c), S2,u\u03022 < gu\u03022 , while in Figure 8(c), S2,u\u03022 < yu\u03022,p. It should be mentioned that, both the Table 1 and Figures from 6 to 8, show only the kinematic chains where the first three kinematic pairs generate the subspace S1u\u03021 , while the last two kinematic pairs generate the subspace S2u\u03022 . However, besides the permissible permutations of the kinematic pairs that form the kinematic chains S1,u\u03021 and S2,u\u03022 proved in the propositions 4 and 5, the following proposition shows that many other permutations of the kinematic pairs are permissible" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003259_gt2013-94105-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003259_gt2013-94105-Figure11-1.png", + "caption": "Figure 11: Simple gear-to-gear test rig", + "texts": [ + " A secondary driven shaft was built onto a platform attached to the lathe saddle, with the cross-slide providing position control for tests. The secondary shaft fed into an electric motor, which acted as a magnetic drag brake, to provide mechanical resistance to the shaft. Each shaft was fitted with precision optical shaft encoders, which provided 500 pulses and 1 pulse per revolution. The 500 per rev outputs were provided as a quadrature pair in order that shaft direction can be established from any recorded data. Figure 11 shows the gear test rig configured for gear tooth inspection by a dynamic inspection idler wheel (white gear wheel), with eddy current sensors embedded in the teeth. The idler inspection wheel is easily replaced with an engineered steel gear wheel for gear-to-gear tests using a static eddy current sensor mounted adjacent to the gear wheels. For the gear-to-gear tests, a standard 10mm-diameter sensor was fitted onto an adjustable mount, and placed in various positions adjacent to the gears. The idler gear wheel used a plastic gear wheel fitted with three ferrite-cored probe coils" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001479_carpathiancc.2015.7145141-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001479_carpathiancc.2015.7145141-Figure3-1.png", + "caption": "Fig. 3 Photograph of test stand after reconstruction.", + "texts": [ + " Petermeier and Ecker deal with parametric vibration damping [9] of beams. We focused on the problem of the active vibration damping of the rotors with the use of parametric excitation of the journal bearings. The research work was carried out within the project (No. P101/12/2520) which was funded by the Czech Science Foundation. The results of tests are described in the second part of the paper. II. TEST STAND DESIGN The longitudinal section of the test stand after reconstruction is shown in Fig. 2 while some details can be seen on a photo in Fig. 3. The test stand consists of a rigid shaft of 30 mm diameter which supported in two cylindrical hydro-dynamic journal bearings of the length of 30 mm. The oil inlet is in the horizontal plane of symmetry of the bushing. The results of the experiments are presented only for the radial clearance of 55 f.tm. The bearing span is of 200 mm. An inductive motor for 400 Hz drives the rotor and therefore the maximum rotational speed is 23k rpm. As is evident from the photograph on Fig. 3 the supports of the piezoactuator were reinforced by additional bars. Bending and torsional load of the piezoactuators are excluded by using flexible tips and appropriate mounting procedure. III. TEST STAND INSTRUMENT ATION The position of the bearing journal is measured by the pair of the proximity probes which are based on the capacitive principle and are originated from the MICRO-EPSILON Company. The sensors are of the capaNCDT CS05 type with a measurement range of 0.5 mm. The advantage of the capacitive sensors is that it is not necessary to ground the shaft" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002274_amm.490-491.342-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002274_amm.490-491.342-Figure4-1.png", + "caption": "Fig. 4 Hit points and measured response points", + "texts": [ + " Exciting points should keep away with vibration nodes and nodal lines of different orders to excite as many modal shapes as possible, and the amplitude of response signals should be large enough to ensure good signal noise ratios of measured signals. MISO method is adopted in the experiment, in vertical direction, 19 hit points (v1-v19) and 1 measured response points are set appropriately. In horizontal direction, 13 hit points (h1-h13) and 1 measured response points are set, which is shown in Fig.4. In the process of test, each measured point is hit 3 times continuously with hammer. By the analysis of test data of HSR series of linear rolling guide, frequencies of the vertical and horizontal former several orders of linear rolling guide could be obtained as listed in Table 2, modal damping ratios are listed in Table 3. Fig.5 shows the comparison of results of finite element analysis and experiment. In the process of test, modality may be lost for various errors caused by instruments or personnel, but the result obtained by finite element calculation is a relatively complete solution, in event of only corresponding to the order, it is likely to produce the corresponding dislocation, so whether the modal shapes are corresponding must be considered in the process of identifying corresponding frequencies" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003617_icems.2014.7013783-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003617_icems.2014.7013783-Figure11-1.png", + "caption": "Fig. 11. The load schematics.", + "texts": [ + " 10, it is apparent that the opposite direction torque is required during the brief period, thus the look-up tables of appropriate direction in Fig. 6 is selected to follow the sign of the reference torque. The magnetic nonlinearity is considered in the torque controller, it results to decrease the ripple of the reference current. As the results, the position control using the MNLC has better performance than the MLC. D. Position Control Results of the MNLC under Load Condition The position control was carried out under two direction load condition. Figure 11 shows the load schematic that the weight is connected directly the rotor shaft through the string. The CW direction of load F as shown Fig. 11(a) is opposite direction of rotor rotation. On the other hand, the CCW direction is same direction of the rotor rotation as shown Fig. 11(b). Figure 12, 13 and 14 show the experimental results of the MNLC from 30 to 345 degree. The PI gains of speed controller were adjusted that the actual position and speed are followed to the reference ones without overshoot under load condition. Figure (a) shows the position and speed response results. The actual rotor position reached the 345 degree without overshoot. The step position response of CW and CCW load direction have almost the same performance comparison under no load condition. Figure (b) shows the reference torque from speed controller and the command current of each winding from torque controller using the look-up tables" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002263_iccas.2014.6987984-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002263_iccas.2014.6987984-Figure3-1.png", + "caption": "Fig. 3 Wake detection region", + "texts": [ + "1 Wake Modelling The wake is generated once and it spreads out slowly during a certain time. After the certain time has passed, the wake disappears. There are trajectory which is connected from the target ship. Because the ship moves continuously, not only the length but also breadth of the trajectory continue to grow. Figure 2 depicts the wake trajectory. For the modelling of the wake trajectory, it is defmed to two parameters which are the spreading angle and the maximum detection distance in Figure 2. Two parameters are dependent on the speed of the target ship. In Figure 3, \ufffd(t) is the position of the vehicle at time t and lk (I) is the width of the wake at the position Pk (I) . In the assumption of the constant speed v, after time interval;'.,t , the position of the vehicle is located in Pk+,(1 2). The width 'k(I,) becomes 'k(1 2) at time 1 2\u2022 The relation of 'k(I,) and 'k(1 2) is as follow: (1) PkRk and \ufffd.Lk can be obtained from simple calculation. (2) The width of the wake expands until the wake disappears. 2.2 Wake detection The wake detection region is determined by the sensor attachment location and the measurement angle of the sensor" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002241_1.4007860-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002241_1.4007860-Figure1-1.png", + "caption": "Fig. 1 (a) Coupled fields in a welding process and (b) strongly coupled weld pool dynamics-thermal fields", + "texts": [ + " The microstructure that develops in different parts of the HAZ depends upon the chemical composition of the material, the thermal cycle, and the austenite grain size. The microstructure evolution during welding affects the mechanical response of the welded structure. Hence, the welding process comprises several highly coupled physical phenomena\u2014thermal evolution of the welded structure, microstructural evolution/phase transformations, stress and distortion of the welded structure. The mechanical response of welds is sensitive to the close coupling between thermal energy distribution, microstructure evolution and mechanical (deformation/stress) behavior. Figure 1(a) depicts the coupling between the different fields in the modeling of welding processes. The solid arrows denote strong coupling and the dotted arrows denote weak coupling. It can be noted that the effect of thermal energy on the microstructure and mechanical response is dominant. However, the effect of microstructure evolution on thermal energy is negligible. Similarly, due to negligible heat generation from deformation, there exists weak coupling from mechanics to thermals. Hence, the thermometallurgical and thermomechanical problems are treated as one-way coupled", + " Kong and Kovacevic [16] developed a finite element model for the thermally induced residual stress analysis of a hybrid laser/arc welded lap joint. They found that increasing the welding speed causes the penetration and width of the weld bead to decrease, and the thermal stress concentration at the welded joint also reduces accordingly. Recently, del Coz Diaz et al. [17] compared the distortions between austenitic and duplex stainless steels under TIG welding through FE simulations. In all of these studies, the strong two-way coupled weld pool dynamics-thermal fields (Fig. 1(b)) have been ignored, i.e., thermal energy redistribution and change in the HAZ due to convective driving forces in the molten weld pool have not been considered. Previous research [18,19] has shown that heat transfer and fluid flow in the molten metal weld pool are driven by a complex interplay of driving forces: buoyancy, electromagnetic (Lorentz) force, surface tension gradient induced shear stress (Marangoni convection), plasma induced shear stress, etc. These driving forces strongly influence the thermal state of the welded structure through redistribution of the input heat from the arc" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003889_kem.584.142-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003889_kem.584.142-Figure1-1.png", + "caption": "Fig. 1 (Up Left): The 3-Dimensional drawing of the chemical \u2013 electric model car", + "texts": [ + " It is required by AICHE that the appropriate size of the car had to fit in the shoebox parameters when disassembled, and no larger than 400mm long, 300mm wide and 18mm high. Secondly, it was necessary to take into consideration the material of the wheels would be used. The main focus was on using rubber wheels. Those wheels could efficiently transfer electrical and mechanical power from the driving motor to the wheels without having to worry much about the wheels slipping on the ground at the start of each run. The Fig. 1 and Fig. 2 below, shows the preliminary design of the chemical \u2013 electric model car. The body had exactly 400mm length, 300mm width and 5mm thickness. This body was made from alloy steel, which had very high Modulus of Elasticity (Young\u2019s Modulus, around 200GPa). Fig. 3 shows the simulated deformation of the car body by using alloy steel under 100 N force uniformly added to the surface of the car body. The maximum displacement is 0.09248mm, which means that this 5mm thick car body can carry all of the chemical equipment on it with nearly zero meter displacement" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003489_amm.630.277-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003489_amm.630.277-Figure1-1.png", + "caption": "Fig. 1. Impulse face seal: 1- flexibly mounted ring; 2- ring mounted on the rotating shaft, 3- springs; 4- elastomer ring, 5- pump gland (housing); 6- cavities; 7- inlet ducts", + "texts": [ + " Also, it has become fairly common to appropriately modify the geometry of seal rings to obtain sealing with a self-regulation capacity adaptable to unsteady operating conditions. The aim of the paper is to determine the influence of the geometric parameters of the seal on the axial vibrations of the flexibly mounted ring. The results should be treated as supplementary to the generally accepted mathematical model of the impulse seal derived by V.A. Martsinkovsky, which is discussed in Ref. [2]. A schematic diagram of the impulse face seal is shown in figure 1. The impulse seal consists of two rings, both with specially prepared faces. The flexibly mounted ring, i.e. ring 1, has some cavities 6 evenly distributed round the face. The rotating ring, i.e. ring 2, on the other hand, has some radial ducts 7 on the edge, which are open to the space containing the working medium under preset pressure, p1. All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002773_aim.2013.6584218-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002773_aim.2013.6584218-Figure4-1.png", + "caption": "Fig. 4. 3-DOF RRR planar robot", + "texts": [ + " For visualization purposes, the proposed method will be aided by a simple 3-DOF RRR planar robot with link length of 2m, 3m and 1m. The extension of this method for higher DOF robots is straightforward with the aid of major multi-body dynamics software such as MSC Adams or FunctionBay Recurdyn. The work-piece placement problem can be summarized as follows: (i) Consider a set of critical target frames which repre- sents the desired trajectory of the robot TCP. Note that only targets at critical location (3 targets in this case) are considered to reduce the computation time. For example, Figure 4 depicts the 3 critical target frames (T1:3) along the desired path w.r.t the work-piece frame (Twobj). (ii) Assume that the following information is available: \u2022 the robot kinematic model: link length in this case \u2022 the robot dynamics: since this searching pro- cess is done in the virtual environment, a rough estimation of the robot dynamics is sufficient. Note that the same model parameters are used in both modelling and control to achieve dynamically decoupling between the task space and null space [15]", + " Although the proposed method can take into account the collision avoidance problem by imposing artificial constraints (such as the potential field) during the searching process, the problem of collision avoidance is not considered in the simulation of this paper. It is worth noting that collision avoidance using artificial force field is a common topic in the literature that requires a separated discussion/treatment. (iii) To determine T wobj robot in such a way that the robot can reach all target points (T1:3 in this example) while staying within the robot joint limits. As can be seen from Figure 4, although this is a 2D problem, it is not trivial to find such a solution when constraints are taken into account (joint limit in this case). To address this problem, the following algorithm is pro- posed: Step 1: Attach the robot TCP (TEE in Figure 4) to each target frames under that assumption that the robot base frame is no longer constrained at Trobot. In other words, TEE now becomes the robot base frame (fixed to a target frame) and Trobot now becomes the robot end-effector frame. The internal robot configuration i.e. qi=1:3 can be assigned arbitrarily within the robot join limit as illustrated in Figure 5. As is seen, the above work-piece placement problem has been transformed into the problem of bringing P1, P2 and P3 into a common location while staying within the robot joint limit" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003570_j.jappmathmech.2011.11.002-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003570_j.jappmathmech.2011.11.002-Figure1-1.png", + "caption": "Fig. 1.", + "texts": [ + " 1 when analysing a mechanical model of the oscillations of molecules. Some problems of the dynamics of such a pendulum were then considered in a number of investigations (see, for example, Refs 2-5). Suppose a point mass M moves in a uniform gravity field. The point is attached to one of the ends of a weightless spring, the other end of which is attached to a fixed point O. Motion occurs in a fixed vertical plane Oxy. The position of the point is specified by its polar coordinates \u03d1, (the upper part of Fig. 1). Suppose m is the mass of the point, g is the gravitational acceleration, 0 is the length of the spring in the unstretched state and k is its stiffness. We have the following expressions for the kinetic and potential energy of the point M where the dot denotes differentiation with respect to time t. The Lagrange function L = T \u2013 . The equations of motion have the form (1.1) We will say that the pendulum is in a state of uniform rotation if during all the time of motion the value of the derivative \u03d1\u0307 is constant and is non-zero", + "2) In what follows, it is more convenient to write condition (2.2) in dimensionless form. Suppose is the length of the spring in the equilibrium state of the pendulum along the vertical. Then (2.3) The quantity is equal to the ratio of the frequencies \u221a g/ and \u221a k/m of small oscillations of the point M in the neighbourhood of its equilibrium position on the vertical. Bearing in mind Eqs (2.3), condition (2.2) can be written in the form (2.4) For uniform rotation of the pendulum, the trajectory of the point M in the Oxy plane is a Pascal limac\u0327 on. In Fig. 1 we show the qualitative form of the trajectory when 2 < 1/4 (on the left) and when 1/4 < 2 < 2/5 (on the right). When investigating the stability of the rotation (2.1) we will use the Hamilton form of the equations of motion. The Hamiltonian is equal to the sum T + , in which the quantities \u03d1\u0307, \u0307 are expressed in terms of the momenta p\u03d1 , p , defined by the equalities We will introduce the dimensionless variables q1, q2, p1 and p2 using the formulae (3.1) Replacement (3.1) is (see, for example, Ref" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001428_9781118869796.ch17-Figure17.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001428_9781118869796.ch17-Figure17.2-1.png", + "caption": "FIGURE 17.2 Schematic structure of an insertion MEFC designed to use biochemical energy in living organisms.", + "texts": [ + " One possible solution is to use a needle-shaped insertion anode to approach fuels through the skin. Another important consideration is that oxygen is limited in many organisms to a lower concentration than that of sugars. In addition, biofluids may contain reaction inhibitors for cathodic enzymes, such as ascorbic acid and urate [14]. Taking these considerations into account, an insertion MEFC with a needle anode and a gas diffusion cathode designed to be exposed to atmospheric air was assembled, as illustrated in Figure 17.2. The insertion MEFC device described above (Figure 17.3) consists of a fructose dehydrogenase (FDH)-modified needle anode for fructose oxidation and a bilirubin oxidase (BOx)-modified carbon paper cathode for reduction of oxygen in ambient air (Figure 17.3). The anode and cathode are assembled using a polydimethylsiloxane (PDMS) chamber and an ion-conducting agarose hydrogel (pH 5.0) as the inner matrix. This structural design allows the use of fructose in grapes and also protects the cathode from reaction inhibitors in the grape juice" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001567_icma.2015.7237783-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001567_icma.2015.7237783-Figure2-1.png", + "caption": "Fig. 2 Eyebrow mechanism", + "texts": [ + " To control its shape and position, each eyebrow was driven by 2 towing points which are connected to motors by elastic steel-wire. The elastic steel-wire crossed a nylon tube which is fixed on head skeleton. By pivot pins, one end of the wire is connected to the motor rocker and another end is connected to the eyebrow towing point. The steel-wire is 2 mm thick to make it plastic. When the motor rotates, the steel wire will make deformation to move through the nylon tube. Eyebrow mechanism is showed in Fig. 2 1944978-1-4799-7098-8/15/$31.00 \u00a92015 IEEE Proceedings of 2015 IEEE International Conference on Mechatronics and Automation August 2 - 5, Beijing, China Fig. 3 is CAD drawing of eye mechanism. Eyeball mechanism has 4 DOFs to pitch and yaw eyeballs. Each motor and eyeball are connected by RSSR mechanism, which is a space linkage mechanism. The two eyeballs have separate mechanism, so they can make independent movement. The eyeball is 30mm diameter. They are manufactured by 3D printer. Each eyeball is equipped with a CCD camera of 13" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003678_1.4024704-Figure14-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003678_1.4024704-Figure14-1.png", + "caption": "Fig. 14 z-direction deformation (in meters) of the workpiece. Welding away from fixed face AA0B0B.", + "texts": [], + "surrounding_texts": [ + "The GTA welding process and the various transport processes involved are discussed in detail in Part I [2]. The mathematical model can be divided into two parts: (a) weld pool dynamics modeling and (b) structural analysis modeling. In the weld pool dynamics modeling, the melting/solidification problem is handled using the enthalpy-porosity formulation. The molten metal flow in the weld pool is obtained using the governing equations of continuity, momentum and energy, based on the assumption of incompressible laminar flow. The Navier\u2013Stokes (N\u2013S) momentum equation takes into account the mushy zone through the momentum sink term, and includes the electromagnetic (Lorentz) force as a body force term. The Lorentz force is determined using the current continuity equation in association with the steady state version of the Maxwell\u2019s equation in the domain of the workpiece for the current density and magnetic flux. The structural analysis model is developed based on isotropic material behavior. The elastic response is handled using the isotropic Hooke\u2019s law with temperature dependent Young\u2019s modulus and Poisson\u2019s ratio. For the inelastic response or plasticity, incompressible plastic deformation is assumed with rate-independent plastic flow and vonMises yield criterion. The yield strength is considered as a function of temperature only. Also, the bilinear isotropic hardening model is employed to consider the material strain-hardening behavior. The mathematical models for both weld pool dynamics and structural analysis have been discussed in extensive detail in Part I and hence is not represented here. However, it is to be noted that the analysis in this study ignores the influences from the arc pressure and a flat weld pool surface is assumed. These assumptions are reasonable for the present study and discussed in detail in Part I of the present study. Also, the boundary and initial conditions used in the mathematical model are described in detail in Part I." + ] + }, + { + "image_filename": "designv11_84_0002868_s12206-015-0143-9-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002868_s12206-015-0143-9-Figure3-1.png", + "caption": "Fig. 3. Operating modes and the nacelle operation for each mode.", + "texts": [ + " Moller international\u2019s M400 skycar could become a representative VTOL PAV if it succeeds in flying. A VTOL PAV brings true \u2018door-to-door\u2019 transportation capability, and requires minimum infrastructure for take-off and landing. Meanwhile, a VTOL PAV with a quad tilt prop has been proposed, which was awarded a grand prize from the 3rd International Unmanned PAV Competition in South Korea [7]. The design concept of the proposed PAV started from a car, as shown in Figs. 1 and 2, with four nacelles each housing a pair of wheels and a propeller. Fig. 3 shows the three modes of the proposed PAV. Four nacelles are tilted outward to lower the height of the center of gravity (CG) point during the driving mode, straight-up for the VTOL mode, and can be tilted for*Corresponding author. Tel.: +82 51 510 2310, Fax.: +82 51 513 3760 E-mail address: bskang@pusan.ac.kr \u2020 Recommended by Associate Editor Deok Jin Lee \u00a9 KSME & Springer 2015 ward and backward during the forward flight mode. The proposed design does not fully transit to a fixed-wing mode, but is intended to maintain a low speed, and use tiltable nacelles", + " 4 shows the flight data that the inboard wing caused a serious uncontrollable state during the VTOL mode flight. It was barely controllable, but the ground effect under the wing caused an uncontrollable state near the ground. The inboard wing had to be modified to a circular support in the initial prototype due to its strong aerodynamic interference that is also prevalent in tilt-rotor [11]. Therefore, unlike a tilt-wing concept, the current prototype has 3 modes: i.e., the \u2018driving\u2019, \u2018VTOL\u2019, and \u2018forward flight\u2019 modes as il- lustrated in Fig. 3. During the forward flight mode, the nacelle tilt angle is limited under \u00b120\u00b0, where the dynamics is similar to a pure quad-rotor, and no additional aerodynamic control surfaces are needed. Fig. 5 shows the frame structure and applied mechanisms. It is mainly constructed with two longitudinal carbon pipes and four aluminum bulkhead frames for a light structure. Two torque tubes are installed on the longitudinal pipes by aluminum housings, where the front torque tube is installed with two nacelles", + " Each front nacelle is equipped with a propulsion motor, two wheels, a steering mechanism and a steering servo actuator. The rear torque tube is also installed with two nacelles, where each nacelle is equipped with a propulsion motor, two wheels, a driving motor with a reduction gear and a brake drum. The mechanism of tilt actuator is also included in the Fig. 5, where there are total of 8 high-torque servo actuator coupled in a pair that control tilt angle. A stopper pin is also added to release the power of servo actuators during the driving mode. As it was also shown in the Fig. 3 for the orientation of tiltable nacelles during each mode, during the driving mode, nacelles are fixed to 45\u00b0 facing each other to lower the overall height, and locked by the stopper pin to save power consumption. When the VTOL mode is initiated, two pairs of servo actuators deliver control forces to each torque tube without any gear to simplify the overall structure [8]. So far, controls are achieved by manual control during the driving mode, and the manual operator controls the steering angle of front wheels, driving motor and brake drums of rear wheels" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003999_s12283-014-0158-y-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003999_s12283-014-0158-y-Figure3-1.png", + "caption": "Fig. 3 The forces acting on the balls during the impact", + "texts": [ + " The assumption of a point contact has also been used by other researchers such as Kondic [22] and Domenech [23]. In this paper, collisions between two snooker balls are studied using the principles of impact mechanics. The next section describes theoretical derivations. Simulation results are subsequently presented and discussed extensively, followed by a conclusion section. The balls are assumed to be moving on a flat surface; hence, the point of their contact will be at a height of R from the surface. This is depicted in Fig. 3. 2.1 General equations of motion In Fig. 3, for ball C, for the linear motion along X, Y and Z directions F1 \u00fe FC R;x \u00bc M\u20acxC G \u00f01a\u00de FI \u00fe FC R;y \u00bc M\u20acyC G \u00f01b\u00de F2 \u00fe FC N mg \u00bc M\u20aczC G \u00f01c\u00de Here, F denotes the instantaneous normal force acting on any of the points of contact between the balls and the table interface, as shown in Fig. 3. In Eq. (1a), superscripts denote the sphere to which a particular parameter belongs to. The subscript G stands for the centroid values of a sphere. Thus, \u20aczC G denotes the centroid acceleration of C along the Z-axis. By considering an infinitesimal time period of Dt, the increment and the accumulated impulse values can be expressed as, DP \u00bc Zt\u00feDt t F dt \u00f02a\u00de and, P \u00bc X DP \u00bc Z t 0 F dt \u00f02b\u00de The impulse\u2013momentum change relationship along the above directions result in the following equations. For cue ball C, at time t, consider an increment Dt in time, from Eqs", + " Slip speeds along the X and Z axes, respectively, are as follows: _xA \u00bc _xC A _xO A \u00bc s PI\u00f0 \u00de cos U PI\u00f0 \u00de\u00f0 \u00de \u00f05a\u00de _zA \u00bc _zC A _zO A \u00bc s PI\u00f0 \u00de sin U PI\u00f0 \u00de\u00f0 \u00de \u00f05b\u00de The normal component of relative velocity, _yA \u00bc _yC A _yO A \u00bc _yC G _yO G \u00f05c\u00de For the nominal slipping speeds to be along the positive X and Z axes, when the balls are sliding on each other at their contact point A, from the Amontons-Coulomb law, DP1 \u00bc lbb cos U PI\u00f0 \u00de\u00f0 \u00deDPI \u00f06a\u00de DP2 \u00bc lbb sin U PI\u00f0 \u00de\u00f0 \u00deDPI \u00f06b\u00de where lbb is the coefficient of sliding friction between the balls. Notably, depending on the value of vertical sliding velocity between the balls, i.e., _zA as given in Eq. (5b), some of the impulses in the equation sets (3a) or (4) will be zero. If _zA is negative, ball C will have more downward velocity (along the Z-axis) at the contact point A, and the frictional impulse along Z, DP2, between the balls will be acting on the balls in the directions as shown in Fig. 3. This condition is given by, DP2 [ 0 \u00f06c\u00de If ball C is to remain on the table, from Eq. (3c), DP2 \u00fe DPC N\\0 \u00f06d\u00de The conditions in (6c) and (6d) can only be satisfied when, DPC N\\0 \u00f06e\u00de and apparently, it is impossible to satisfy the condition in (6e) as the table cannot apply a \u2018negative\u2019 reaction on the ball. Thus, DPC N \u00bc 0 \u00f06f\u00de Condition (6c) in turn says that the associated frictional impulses are also absent, i.e., DPC x \u00bc 0 and DPC y \u00bc 0 \u00f06g\u00de Here, ball C will lift up from the table, like the cue ball in a \u2018jump\u2019 shot; Kondic [22] and Domenech [23] also acknowledge this effect. However, it is assumed in this paper that during the time of the impulse, it remains at the same spatial location, just above the table, without altering the geometrical configuration presented in Fig. 3. This assumption is reasonable since the time of impulse between two balls is very small and is in the range of 0.3 ms [21]. Conversely, if _zA is positive, then, DPO N \u00bc 0; DPO x \u00bc 0 and DPO y \u00bc 0 \u00f06h\u00de Finally, if _zA is zero, DP2 \u00bc 0 \u00f06i\u00de When (6i) prevails, both the balls will be either in contact with the table or airborne. Hence, DPC N \u00bc DPO N \u00bc 0 \u00f06j\u00de and the associated friction components will also be zero. If ball C is touching the table at any instant during impact, to satisfy the condition, _zC G t \u00fe Dt\u00f0 \u00de _zC G t\u00f0 \u00de \u00bc 0 and from (3c) DPC N \u00bc DP2 \u00bc lbb sin U PI\u00f0 \u00de\u00f0 \u00deDPI \u00f07a\u00de Else, if ball O is on the table, to satisfy the condition, _zO G t \u00fe Dt\u00f0 \u00de _zO G t\u00f0 \u00de \u00bc 0 it can be shown that DPO N \u00bc DP2 \u00bc lbb sin U PI\u00f0 \u00de\u00f0 \u00deDPI \u00f07b\u00de If ball C touches the table at point D with the table plane, slip s0 and slip angle U0 with the X-axis (s0 will lie on the XY plane), and from (7a), for the sliding condition, DPC x \u00bc ls cos U0 PI\u00f0 \u00de\u00f0 \u00deDPC N \u00bc lbbls sin U PI\u00f0 \u00de\u00f0 \u00de cos U0 PI\u00f0 \u00de\u00f0 \u00deDPI \u00f08a\u00de Similar expressions can be derived for along the Y-axis and those for ball O along X and Y involving slip s00 and slip angle U00 with the X-axis (s00 will be on the XY plane)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.6-1.png", + "caption": "Fig. 2.6. Successive rotations of an object about axes of current frame", + "texts": [ + " By recalling the meaning of a rotation matrix in terms of the orientation of a current frame with respect to a fixed frame, it can be recognized that its columns are the direction cosines of the axes of the current frame with respect to the fixed frame, while its rows (columns of its transpose and inverse) are the direction cosines of the axes of the fixed frame with respect to the current frame. An important issue of composition of rotations is that the matrix product is not commutative. In view of this, it can be concluded that two rotations in general do not commute and its composition depends on the order of the single rotations. Example 2.3 Consider an object and a frame attached to it. Figure 2.6 shows the effects of two successive rotations of the object with respect to the current frame by changing the order of rotations. It is evident that the final object orientation is different in the two cases. Also in the case of rotations made with respect to the current frame, the final orientations differ (Fig. 2.7). It is interesting to note that the effects of the sequence of rotations with respect to the fixed frame are interchanged with the effects of the sequence of rotations with respect to the current frame" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003852_iceceng.2011.6057719-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003852_iceceng.2011.6057719-Figure1-1.png", + "caption": "Fig. 1 The Arduino Duemilanove together with its connections with the servo motors", + "texts": [], + "surrounding_texts": [ + "The Arduino Duemilanove is a microcontroller board with either ATmega168 or ATmega328. The latter one was used for the development of the control for the system. It has a 16MHz crystal oscillator, 6 analog and 14 digital input output pins. A universal serial bus connection was also available to burn the code onto the controller and for serial communication. Arduino IDE was used for compiling the C/C++ code and burning it onto the controller. Proteus VSM, Virtual Breadboard and MATLAB were used for preliminary simulations." + ] + }, + { + "image_filename": "designv11_84_0002000_scis-isis.2014.7044727-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002000_scis-isis.2014.7044727-Figure1-1.png", + "caption": "Fig. 1 Definition of collision-free area", + "texts": [ + " However, at present, because of its cost, LIDAR is only used for research purposes and not for consumer application [3]. Depth cameras operating with projected infrared patterns, such as the Microsoft Kinect and Asus xtion series, have recently become available on the consumer market, enabling a low cost, compact, and lightweight device option compared with other conventional sensors. These cameras can capture three-dimensional range information. In this paper, we present a new collision-free area detection system using a depth camera for mobile robot navigation. II. PROPOSED SYSTEM AND PROBLEM DESCRIPTIONS Fig. 1 illustrates the configuration of the proposed system, which comprises a mobile robot and an on-board depth camera. The depth camera can capture three-dimensional range information, including obstacles, ground, and slope, as threedimensional coordinate-based point cluster. Based on the measured point cluster, we need to find the collision-free areas for the mobile robot. To detect collision-free areas using the depth camera, we define the following two assumptions: A1) The ground surface of the collision-free area is smooth and has no bumps that cannot be overridden by the mobile robot" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003671_detc2011-48794-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003671_detc2011-48794-Figure4-1.png", + "caption": "Figure 4. Toroidal CVT drive model. The yellow cylinders are the revolute joints at the center of the shafts, rollers and roller angle controllers.", + "texts": [ + " This step is done in parallel by running each list of elements identified in step 1.c on one processor. iii. Find the nodal values at the current time step using the semi-discrete equations of motion and the trapezoidal time integration rule (Eqs. 1-5). iv. Execute the prescribed motion constraints which set the nodal value(s) to prescribed values. v. Go to the beginning of step 2. The model of a typical toroidal CVT drive used in automotive applications that was presented in [6] is modeled using the present method. Figure 4 shows the toroidal drive model. It consists of an input toroidal disk, an output toroidal disk, two rollers, and a two roller angle controllers. The input shaft, output shaft, and roller angle controllers are connected to ground using revolute joints. The rollers are mounted on the roller angle controllers using revolute joints. Figure 5 shows a sketch of the system along with the major system dimensions. Table 2 shows the values of the dimensions, mass/inertia and discretization parameters of the toroidal drive model" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003632_amr.903.215-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003632_amr.903.215-Figure1-1.png", + "caption": "Fig. 1 Geometry of two-layered film journal bearing with partial slip configuration", + "texts": [ + " Results of nondimensional load capacity and coefficient of friction in one dimensional journal bearing with partial slip under steady state are analyzed for different values of nondimensional slip coefficient ( ) and eccentricity ratio ( ). All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 132.174.254.159, Pennsylvania State University, University Park, USA-26/05/15,11:28:21) A one-dimensional analysis of two-layered partial slip journal bearing is considered in the analysis. Figure 1 shows the schematic of two-layered film journal bearing with partially slip configuration. The partially slip extent is . Neglecting the pressure variation along the film thickness, and considering that pressure variation is only along the sliding direction, the simplified momentum equations for each of the velocity components is in 0 , and in . (1) Navier slip boundary conditions are imposed on the part of bearing surface, no slip conditions are imposed on the other part of the plain bearing surface and on the journal surface" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002958_20110828-6-it-1002.00858-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002958_20110828-6-it-1002.00858-Figure3-1.png", + "caption": "Fig. 3. (a) GLMAV prototype - (b) Aerodynamic balance.", + "texts": [], + "surrounding_texts": [ + "From this point onward, the purpose consists of determining of the aerodynamic parameter values \u03b1, \u03b2, Cz, \u03b31 and \u03b32, by using the GLMAV model described in section (2) as well as the measured input-output data collected from experiments." + ] + }, + { + "image_filename": "designv11_84_0002937_amc.2014.6823323-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002937_amc.2014.6823323-Figure5-1.png", + "caption": "Fig. 5. The robot is undergoing a pure rotation when starting to ascend a slope", + "texts": [ + " Then we introduce the iterative learning control into the admittance in order to achieve the same \u201cload feeling\u201d as the horizontal plane. When the user and the robot are not on the same plane, for example, when the robot starts to ascend the slope (as phase B shown in Fig. 4), in which, the user is on a horizontal plane while the front wheels of the robot are already on the slope, the respective velocities of the user and the robot will be different. Now we analyze the movement in phase B of starting to ascend a slope. As shown in Fig. 5, F and R are respectively the centers of the front and rear wheels of the robot; A is the user\u2019s holding point at the handle bar of the robot; H is the height of the handle bar from A to R. Since the front wheels of the robot are on the slope, their velocity VF is along the slope too. However the rear wheels of the robot is still on the horizontal plane, thus their velocity VR is along the horizontal direction. Therefore, at this instant of time, the whole robot can be considered undergoing a planar motion and rotating around an instantaneous center of rotation at an angular velocity \u03c9. The instantaneous center of rotation can be easily found with the two different velocity directions of its front and real wheels. As shown in Fig. 5, P is the instantaneous center of rotation of the robot, and two straight lines PR(LPR) and PF (LPF ) are respectively vertical to the velocity VF and VR. In ascending the slope, since the robot can be considered undergoing a pure rotation around an instantaneous center of rotation, the velocity of each point of the robot is different and dependent on the distance from the instantaneous rotation center. The velocity of point R (VR) and point A (VA) are respectively given by VR = LPR \u00b7 \u03c9 (4) VA = LPA \u00b7 \u03c9 (5) where, LPR and LPA are respectively the distance from point P to R and the distance from point P to A; VA is the velocity of the user\u2019s holding point A at the handle bar of the robot. When \u03b1 is the angle between straight line PA and PR, the relationship between VH and VA is given by VH = VA cos\u03b1 = LPB \u00b7 \u03c9 (6) Note that VH is in fact the user\u2019s walking velocity. In Fig. 5 there is LPB = LPR \u2212H cos \u03b8 (7) Consequently, the relationship between VH and VR is VH VR = LPB LPR < 1 (8) On the other hand, according to sine law in \u0394PRF shown in Fig. 5, there is LPR = cos(\u03b80 \u2212 \u03b8) sin \u03b80 LFR (9) Thus, the desired walking velocity of the user is given by VH1 = s1 \u00b7 VR (10) where, s1 = 1\u2212 H cos \u03b8 sin \u03b80 L cos(\u03b80 \u2212 \u03b8) < 1 (11) The above two equations show that the velocity at the point A of hander bar where the user is holding is less than the velocity of the robot, when the robot starts to ascend the slope. This is because there is a backward component caused by rotation movement when the inclination angle of the robot gradually increases during the robot traveling along the slope while its rear wheels are still on the horizontal plane" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001539_chicc.2015.7260615-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001539_chicc.2015.7260615-Figure10-1.png", + "caption": "Fig. 10: The picture of the roller bearing test rig", + "texts": [ + "9 that LMD marginal spectrum is capable of identifying the frequency and the corresponding energy distribution in the multi-component simulated signal, which making the overall energy-frequency distribution displayed more clearly. In this paper, a rotating machinery vibration fault testing platform produced by Jiangsu Qianpeng Diagnostic Engineering Company Limited is selected to implement the experiment. The platform is composed of signal collection and processing system, lubrication system, loading system and transmission system, as shown in Fig. 10. As is shown in structure diagram Fig. 11 of the experiment, the teeth number of small gear is 37 and the large gear number is 59. A B Fig.11: The structure sketch of gear in the experiment As shown in Fig. 11, A stands for the small gear and B represents the big gear. I and II represent respective transmission shaft. The crack of the small gear is simulated by processing small grooves whose depth is 0.5mm and width is 0.2mm with an electric discharge machine. In the experiment, the vibration signal of axial and radial direction is collected by an acceleration sensor" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002418_nano.2011.6144588-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002418_nano.2011.6144588-Figure3-1.png", + "caption": "Fig. 3: Magnetic actuation of Ni NW using a rotating magnetic field. (a) Schematic of a NW rotating horizontally in the X-Y plane induced by the magnetic torque \u03c4m. (b) Schematic of a NW rotating vertically with respect to the X-Y plane. Vector A is the normal of the plane of rotation of the rotating magnetic field.", + "texts": [ + " The Helmholtz coils are placed around the lens of the probe station (Signatone model S-1160) and the tank is placed at the centre of the coils. Before the magnetic actuation and manipulation, the as-fabricated Ni NWs suspension were separated in DI water by a 15 min sonication, and then a drop including 6 \u03bcm diameter microparticles was added in the tank. The magnetic field (\u23d0B\u23d0=1-10 mT) was turned on after the NWs and microparticles sank near the Si substrate. In the experiments, the NWs were actuated by a rotating field that was either parallel or perpendicular to the substrate, as shown in Fig. 3a and Fig. 3b, respectively. When the field is turned on, the NWs rotate simultaneously due to the induced magnetic torque (\u03c4m), which is given by: \u03b8\u03c4 sinBMVVm =\u00d7= BM (1) where M is the magnetic moment, B is the magnetic field vector, V and M are the volume and the spontaneous magnetization of the Ni NW, respectively, and \u03b8 is the angle between the magnetization vector M and B [14-15]. During experiments, the NWs rotate either in the horizontal plane and remain simple rotation (see Fig. 3a), resembling remotely controlled microrotors; or rotate in a vertical plane and are propelled along the X-Y plane (see Fig. 3b). The steering of a NW swimmer was conducted by updating the orientation of the applied magnetic field, which is indicated by the vector A , the normal of the plane of the rotating field B (see Fig. 3). When the orientation of the vector A is changed, the NW attempts to align its own plane of rotation with the plane of rotation of the input field [14]. When the NW rotates vertically with respect to the X-Y plane, the steering is conducted by tuning the yaw angle \u201c\u03b1\u201d (see Fig. 3b). In the user interface, pitch angle is applied to adjust the angle of the vector A with respect to the horizontal X-Y plane. Thus, the NW swimmers were controlled by four parameters: magnetic field strength, input frequency, yaw angle and pitch angle, in which two pitch angles (0\u00b0 and 90\u00b0) were used for the tests. C. Assembly of a Ni NW with PS microparticles For colloidal cargo transport near a solid surface, a PS microparticle can be assembled onto one end of the tumbling NW by steering the NW towards the particle at a sufficiently high rotational speed" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002586_iccas.2013.6704206-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002586_iccas.2013.6704206-Figure6-1.png", + "caption": "Fig. 6 Drawing of Manipulator (left), Pan-tilt camera module (right).", + "texts": [ + ", controls the integrated system including the mobile platform and the manipulator. Windows-based and network supported UI software was developed in Fig. 5. A pan-tilt camera and CCD cameras are installed with surveillance purpose at middle, front and gripper-end on the platform and manipulator, respectively. Also MIDER-3 equipped CVT (Continuously Variable Transmission) for controlling suitable speed and torque in mobile platform. Performance specifications of the mobile platform are shown in Table 1. The manipulator system of MIDERS-3 is depicted in Fig. 6 and consists of three modules; a manipulator, a pan-tilt camera module, a gripper. The manipulator and pan-tilt module possess five and one degrees of freedom, respectively. The pan-tilt module attached at the mobile base. MIDERS-3 can be operated for EOD and surveillance missions by equipped a gripper at the end-effector. For the mine detection the MD and GPR sensor module is grasped by the gripper. Equipped a gripper have 2\ufffd5 km/g grasp force and is shown in Fig. 7. For accurate gripping, two laser pointers are attached" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002570_ever.2013.6521586-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002570_ever.2013.6521586-Figure2-1.png", + "caption": "Fig. 2. Distribution of the magnetic vector potential A.", + "texts": [ + " Rotor fault is modeled as 3 broken adjacent rotor bars with symmetrical power supply. Figure 1 shows the current density distribution. There is no current flow in the broken rotor bars and the highest current density is in adjacent bars which causes a local increase of temperature. Presentation of this effect points the most common localization of the rotor bar fault and explains why the adjacent bars are usually damaged. Deformation of the electromagnetic field resulting from the asymmetry is visible in the distribution of the magnetic vector potential (Fig. 2). The penetration of the flux to the damaged bars is clearly visible. In this case the mutual inductance is decreased while the leakage inductance is increased. Distribution of magnetic induction corresponding to the same time of the simulation is presented in Figure 3. After calculation of a series of FEM simulations three broken rotor bars were selected, since the electromagnetic field deformation in this case is visible globally. Field analysis provides studying the effects of the damaged rotor bars to the equivalent circuit parameters of the machine" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002000_scis-isis.2014.7044727-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002000_scis-isis.2014.7044727-Figure5-1.png", + "caption": "Fig. 5 Arrangement of the depth camera", + "texts": [ + " ( ) ( )= = = areaotherP areaobjectbumpP I ji ji jibin ),( ),( ),( 1 /0 (3) Even if the area has been classified as a motorable area, depending on the magnitude of slope angle, the mobile robot cannot always move safely. To discriminate impassible areas from safe navigation areas, we estimate the magnitude of the slope angle and width by applying the plane approximation method by principal component analysis. 978-1-4799-5955-6/14/$31.00 \u00a92014 IEEE 566 areaobstacleisslopeelse areafreecollisionisslopethenif max\u03b8\u03b8 > (4) Note that the plane approximation must apply only to the motorable area data. To distinguish the flat slope and ground plane areas, we arrange the depth camera as illustrated in Fig. 5. To prevent the observation of negligible ups and downs on the ground surface, the height of the depth camera is set to h. IV. EXPERIMENTS To confirm the validity of our proposed algorithm, we conducted outdoor experiments. Detailed Experimental setup is summarized in Table.1. A. Various area segmentation by applying proposed method Fig. 6 shows the experimental environment that includes slope, step, and wall. In order to prevent false detection of obstacles due to the strong sunlight effect, we carried out experiments at night time" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure3.41-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure3.41-1.png", + "caption": "Fig. 3.41 Screw offset. Lambda plate", + "texts": [ + " In accordance to the displacement analysis, it can be observed that the lambda plate serves as a joint that allows the homogeneous movement of the mandibular structure. This can be verified looking at the displacement, a displacement close to 3 Biomechanical Evaluation of Sharped Fractures \u2026 109 Fig. 3.39 Assignment of forces for simulation 2 Table 3.26 Results of simulations with muscle forces applied to the jaw Case of study Lambda plate Champy plate Von Mises stress (MPa) 451.897 675.076 Displacement (mm) 0.57504 1.61328 Maximum shear stress (MPa) 393.380 432.815 110 J. A. Beltr\u00e1n-Fern\u00e1ndez et al. 0 mm (Fig. 3.41), in contrast to the Champy plate which has a greater displacement in the chin area of up to 10 times the displacement in the condylar area. It is also important to emphasize that the lambda plate had a significant displacement in the last screw of the system (Fig. 3.42), which represents that this element will have the tendency to move from its site. Such behavior is common in the treatment of fractures by means of osteosynthesis with plates, but it is equally of the utmost importance to anticipate such situations to avoid future complications for the patient" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002868_s12206-015-0143-9-Figure23-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002868_s12206-015-0143-9-Figure23-1.png", + "caption": "Fig. 23. Schematics and notations for PN guidance.", + "texts": [ + " (3) For an aircraft traveling with a velocity vector of V and a line of sight (LOS) angle \u03bb to a reference point, CV is the closing velocity and Na is the control force normal to the CV vector. Na is proportional to the rate of change of LOS and the closing speed using the proportional gain value of K. Fig. 22 shows the schematic diagram for PN guidance. In an actual application, PN guidance is applied as: N G cmd N CK l= &V V , (4) ( )1 /N cmd cmd ctany l -= + V V , (5) where Na is altered by N cmdV , the velocity command normal to the closing velocity. The altered schematic and notations are shown in Fig. 23. The LOS is calculated as the heading angle referenced to the northerly direction, and the heading command cmdy is tilted by the arctangent of /N cmd cV V . If the aircraft\u2019s flight path is a series of waypoints, and if the direction of the course is changed by a certain amount of angle, the PN guidance may create an abrupt change in heading angle and side slip, which results in abrupt roll rate changes. Therefore, circular path guidance is used during a change in path direction - see Fig. 24. During waypoints #1 and #2, the aircraft was flying using PN guidance" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003224_pesgm.2014.6939394-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003224_pesgm.2014.6939394-Figure3-1.png", + "caption": "Fig. 3. Wind Turbine Drive Train [15] Similarly, the torque that produced by the high-speed shaft T accelerates the rotor of the synchronous generator and is balanced with the electromagnetic torque T produced by the generator. The relation between the electromagnetic torque T", + "texts": [ + " The mechanical torque of the prime mover T can be defined from the next relation: T = T + T (1) where T is the acceleration torque which can be determined as follows: T = J d\u03c9 dt (2) where J is the combined generator rotor and wind turbine inertia coefficient in steady-state at a fixed speed, and is the change in the rotational angular speed of the high speed shaft per time which is equal to the change in the rotational angular speed of the generator rotor shaft per time [8, and 15]. The mechanical torque T that produced by the wind accelerates the wind turbine and is counterbalanced with the torque of the low speed side shaft T (the torque produced by the torsional movement of the low speed side shaft). From Figure 3, the relation between T and T as follows: T \u2212 T = J , d\u03c9 dt (3) where \u03c9 is the rotational angular speed of the low speed shaft, and J , is the total moment of inertia for both the blades and hub of the wind turbine!kg. m \" which can be calculated as follows: J , = J + J (4) J = 3 12 [ % + & + '() * + 3 +, ' ] (5) -. = +. . D0 /8 (6) where J is the turbine\u2019s blades moment of inertia, J is the turbine\u2019 hub moment of inertia, % is the blades measured length, & is the average width the blades, * is the blade angle, ' is the center of mass displacement of the blades, +" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003715_robio.2014.7090476-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003715_robio.2014.7090476-Figure3-1.png", + "caption": "Fig. 3. Spherical center calibration", + "texts": [ + " Planning with the offline programming system requires that the coordinates of the spherical workpiece center in the robotic coordinate frame is determined accurately, so that the coordinates of the trajectory points can be converted to the coordinates of TCP. It is impossible to obtain these coordinates accurately by measurement with a ruler, and calibration of the center coordinates is therefore necessary. A center point calibration method is proposed, which needs only one gradienter and a series of calculation [11]. As shown in Fig. 3(a), the gradienter is first put on the supporting frame of the sphere. The positioner orientation is adjusted so that its table and the supporting frame of the sphere are horizontal (indicated by the gradienter). Let the vertical arm (indicated 90o by the protractor) of the gradienter contact the sphere, and mark the contact point (this point and the center point are in the same horizontal plane). Then drive the robot with the teach pendant to let the TCP touch the mark, as shown in Fig.3(b), record the configuration (joint angles) of the robot. Next, let the table of the positioner and the sphere rotate about their vertical axis by five degrees, and let the TCP touch the mark again. Repeat this procedure until the positioner has rotated by \u00b1120\u25e6 (the range is limited due to the workspace of the robot), and record all the 48 configurations (sets of joint angles). According to above configurations, the corresponding coordinates (x, y, z) of the TCP, in the base coordinate frame of the robot, can be calculated using forward kinematics of the robot, and the mean of the coordinate z, z, can be obtained" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001956_1548512911407647-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001956_1548512911407647-Figure1-1.png", + "caption": "Figure 1. Two-part missile model.", + "texts": [ + " Taking a moving surface target model into account, the performance characteristics of the mentioned types of missiles are examined. In this study, both single- and two-part air-to-surface missile configurations are considered. Once the governing differential equations of motion for a more complicated two-part missile have been derived, they can be easily adapted to single-part ones. Here, an aerodynamically controlled canard-type two-part missile configuration is considered as schematically represented in Figure 1. The mentioned missile model is a combination of two relatively rotating parts that are connected to each other by means of a roller bearing. Here, ui b( ) (i = 1, 2 and 3) represents the unit vectors of the body-fixed frame of the missile (Fb). C1, C2 and CM denote the mass centres of the front body, or body 1, rear body, or body 2, and entire missile body, respectively, with definitions that xM and d12 indicate the distances from C1 to CM and from C1 to C2. As shown, dj (j = 1, 2, 3 and 4) represents the fin deflections of the aerodynamic control surfaces, that is, the canards" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001652_978-3-319-10891-9_10-Figure10.6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001652_978-3-319-10891-9_10-Figure10.6-1.png", + "caption": "Fig. 10.6 Furuta pendulum", + "texts": [ + " Experiments show that the proposed algorithm manages to produce results resembling the ideal closed loop behavior. If the proposed algorithm is not used, the behavior is altered; for instance\u2014for the given example\u2014the steady state error is not zero. The PBC approach has been experimented with on a real plant. These experiments assess the real-world applicability of the approach. A comparison between netwoked control and local control is shown. The test-bed is the Furuta pendulum is represented in Fig. 10.6 and its parameters are contained in Table 10.2. The vector q = [q1, q2]T describes the vector of the state variables: q1 is the angular position of the arm and q2 is the angular position of the pendulum. The system is under-actuated, meaning that only the arm joint is actuated by means of the torque \u03c4 . The dynamics of the nonlinear model for the plant is given by: Table 10.2 Parameters of Furuta pendulum Physical quantity Symbol Value Units Arm mass m1 200 \u00d7 10\u22123 kg Pendulum mass m2 72 \u00d7 10\u22123 kg Arm length L1 224 \u00d7 10\u22123 m Arm COM l1 144 \u00d7 10\u22123 m Pendulum COM l2 106 \u00d7 10\u22123 m Arm z0 inertia Jz0 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003090_ecce.2013.6647199-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003090_ecce.2013.6647199-Figure2-1.png", + "caption": "Fig. 2. Experimental setup.", + "texts": [ + " 6) Decide the command-tracking performance by selecting \u03c9l and \u03b6l and construct the prefilter matrices Af , Bf , Cf , and Df using (15). 7) Check the stability of the system, e.g., applying the Nyquist diagram for the loop gain of the system. More information about the stability analysis is provided in Section IV-G. 8) Discretize the continuous-time subsystems (3), (8), (11), and (15), e.g., by applying Tustin\u2019s method [30]. Simulation parameters, considered in the design example, are based on the experimental setup shown in Fig. 2. The setup consists of two 4-kW 2400-rpm servo motors coupled together with a toothed belt. In order to vary the coupling stiffness, different belts can be used. An additional inertia disk can be added to the shaft of the load motor. The torque control is accomplished using a field-oriented control (FOC) and when the system is operating in linear region, the closed torque-control loop can be modeled as a transfer function Gt(s) = e\u2212sTd \u03b1t s+ \u03b1t (16) where \u03b1t = 1.8 krad/s is the bandwidth and Td = 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003026_amm.401-403.254-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003026_amm.401-403.254-Figure2-1.png", + "caption": "Figure 2 The geometric relationship of sphere contact with each other", + "texts": [ + " 1: K\u2014load - displacement coefficient Under the action of load, the normal contact deformation amount which between two raceways separated by rolling body is equal to the sum of contact deformation of rolling body to each raceway. Therefore, oin \u03b4\u03b4\u03b4 += (2) In Eq. 2: \u03b4i, \u03b4o\u2014The inner ring and the outer ring contact deformation 2 3 3 2 3 2 )/1()/1( 1 + = oi n KK K (3) In Eq. 3: Ki, Ko\u2014The load - displacement coefficient of the inner ring and the outer ring The geometric relationship of sphere which contact with each other as shown in Figure 2. In the figure, r11, r12, r21, r22\u2014is the radius of objects 1 and 2 in two orthogonal planes For the steel roller and raceway contact relation is: 2 3 *2 1 , 5 , )()(1015.2 , \u2212\u2212 \u2211\u00d7= oioioiK \u03b4\u03c1 (4) In Eq. 4: \u03a3\u03c1i, \u03a3\u03c1o\u2014The curvature of The inner ring and the outer ring. Where * ,oi \u03b4 \u2014The contact deformative coefficient of the inner ring and the outer ring. 22211211 1111 rrrr +++=\u2211\u03c1 (5) \u2211 \u2212+\u2212 = \u03c1 \u03c1 22211211 1111 )( rrrr F (6) The contact deformation coefficients \u03b4 * related with curvature difference F(\u03c1)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.35-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.35-1.png", + "caption": "Fig. 2.35. Cylindrical arm", + "texts": [ + " Prove that the unit quaternion corresponding to a rotation matrix is given by (2.34), (2.35). 2.9. Prove that the quaternion product is expressed by (2.37). 2.10. By applying the rules for inverting a block-partitioned matrix, prove that matrix A1 0 is given by (2.45). 2.11. Find the direct kinematics equation of the four-link closed-chain planar arm in Fig. 2.34, where the two links connected by the prismatic joint are orthogonal to each other. 2.12. Find the direct kinematics equation for the cylindrical arm in Fig. 2.35. 2.13. Find the direct kinematics equation for the SCARA manipulator in Fig. 2.36. 2.14. Find the complete direct kinematics equation for the humanoid manipulator in Fig. 2.28. 2.15. For the set of minimal representations of orientation \u03c6, define the sum operation in terms of the composition of rotations. By means of an example, show that the commutative property does not hold for that operation. 2.16. Consider the elementary rotations about coordinate axes given by infinitesimal angles. Show that the rotation resulting from any two elementary rotations does not depend on the order of rotations", + " Further, define R(d\u03c6x, d\u03c6y, d\u03c6z) = Rx(d\u03c6x)Ry(d\u03c6y)Rz(d\u03c6z); show that R(d\u03c6x, d\u03c6y, d\u03c6z)R(d\u03c6\u2032 x, d\u03c6 \u2032 y, d\u03c6 \u2032 z) = R(d\u03c6x + d\u03c6\u2032 x, d\u03c6y + d\u03c6\u2032 y, d\u03c6z + d\u03c6\u2032 z). 2.17. Draw the workspace of the three-link planar arm in Fig. 2.20 with the data: a1 = 0.5 a2 = 0.3 a3 = 0.2 \u2212\u03c0/3 \u2264 q1 \u2264 \u03c0/3 \u2212 2\u03c0/3 \u2264 q2 \u2264 2\u03c0/3 \u2212 \u03c0/2 \u2264 q3 \u2264 \u03c0/2. 2.18. With reference to the inverse kinematics of the anthropomorphic arm in Sect. 2.12.4, discuss the number of solutions in the singular cases of s3 = 0 and pWx = pWy = 0. 2.19. Solve the inverse kinematics for the cylindrical arm in Fig. 2.35. 2.20. Solve the inverse kinematics for the SCARA manipulator in Fig. 2.36." + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001960_012020-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001960_012020-Figure2-1.png", + "caption": "Figure 2. The free body diagrams for disks 2 and 3 of the four-to include a model of a gearbox inertia idealisation of a drive system.", + "texts": [ + " The four equations of motion of this simple system, assuming viscous damping in the shaft, are I1\u03d5\u03081 + c1 ( \u03d5\u03071 \u2212 \u03d5\u03072 ) + k1 (\u03d51 \u2212 \u03d52) = T1 I2\u03d5\u03082 + c1 ( \u03d5\u03072 \u2212 \u03d5\u03071 ) + k1 (\u03d52 \u2212 \u03d51) = T2 + R2F23 I3\u03d5\u03083 + c2 ( \u03d5\u03073 \u2212 \u03d5\u03074 ) + k2 (\u03d53 \u2212 \u03d54) = T3 + R3F23 I4\u03d5\u03084 + c2 ( \u03d5\u03074 \u2212 \u03d5\u03073 ) + k2 (\u03d54 \u2212 \u03d53) = T4 (1) where F23 represents the instantaneous force acting between gears 2 and 3 along the common tangent and where {T1, T2, T3, T4} are the instantaneous torques being exerted on inertias {1, 2, 3, 4} respectively from external sources. It appears anomalous at first that F23 contributes positively to both gear wheels but inspection of the free body diagrams for disks 2 and 3, shown in Figure 2, reveals that this is correct. There are four rotation co-ordinates in Equation (1) but only three of these are independent because of the constraint between \u03d52 and \u03d53. We choose {\u03d51, \u03d52, \u03d54} as independent coordinates, and \u03d53 is eliminated using \u03d53 = \u2212\u03b3\u03d52. The forcing term F23 in Equation (1) is unknown, and this is removed by subtracting \u03b3 times the third equation from the second. The resulting three equations in {\u03d51, \u03d52, \u03d54} are I1\u03d5\u03081 + c1 ( \u03d5\u03071 \u2212 \u03d5\u03072 ) + k1 (\u03d51 \u2212 \u03d52) = T1( I2 + \u03b32I3 ) \u03d5\u03082 + c1 ( \u03d5\u03072 \u2212 \u03d5\u03071 ) + c2 ( \u03b32\u03d5\u03072 + \u03b3\u03d5\u03074 ) +k1 (\u03d52 \u2212 \u03d51) + k2 ( \u03b32\u03d52 + \u03b3\u03d54 ) = T2 \u2212 \u03b3T3 I4\u03d5\u03084 + c2 ( \u03d5\u03074 + \u03b3\u03d5\u03072 ) + k2 (\u03d54 + \u03b3\u03d52) = T4 (2) These equations are equivalent to those obtained from a three-inertia model of a simple rotor but comprising only three inertias" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003942_s003602441309032x-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003942_s003602441309032x-Figure7-1.png", + "caption": "Fig. 7. Dependence of temperature on reaction rate in the absence (1) and presence of surfactant (2).", + "texts": [ + " The Gibb\u2019s energy of micellization ( = 13.68 kJ mol\u20131), [53] and The Gibb\u2019s energy of adsorption ( = 13.88 kJ mol\u20131) [54] can be cal culated by using the following equations: (18) (19) To study the effect of temperature on kobs, kinetics run was carried out at pH 3 in the absence and pres ence of octadecylamine by varying the temperature in the range of 15 to 35\u00b0C. The residual conditions were kept unaltered. Various thermodynamic functions were calculated by applying different kinetic equations (Tables 3, 4 and Fig. 7). Value of activation energy (\u0394Ea) was calculated from the slope when Arrhenius equation i.e., kobs = lnA \u2013 Ea/RT, was plotted between \u0393max 1 RT \u0394\u03b3 \u0394 cln T ,\u2013= Amin \u0393max/NA,= a 1 10 20 /\u0393maxNA.\u00d7= \u03a0max \u03b30 \u03b3CMC,\u2013= \u0394GM\u00b0\u2013 \u0394Gad\u00b0\u2013 \u0394GM\u00b0 RT CMC[ ],ln= \u0394Gad\u00b0 \u0394GM\u00b0 \u03a0max/\u0393max.\u2013= lnkobs vs. 1/T (Fig. 8). The Erying equation i.e., ln(kobs/T) = ln + \u2013 \u2013 , was used to determine the enthalpy and entropy of activation. The enthalpy of activation (\u0394H#) was determined with the help of a plot lnkobs/T vs. 1/T (Fig. 9); \u0394H# was calculated from the slope of graph" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure3.42-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure3.42-1.png", + "caption": "Fig. 3.42 Displacement and maximum von Mises stress on the Champy plate", + "texts": [ + "26 Results of simulations with muscle forces applied to the jaw Case of study Lambda plate Champy plate Von Mises stress (MPa) 451.897 675.076 Displacement (mm) 0.57504 1.61328 Maximum shear stress (MPa) 393.380 432.815 110 J. A. Beltr\u00e1n-Fern\u00e1ndez et al. 0 mm (Fig. 3.41), in contrast to the Champy plate which has a greater displacement in the chin area of up to 10 times the displacement in the condylar area. It is also important to emphasize that the lambda plate had a significant displacement in the last screw of the system (Fig. 3.42), which represents that this element will have the tendency to move from its site. Such behavior is common in the treatment of fractures by means of osteosynthesis with plates, but it is equally of the utmost importance to anticipate such situations to avoid future complications for the patient. It is important to mention that the simulation is a static test, and it represents that the displacement presented by the screw should not be expected to occur in this way unless the jaw (in question), receives a firm blow/hit similarly to that of the simulation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002586_iccas.2013.6704206-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002586_iccas.2013.6704206-Figure4-1.png", + "caption": "Fig. 4 MIDERS-3: Drawing of MIDERS-3 (left), Integrated system (right).", + "texts": [ + " This chapter demonstrates the hardware composItIOn, specifications, and functions for hardware architectures of MIDERS-3. MIDERS-I and MIDERS-2 system can be confirmed our previous paper [13]. MIDERS-3 is developed for precise scan of a mine. MIDERS-3 follows MIDERS-I likewise MIDERS-2. Once MIDERS-I marks the location of a mine with paint spray and goes further for fast scan, the location is transmitted to MIDERS-3 and it is approached to the marked area to perform precise scan of a mine. The mobile platform, MIDERS-3, that we developed is shown in Fig. 4. The rubber track type mobile platform is designed for the rough terrain operations. Since it should be able to navigate on the rough terrain and maintain it stability for detecting mine with more detailed manner, the mobile platform become smaller than other MIDERS platform. Minimized size also helps more precise control of the mobile platform. In order to obtain more stable and accurate positioning along the path with many obstacles, two sets of flippers are implemented as in Fig. 4. It can proceed forward with maximum speed at 4kmlh. A Single Board Rio as a control PC, the Product of the National Instrument INC., controls the integrated system including the mobile platform and the manipulator. Windows-based and network supported UI software was developed in Fig. 5. A pan-tilt camera and CCD cameras are installed with surveillance purpose at middle, front and gripper-end on the platform and manipulator, respectively. Also MIDER-3 equipped CVT (Continuously Variable Transmission) for controlling suitable speed and torque in mobile platform" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.21-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.21-1.png", + "caption": "Fig. 2.21. Parallelogram arm", + "texts": [ + " Obviously, pz = 0 and all three joints concur to determine the end-effector position in the plane of the structure. It is worth pointing out that Frame 3 does not coincide with the end-effector frame (Fig. 2.13), since the resulting approach unit vector is aligned with x0 3 and not with z0 3. Thus, assuming that the two frames have the same origin, the constant transformation T 3 e = \u23a1 \u23a2\u23a3 0 0 1 0 0 1 0 0 \u22121 0 0 0 0 0 0 1 \u23a4 \u23a5\u23a6 . is needed, having taken n aligned with z0. Consider the parallelogram arm in Fig. 2.21. A closed chain occurs where the first two joints connect Link 1\u2032 and Link 1\u2032\u2032 to Link 0, respectively. Joint 4 was selected as the cut joint, and the link frames have been established accordingly. The DH parameters are specified in Table 2.2, where a1\u2032 = a3\u2032 and a2\u2032 = a1\u2032\u2032 in view of the parallelogram structure. Notice that the parameters for Link 4 are all constant. Since the joints are revolute, the homogeneous transformation matrix defined in (2.52) has the same structure for each joint, i.e" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure9.3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure9.3-1.png", + "caption": "Fig. 9.3 Schematic representations of the adult intervertebral disc: a midsagittal cross section showing anatomical regions. b Three-dimensional view illustrating AF lamellar structure [5]. Reprinted with permission from SpringerNature publishers", + "texts": [ + " Each vertebra consists of two parts, the vertebral body and the vertebral arch, which also enclose the vertebral foramen that contains the spinal cord. Between the vertebrae, fibrocartilaginous joints called intervertebral discs can be found. These hydraulic systems separate the vertebrae from each other and provide optimal cushioning protection against repeated impacts, allowing the performance of the natural movements of each functional unit. An intervertebral disc is also formed of two parts (as shown in Fig. 9.3): (a) an outer fibrous ring formed by several layers that cross one another, forming a mesh-like structure. The fibrous layers unite the vertebrae and protect the nucleous. (b) The central pulposus structure called the nucleous, composed of 80% of water, distributes the hydraulic pressures in all directions within each intervertebral disc under compressive loads. The combination of two vertebrae, together with the intervertebral disc is known as the vertebral column\u2019s functional unit. A typical vertebra consists of two parts: the vertebral body and the vertebral arch" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002757_imece2011-63090-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002757_imece2011-63090-Figure4-1.png", + "caption": "Figure 4. The Full Vehicle Model (12-DOF) [7]", + "texts": [ + " 5 1 6 2 7 3 8 41 2 3 4 05 1 6 2 7 3 8 41 2 3 4 M Z C Z Z C Z Z C Z Z C Z Zb s k Z Z k Z Z k Z Z k Z Z (7) 5 1 6 2 7 3 8 41 2 3 4 05 1 6 2 7 3 8 41 2 3 4 I C Z Z a C Z Z b C Z Z b C Z Z ayy b k Z Z a k Z Z b k Z Z b k Z Z a (8) 5 1 6 2 7 3 8 41 2 3 4 05 1 6 2 7 3 8 41 2 3 4 I C Z Z c C Z Z c C Z Z c C Z Z cxx b k Z Z c k Z Z c k Z Z c k Z Z c (9) 05 1 1 1 5 1 1 11 1 1 1m Z C Z Z C Z q k Z Z k Z qt t (10) 06 2 2 2 6 2 2 22 2 2 2m Z C Z Z C Z q k Z Z k Z qt t (11) 07 3 3 3 7 3 3 33 3 3 3m Z C Z Z C Z q k Z Z k Z qt t (12) 08 4 4 4 8 4 4 44 4 4 4m Z C Z Z C Z q k Z Z k Z qt t (13) 5Z Z c ab (14) 6Z Z c bb (15) 7Z Z c bb (16) 8Z Z c ab (17) Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/imece2011/70883/ on 02/08/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use 4 Copyright \u00a9 2011 by ASME 2.4 The Full Vehicle Model (12-DOF) The vehicle body is assumed to be rigid, with mass ms and moments of inertia. The rigid body has six degrees of freedom as shown in Fig. 4, which includes three translations namely; forward velocity U in xL -direction, lateral velocity V in yL - direction and vertical velocity W in zL -direction, in addition to three rotations namely; roll rate p about xL -axis, pitch rate q about yL - axis and yaw rate r about zL - axis. The wheels are connected to the vehicle body via springs and shock absorbers. It is assumed that each wheel has two degrees of freedom, one for the vertical displacement zwi , and the other for wheel rotational driving speed i , the equations of motion for the lumped mass can be derived as follow [7]: F m U Vr Wqx s (18) F m V Wp Ury s (19) F m W Uq Vpz s (20) M I p I I qr I pq rx xx yy zz zx (21) 2 2M I q I I rp I p ry yy zz xx xz (22) M I r I I pq I rq pz zz xx yy zx (23) 1 : 4 m Z m g C Z Z k Z Z Fbi wi bi wii wi wi i i zi where i (24) , 1:2I M M M where iwi i wi Bi Ui (25) 3" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001077_978-1-4419-7979-7_6-Figure6.45-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001077_978-1-4419-7979-7_6-Figure6.45-1.png", + "caption": "Fig. 6.45 Three-voltmeter method for measuring the loss of an inductor and corresponding phasor diagram", + "texts": [ + "5 Measurement of Low Losses of High-Efficiency Inductors 269 Three-Voltmeter Method for Power Measurement: The losses for some inductors are very small; nevertheless, one has to raise the question about the accuracy of the measured results obtained from the computer-aided measurement circuit of Fig. 6.44. For this reason an alternative approach is chosen and the results of both measurement approaches are compared. The three-voltmeter method is based on (sinusoidal) voltage measurements, as shown in the circuit of Fig. 6.45. The maximum value of the phase angle y is 90 , which corresponds to an ideal lossless inductor. The value of resistor R1 must be known in order to compute the loss Pvlossand AC resistance Rs of the inductor. The three-voltmeter method can be implemented together with the computer-aided measurement circuit of Fig. 6.44 if the shunt resistance Rsh of Fig. 6.44 is used as resistance R1 of Fig. 6.45. The loss is then computed from Pvloss \u00bc I2Rs \u00bc \u00f0V2 1 V2 2 V2 3\u00de 2R1 ; (6.170) where V1,V2, and V3 are the rms values of ~V1;~V2, and ~V3, respectively, and I is the rms value of the sinusoidal current ~I. The derivation of the formulae for the three-voltmeter, and three-ampere meter methods will be addressed in Application Example 6.16. (a) Show that for the three-voltmeter method (6.170) is valid. (b) How can we use three current meters (in analogy to the three-voltmeter method) for the measurement of the power dissipated in an inductor? You may assume sinusoidal currents. Devise a circuit diagram similar to that of Fig. 6.45 employing three ampere meters, and derive a similar equation to (6.170) for expressing the measured real power as a function of ~I1 , ~I2 , and ~I3 (a) There are two methods to show that (6.170) is true: method #1 is based on the phasor diagram and method #2 on Ohm\u2019s law. 270 6 Magnetic Circuits: Inductors and Permanent Magnets Method #1: Figure 6.46 shows the definitions of the 3 voltmeters, inductor resistance RL and reactance XL, and calibrated resistor with resistance R. The phasor diagram for circuit Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003277_icems.2011.6073426-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003277_icems.2011.6073426-Figure1-1.png", + "caption": "Fig. 1. Cross-sectional geometry of the 37-kW cage induction motor. The flux shown is associated with the locked rotor condition.", + "texts": [ + " The motor was supplied from a 750-kVA synchronous generator and loaded by an 80-kW dc machine controlled by a thyristor bridge. A 1000-Nm torque transducer set a torque limit that forced us to reduce the supply voltage of the motor to 320 V from the rated 380 V. To control the temperature rise, the measurements were done a torque-speed point by a torque-speed point. One measurement took about 10 s and after this the machine was run at no load long enough to reduce the stator end winding temperature back to 80 \u02daC. The cross-sectional geometry of the 37-kW cage induction motor studied is shown in Fig. 1. Its main parameters are given in Table I. The machine that was measured and simulated has a non-skewed rotor. A. Validation of the numerical method of analysis Fig. 2 shows the measured and computed torque versus speed curve of the 37-kW machine. The measurements and simulations were done at a reduced line-to-line voltage of 320 V. The synchronous torque dip at 150 rpm predicted by the analytical model is clearly present. The FEA predicts the magnitude of the torque dip with good accuracy. There is a smaller synchronous torque at \u2013300 rpm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003806_isie.2013.6563680-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003806_isie.2013.6563680-Figure6-1.png", + "caption": "Fig. 6 Circuit and the current flowing path under", + "texts": [ + " The total voltage drop dropV across the buck inductor BL and the line- to-line inductance sL2 and line-to-line resistance sR2 are the difference between the diode voltage oiV and the inducted voltage se . Thus, the transfer function from the voltage drop dropV to the yielded bus current dI can be expressed as ssBdrop d RsLLsV sI 2)2( 1 )( )( (6) When both 1T and 2T are conducting (i.e. 3/0 e ), the bus current dI flowing through 1T and 2T . It follows that cad iiI and the resulting circuit under switch BT turning on and turning off are plotted in Fig. 6(a) and Fig. 6(b), respectively. Based on KVL, the terminal voltages av and bv may be expressed as the following two equations, respectively sdca sB s sdc ainTB a RIee LL L RIe eVG v 2)( 2 2 2 (7) sdcb sB s sdc ainTB b RIee LL L RIe eVG v 22 (8) where TBG is the switching signal of controllable switch BT in the buck converter and it can be defined as offturningTswitch ontruningTswitch G B B TB 0 1 (9) (a) BT turning on, and (b) BT turning off. By neglecting the voltage drop across the equivalent resistance, the terminal voltages av and bv can be approximated to sB B c sB B a sB s inTBa LL L e LL L e LL L VGv 222 2 (10) sB sB cb sB s a sB s inTBb LL LL ee LL L e LL L VGv 222 (11) The illustrated waveforms for buck-type CSI-fed BDCM are plotted in fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001291_chicc.2015.7260301-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001291_chicc.2015.7260301-Figure1-1.png", + "caption": "Fig. 1. Model of tracking control", + "texts": [ + " Then the desired forward and angular velocities can be deduced by cos sind d d d du x y (10) ( sin ) cosd d d d dv x y (11) d dw z (12) 2 2 d d d d d d d d x y x y r x y (13) As the objective of the tracking controller is to make UUV follow the known path by controlling the velocity and angular velocities, so the tracking error x de x x , y de y y , z de z z , de converges to zero. Here T d x y ze e e e-e is the tracking error in the inertial frame. A detailed model of tracking control is given in Fig. 1. The cascaded adaptive design consists of two parts: the kinematic control part and dynamic control part. The kinematic control part is mainly based on the authors\u2019 former work [14] while the dynamic part is based on adaptive design. The ocean current is added in the control design and the block diagram is illustrated in Fig. 2. As roll and pitch are usually the unwanted motion actions in the full degrees, so in this paper, only a 4-DOF (surge, sway, heave, and yaw) tracking control problem is represented" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002281_12.2062006-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002281_12.2062006-Figure6-1.png", + "caption": "Figure 6. (a) EWD is after the afocal micro lens array. (b) EWD is inserted in the middle of afocal micro lens array.", + "texts": [ + "138 mm of simulation, therefore about 1% error between the measurement and the simulation. The interval of sinc function are 2.08 mm of measurement and 2.02 mm of simulation, therefore about 2.9% error between the measurement and the simulation. In order to expand the aperture of EWD pixels for reducing diffraction effect, the micro leans array is adopt to avoid that the light is blocked by the outer frame of EDW pixel. As Fig. 5 showing, the micro lens array avoid the light touches the outer frame of EWD pixels. There are two method to add the micro lens array on EWD as Fig. 6 showing. One is that EWD is after the afocal micro lens array as Fig. 6(a) showing, another is that EWD is inserted in the middle of afocal micro lens array as Fig. 6(b) showing. We defined a definition, decreasing percentage of width, to evaluate the diffraction improvement. The decreasing percentage of width is equal to 1-(improved width/original width), where width is half diffraction width as Fig. 7 showing. The light distribution after EWD 30 cm, 50cm, and 1m are simulate with 3 condition as Fig. 8 showing. The light distribution of simulation without micro lens array at 30 cm, 50cm, and 1m are showing in Fig. 8 (a), (b), and (c). The diffraction width of Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.1-1.png", + "caption": "Fig. 2.1. Position and orientation of a rigid body", + "texts": [ + " With reference to a minimal representation of orientation, the concept of operational space is introduced and its relationship with the joint space is established. Furthermore, a calibration technique of the manipulator kinematic parameters is presented. The chapter ends with the derivation of solutions to the inverse kinematics problem, which consists of the determination of the joint variables corresponding to a given end-effector pose. A rigid body is completely described in space by its position and orientation (in brief pose) with respect to a reference frame. As shown in Fig. 2.1, let O\u2013xyz be the orthonormal reference frame and x, y, z be the unit vectors of the frame axes. The position of a point O\u2032 on the rigid body with respect to the coordinate frame O\u2013xyz is expressed by the relation o\u2032 = o\u2032xx+ o\u2032yy + o\u2032zz, where o\u2032x, o \u2032 y, o \u2032 z denote the components of the vector o\u2032 \u2208 IR3 along the frame axes; the position of O\u2032 can be compactly written as the (3 \u00d7 1) vector o\u2032 = \u23a1 \u23a3 o\u2032x o\u2032y o\u2032z \u23a4 \u23a6 . (2.1) Vector o\u2032 is a bound vector since its line of application and point of application are both prescribed, in addition to its direction and norm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003959_s105261881101002x-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003959_s105261881101002x-Figure4-1.png", + "caption": "Fig. 4. Mutual position of the carriage and the balancing weight.", + "texts": [], + "surrounding_texts": [ + "104\nJOURNAL OF MACHINERY MANUFACTURE AND RELIABILITY Vol. 40 No. 2 2011\nALESHIN\nLet the unknown position vector of the displacement be R1 in the coordinates W\u03be1\u03b71, and the unknown mass of the body be M.\nThe measurement system for the time intervals (Fig. 2) is formed by photoelectric detector 11 with optical axis 12 and the coordinate axes \u03be1 and \u03b71 modulating the optical path; these axes are plotted on an optically transparent disk 13. The disk is rigidly bound to the platform. Rotation of the platform with the frequency \u03a9 causes a centrifugal force F1: F1 = \u03a92MR1. The measurement process consists of three con secutive steps.\nStep 1. Subject to the force F1, the platform with the base (Fig. 2) are displaced by the position vector r1. The phase angle \u03b3 of the lag of the vector r1 from R1 is always constant, provided that \u03a9 and M do not change. The value and position of the radius r1 with the polar angle \u03b21 are determined from three time intervals in the frame W\u03be1\u03b71 by the above algorithm.", + "JOURNAL OF MACHINERY MANUFACTURE AND RELIABILITY Vol. 40 No. 2 2011\nA METHOD TO DETERMINE THE MASS AND COORDINATES 105\nStep 2. Carriage 8 with fixed body 7 shifts for an arbitrary yet known position vector k whose direction is set by the platform\u2019s guide bearings and is the same as that of the axis \u03b71. At the same time, weight 9 travels in the opposite direction for the distance k. The center of mass of body 7 takes a new position whose radius vector R2 is R2 = R1 + k (Fig. 3). The respective centrifugal force is F2 = \u03a92MR2. It causes a dis placement of the platform by the position vector r2 with the polar angle \u03b22. These coordinates are found from the time intervals.\nStep 3. Gear wheels 10 are disconnected, and the balancing weight 9 comes back to the central posi tion. The carriage 8 with body 7 stays in the same place.\nNow the rotating platform is simultaneously affected by two centrifugal forces. The force from the unbalanced carriage is F3 = \u03a92mk. The direction and value of this force are known, because we know m and k. The second force is the unknown centrifugal force F2. Their resultant is the source of the platform displacement by the position vector r. The polar coordinates r and \u03b2 are found from the time intervals.\nThe now known values r, r1, r2, \u03b2, \u03b21, \u03b22, mass m, and the value of the vector k allow us to clearly deter mine the mass M of the body and the position vector of center of mass R1 in the frame W\u03be1\u03b71.\nTo solve this problem, we need to prove the property of the position vectors r, r1, and r2 used in the pro posed method.\nStatement. For any vector k and any mass m and M, the endpoints of the vectors r, r1, and r2 always lie on the same line, and the angle between the line and the vector k is the angle \u03b3 of the phase lag.\nProof. Transformation (1) of the unbalance vector di into the displacement vector ri by the dynamic sys tem of the bench in the coordinates \u03be1, \u03b71 can be presented as a linear transformation\n(2)\nwhere A is a finite dimensional nondegenerate linear operator given in the basis \u03be1, \u03b71. It is known that any nondegenerate operator A can be presented as a product of a nondegenerate pos itive defined operator H and unitary operator U [4]\nIn this case, the idea of vector transformation (2) is that the operator H produces a \u03bb fold change in the modulus of the vector di, and the unitary operator U is responsible for the rotation of the vector di by the angle \u03b3 of the phase lag.\nThus, in the frame W\u03be1\u03b71, the structure of the operator A is given by the matrix A, which is the product of the diagonal matrix H and orthogonal matrix U:\nAdi ri,=\nA HU.=", + "106\nJOURNAL OF MACHINERY MANUFACTURE AND RELIABILITY Vol. 40 No. 2 2011\nALESHIN\nThe vectors ab and bc are equal\n(3)\nIt follows from (3) that the vectors ab and bc are proportionate and, hence, parallel to the vector Uk. Then, they are parallel to each other yet have a common point b. Hence, they lie on the same line.\nIt also stems from (3) that this line forms the angle \u03b3 with the vector k; this is the effect of the rotation operator U upon the vector k. Another corollary of the proved statement is an equality for the segments ab, ac, and bc:\n(4)\nand a rule for finding the angle \u03b3: cos\u03b3 = (ac, )/|ac||k|. The segments ab and ac are found from the cosine theorem.\nIn the triangle Wab, the side ab = , and in the triangle Wac, the side ac =\n.\nIt follows from (3) that\n(5)\nHere, we take into account the orthogonality of the matrix U.\nHence, considering (4), we find the unknown mass M of the sample M = m(ab)/(ac \u2013 ab). It follows from equality (2) that |Ad1| = |r1|, or\n(6)\nThe value M\u03bb = (ab)/|k| can be found from equality (5). Substituting this value in (6), we get the mod ulus of the position vector |R1| of the center of mass |R1| = |r1||k|/(ab).\nIt is known that r1 is behind R1 by the angle \u03b3, but the position of r1 is known in the frame W\u03be1\u03b71. Hence, we have found the position and value of the position vector |R1| of the center of mass of the body in the plane of the platform and the body mass M.\nREFERENCES\n1. Aleshin, A.K., Chronometrical Method for Detecting the Value and Position of Rotor\u2019s Disbalance, Izv. Akad. Nauk: Mekh. Tverd. Tela, 2008, no. 2, pp. 43\u201348.\n2. Kiselev, M.I., Kozlov, A.P., Morozov, A.N., et al., The Way to Measure the Rotation Period of Turbounit\u2019s Shaft by Means of Photoelectric Method, Izmerit. Tekhn., 1996, no. 12, pp. 28\u201329.\n3. Levitskii, A.A., Losev, S.A., and Makarov, V.N., Matematicheskie metody v khimicheskoi kinetike (Mathematical Methods in Chemical Kinetics), Bykov, V.I., Ed., Novosibirsk: Nauka, 1990.\n4. Gel\u2019fand, I.M., Lektsii po lineinoi algebre (Lectures on Linear Algebra), Moscow: Nauka, 1971.\nA \u03bb 0 0 \u03bb\u239d \u23a0 \u239c \u239f \u239b \u239e \u03b3cos \u03b3sin\u2013 \u03b3sin \u03b3cos\u239d \u23a0 \u239c \u239f \u239b \u239e \u03bbU.= =\nab r2 r1\u2013 Ad2 Ad1\u2013 A d2 d1\u2013( ) A M R1 k+( ) MR1\u2013( )= = = =\n= A Mk( ) M Ak( ) M \u03bbUk( ) M\u03bb Uk( );= = =\nbc r r2\u2013 A d d2\u2013( ) A MR2 mk+( ) MR2\u2013( ) m\u03bb Uk( ).= = = =\nbc ac ab\u2013=\nk\nr1 2 r2 2 2r1r2 \u03b21 \u03b22+( )cos\u2013+\nr1 2 r 2 2r1r2 \u03b2 \u03b21+( )cos\u2013+\nab ab M\u03bb Uk( ) M\u03bb Uk M\u03bb k , bc bc m\u03bb k .= = = = = =\nM\u03bb R1 r1 .=" + ] + }, + { + "image_filename": "designv11_84_0002496_s10846-013-9971-y-Figure17-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002496_s10846-013-9971-y-Figure17-1.png", + "caption": "Fig. 17 Bounded eigenvalues of det(\u03bbI \u2212 Am(t)T Am(t))", + "texts": [ + " Figure 14 shows as the amplitude of the angular motion is reduced when the term k\u03b2 is included in the model. Figure 15 shows the amplitude of thrust. Figure 16 shows as the term k\u03b2 affects the amplitude of the longitudinal cyclic input. 6.2 Simulation of the Stability of System Am(t) Theorem 2 in the Appendix is used to verify that system (32) is uniformly exponential stable. The first condition that should be satisfied is that in \u2016Am(t)\u2016 \u2264 \u03b1 where \u03b1 is a finite positive constant. Then \u2016A(t)\u2016 = \u221a \u03bbmax(Am(t)T Am(t)). The corresponding graphics (Fig. 17) were obtained in order to observe that the eigenvalues are bounded, considering that the desired trajectories xd and yd are slow. The second condition Re[\u03bb(t)] \u2264 \u2212\u03bc is verified by means of a graphic showing the eigenvalues of Am(t); this is showed in Fig. 18. The third condition \u2016A\u0307(t)\u2016 \u2264 \u03b2 for all t is verified by means of the graphic in the Fig. 19. 6.3 Simulation of the Control Technique With/Without Stabilizer Bar The control strategy was evaluated with and without the stabilizer bar. The graphics in Figs" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003077_s40632-014-0010-3-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003077_s40632-014-0010-3-Figure1-1.png", + "caption": "Fig . 1. 12-DOF test rig model using mass stations.", + "texts": [ + " The ultimate objective of this study was to investigate the influence of bearings and supports upon the entire rotor dynamics. Numerical examples and experimental results were used to demonstrate new approach capabilities. The enhanced model with the newly identified parameters accurately predicts the dynamic responses of the entire rotorbearing system under different static loads. MATHEMATICAL MODEL The model used to simulate the entire rotor system contains 12 generalized coordinates (DOFs), two angular displacements and two radial displacements at each of the three locations shown in Figure 1: Tyxyxyx ,,,yx,,,yx,,,yxz ],,[ 333322221111 \u03b8\u03b8\u03b8\u03b8\u03b8\u03b8= (1) where i ix , y are the radial displacements and x y i i,\u03b8 \u03b8 are the angular displacements of the three locations of i=1, 2 and 3. The complete model is shown in Figure 1. However, the third disk in the middle of the shaft, where the journal bearing is located, is actually not present in the test rig. This \u201cvirtual\u201d disk is just used to obtain the shaft\u2019s stiffness matrix using linear beam theory. The shaft was subdivided into four beams that are classified into two different types of the same length. The masses were divided and concentrated at three stations. In contrast to stations 1 and 3, station 2 has no polar inertia at the disk and is just assumed to contribute to the stiffness matrix, (Figure 2)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002157_detc2013-12233-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002157_detc2013-12233-Figure6-1.png", + "caption": "Fig. 6. COORDINATE SYSTEMS FOR A LEFT-HANDED CUTTER HEAD [3].", + "texts": [ + " And in the cutting position 2, figure 5 (b) shows that the left and right flanks are convex and concave. This lengthwise modification evidently requires twice single-flank cutting processes. Therefore, the modification is only implemented on the pinion considering the machining efficiency. Moreover, the modification is applied on the finish cutting process to reduce machining time. A crowned tooth can be achieved as shown in Fig. 5 (c). A FACE-HOBBING CUTTER HEAD The mathematical model of face-hobbing cutter head has been proposed in reference [3]. As shown in Fig. 6, the cutter head has 0z blade groups. Each blade group has inner and outside blades for cutting left and right flanks of gear, respectively. b\u03b1 is the pressure angle, h\u03b1 is the hook angle, and i\u03b2 is the initial setting angle. 0\u03b4 is the offset angle to let the normal of plane T align the cutting direction. 3 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 02/01/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use EDGE. As shown in Fig. 7, the profile of cutting edge can be straight-lined or circular ( ( )l lr ) with a circular-arc tip fillet ( ( )f lr )", + " The position vector of the blade edge may be specified using homogeneous coordinates as [ ]( )= ( ) 0 ( ) 1 T l l lu x u z ur (5) For the straight-lined cutter edge, ( )( ) 2 21 1 ( ) ( ) 1 1 2 2 ( ) ( cos )( ) sin ( ) cos ( ) sin fl l cf fl b l f l b l cf f x u x ux u u z u u z u z u \u03c1\u03b1 \u03b1 \u03c1 \u23a7 = \u00b1 \u2212\u23a7 = \u00b1\u23aa \u23aa \u23a8 \u23a8 = = +\u23aa \u23aa\u23a9 \u23a9 (6) where tan tan( / 4 / 2)cf r b f b cf r f x h z h \u03b1 \u03c1 \u03c0 \u03b1 \u03c1 = + \u2212\u23a7\u23aa \u23a8 = \u2212\u23aa\u23a9 For the circular cutter edge, ( ) 1 1 ( ) 1 1 ( ) 2 2 ( ) 2 2 ( ) ( cos( )) ( ) sin( ) ( ) ( / cos ) ( ) sin l l cl c b l l cl c b f l cf f c c f l cf f x u x u z u z u x u x u z u z u \u03c1 \u03b1 \u03c1 \u03b1 \u03c1 \u03c1 \u03c1 \u03c1 \u23a7 = \u00b1 + \u2212 \u23aa = \u2212 \u2212\u23aa \u23a8 = \u00b1 \u2212\u23aa \u23aa = +\u23a9 (7) where 2 2 cos( ), sin( ) ( cos ( ) ( sin( )) ) cl c b cl c b cf c b c f r f c b cf r f x z x h z h \u03c1 \u03b1 \u03c1 \u03b1 \u03c1 \u03b1 \u03c1 \u03c1 \u03c1 \u03c1 \u03b1 \u03c1 = \u2212 =\u23a7 \u23aa\u23aa = \u2212 \u2212 + \u2212 \u2212 \u2212\u23a8 \u23aa = \u2212\u23aa\u23a9 f\u03c1 and c\u03c1 are the fillet radius and the curvature radius of blade, respectively. The upper and lower signs in the equations correspond to the inner and outside cutter blades, respectively. As shown in Fig. 6, the coordinate systems lS and tS are rigidly connected to the cutting edge and cutter head, respectively, while the auxiliary coordinate systems mS , nS , and pS denote the relative positions of the cutter edge on the cutter head. Through coordinate transformation, the position vector of the cutter blade can be represented in the coordinate system tS as follow: ( )= ( )t tp pn nm ml lu ur M M M M r (8) where 0 0 0 0 0 cos sin 0 0 1 0 0 sin cos 0 0 0 1 0 0 = 0 0 1 0 0 0 1 0 0 0 0 1 0 0 0 1 cos sin 0 0 1 0 0 0 sin cos 0 0 0 cos sin 0 0 0 1 0 0 sin cos 0 0 0 0 1 0 0 0 1 i i i i tl h h h h r\u03b2 \u03b2 \u03b2 \u03b2 \u03b4 \u03b4 \u03b4 \u03b4 \u03b1 \u03b1 \u03b1 \u03b1 \u2212\u23a1 \u23a4 \u23a1 \u23a4 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a3 \u23a6 \u23a3 \u23a6 \u2212\u23a1 \u23a4 \u23a1 \u23a4 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a3 \u23a6 \u23a3 \u23a6 M - According to the tooth number of generating gear and the selected aR , the number of blade groups 0z is calculated as Eq" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002645_amm.556-562.2677-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002645_amm.556-562.2677-Figure2-1.png", + "caption": "Fig. 2 The MFS-MG experimental platform", + "texts": [ + " Considering high-frequency band of vibration signal contains the key fault information of rolling bearing, the fault characteristics could be extracted from the first numbers of IMFs [8]. Therefore, the first two IMFs will be reserved simultaneously. The overall diagnostic process is shown in Fig. 1. In the section, the approach proposed is applied to detect faults of rolling bearing. The bearing fault data with the sample frequency 12.800KHz are acquired from MFS-MG experimental platform, as shown in Fig. 2. The fault bearing of the type ER-12K is installed at the bearing holder on left side and the normal bearing is on right side, the spindle speed is 1790rpm (29.83Hz) and the end of the gearbox is free of loading. A computer online monitoring system is available for data acquisition and the vibration signals of bearings with four fault types (including outer race fault, inner race fault, rolling element fault and compound fault) were collected. In the present, the outer race fault data will be used" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001813_mfi.2015.7295819-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001813_mfi.2015.7295819-Figure1-1.png", + "caption": "Fig. 1. Button spinner (One-handed manipulation) [8]", + "texts": [ + " In his work, the linearly approximated model is suggested in short control time for the ribbon as a thread-like object, and he showed that we can make any shape with ribbon as long as the two conditions of constant velocity and high-speed motion are met, theoretically and experimentally. His method is highly effective when it comes to the bending control of a flexible object itself. In this paper, we deal with evaluating the high-speed visual sensing and feedback control in non-linear system, using an target object connected to other flexible object. Especially, we focused on high-speed rotating object via twisted thread and chose button spinner as an example task. Prior to addressing the research goal, we want to explain the button spinner first. Button spinner, shown in Fig. 1, is a traditional hand-toy consisting of a button and a thread. The button has a two holes and a thread goes through the holes and tie a knot to make a loop. After hanging two finger tips to inner side of the thread loop, we twist the thread by turning the button some turns to any direction, to set the initial state. Pulling two fingers simultaneously to outside of the loop makes the button start spinning. Repeating the pulling out and restoring motion of the fingers spins the button periodically" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001424_intmag.2015.7157215-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001424_intmag.2015.7157215-Figure1-1.png", + "caption": "Fig. 1 Unbalance mass in vertical axis the washing machine", + "texts": [ + " vd = Ra iod +(1+Ra /Rc )Vod +pLd iod, vq = Ra ioq +(1+Ra /Rc )Voq +pLq ioq (1) vod = - \u03c9eLq, voq =\u03c9eLd +\u03c9e\u03a8a (2) Te=Pn [\u03a8aioq +(Ld - Lq) iod ioq] (3) Here,vd and vq refer to the d- and q-axis terminal voltage, respectively; iod and ioq refer to the d- and q-axis current excluding core loss, respectively;Ra and Rc refer to the winding resistance and equivalent core-loss resistance, respectively;\u03a8a refers to the flux linkage of the permanent magnet; Pn refers to the pole pair; p=d/dt;Te and refers to the motor torque. \u03b8(t)=\u222b(\u03c9r -\u03c9 )dt+\u03b80 (4) Ft=mubgsin\u03a6sin\u03b8 (5) Te = Jzd\u03c9r/dt+mubgRsin\u03a6sin\u03b8+Tf (6) nioq = (2|mubgRsin\u03a6|+Tf )/(Pn\u03a8a) (7) Fig.1 shows the washing machine system. As the drum of the washing machine rotates, the rotation axis is tilted because of the centrifugal force of the unbalance mass [3]. \u03a6 refers to the angle between axis and axis; R refers to the radius of the drum; mub refers to the unbalance mass;Tf refers to the combined friction torque and cogging torque; Jz refers to the moment of inertia of the drum. Additionally, (5) determines the tangential force applied to drum by the unbalance mass. In the steady state, when the electric motor is controlled according to id=0 vector control, a current ripple can be obtained, as defined by (7)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001965_978-3-658-05978-1_7-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001965_978-3-658-05978-1_7-Figure3-1.png", + "caption": "Figure 3: The chassis and the powertrain of the BMW i8.", + "texts": [], + "surrounding_texts": [ + "The mechanical chassis is divided into two modules: the front and the rear module. The front module carries the electric engine and is connected to the tire rods by a double wishbone suspension. The wheels have a special slim wheel design with a reduced rolling resistance. The rear module integrates the combustion engine, the gearbox and the fuel tank. The five link axle is mounted directly to the rear module. The chassis is equipped with electronic actuators, which bring an additional margin for shaping the driving dynamic character. The front wheels are engaged by a torque feedback controlled axial parallel Electric Power Steering system (EPS). This steering system enables a haptic feedback which gives the transparent feeling for the tire-road contact that is generally expected from a sports car. Furthermore, the vertical damping control system (VDC) allows variable adaption of the dampers to ensure a sportiness as well as good riding comfort. The brakes on both axles are equipped with lightweight disks. The hydraulic system brings the pressure to the brake lining. It is a hybrid brake hydraulic system which balances the deceleration torque between the electric engines of the powertrain and the brakes. If the driver pushes the brake pedal, at first the electric engine is engaged to recuperate energy back to the battery. If more brake torque is required by the driver, the electric engine is assisted the by friction brakes. If the electric motor cannot sustain the deceleration torque, assisting pressure can be requested from compensation reservoir overlaps. The brake force of the driver is assisted by an electrical brake force amplifier (ELUP) to ensure assistance of brake force even when the combustion engine is switched off. Finally because of the split axle hybrid powertrain layout nearly ideal all-wheel drive features are possible as the vehicle can be driven purely by the front engine as well as purely by the rear engine. With both engines even an arbitrary power distribution between front and rear can be achieved. Challenges and requirements to driving dynamics First and foremost a sports car, the driving dynamics have to fulfill high requirements to reflect the BMW\u2019s claims for sheer driving pleasure. Even under efficiency aspects the driving dynamic features have priority. In detail, it must be assessed if the actual driving situation is more on the economical side or on the high driving dynamics side. This means changing from a pure electrical front axle driven vehicle to a torque vectoring controlled two axle driven vehicle. The BMW i8 demands absolutely reproducible driving behavior with an intuitive handling of a state of the art sports car. A state of the art plug in hybrid sports car should not only meet all known characteristics of a conventional sports car and also the characteristics of a hybrid car. It must be very efficient, always provide enough energy to drive pure electric and allow the driver to request the maximum system power of both engines in mixed driving modes. If driving in efficient driving mode, there should also be the possibility to regenerate energy while braking and driving in steady state situations in all possible friction conditions. Typical hybrid vehicles have the problem of a variant driving behavior, which means a significant change of driving dynamic characteristics when the high voltage battery is empty or completely charged. This needs to be avoided for a state of the art sports car. That means a compensation of variances should be developed to fulfill the target of a high performing state of the art sports car. For a sports car it is obvious that this should all go together with a high reproducibility, with a highly foreseeable response, good controllability and feedback of the driving situation, also intuitive handling and controllable dynamics and agility. The BMW i8 must provide enough convenience regarding long range traveling ability and appropriate mid range electric drivability to be suitable for everyday use." + ] + }, + { + "image_filename": "designv11_84_0000947_978-3-540-73958-6_5-Figure5.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000947_978-3-540-73958-6_5-Figure5.2-1.png", + "caption": "Fig. 5.2. Airframe\u2019s mechanics", + "texts": [ + " MARVIN (multi-purpose aerial robot vehicle with intelligent navigation) Mark II (M2) is an autonomous unmanned aerial vehicle (UAV) that was built in 2005 (see Fig. 5.1). Its predecessor [42] was developed from 1997 to successfully win the International Aerial Robotic Competition (IARC) [15] in 2000. From 2002 to 2005 it was used in cooperation with other UAVs in the European research project COMETS [7]. The M2 airframe is a commercially available helicopter made in Germany by Aero-Tec [2]. Its main feature is the sealed, full metal, single-stage main rotor gear which is mounted directly on the engine, building a self-supporting unit (see Fig. 5.2). Its additional features make it a good choice for an autonomous helicopter airframe. The setup acquired by TUB is equipped with a magnetic sensor on the main rotor axis to be able to measure the rotor\u2019s rotation per minute (r p m), a high-stand landing gear with mounting capabilities and an electric starter. The remote control (RC) equipment and gyro uses standard parts produced by Graupner [24]. The engine is a Zenoah G230RC two-stroke 23 cm3 gasoline-oilmix engine with a maximum power output of 1", + " Changes in lift, which are needed to move the helicopter, are produced by changing the main rotor blades\u2019 pitch. This is done symmetrically (collective pitch) to change the lift and asymmetrically (cyclic pitch) to accelerate in a direction. To set the pitch relative to the helicopter a so-called swashplate (SP) is used. It is moved by three servos. The tail rotor is used to compensate for the main rotor and engine torque and changes the heading of the helicopter. Its (collective) pitch is controlled by a single servo. A fifth servo moves the throttle. The SP and the throttle\u2019s servo can be seen in Fig. 5.2. The three SP servos are inside the dark front box of the helicopter, but their linkage is visible. The servos are controlled by pulse width modulated (PWM) signals where the PW is proportional to the servo anchor\u2019s position. In radio controlled flight the servos are driven by the RC receiver as received from the RC. The RC pilot does not control directly the servos but uses a higher-level approach. The pilot has two 2-way sticks, one is controlling the cyclic pitch, which has 2 degrees of freedom (DOF) and the other one is controlling the collective and the tail rotor pitch with 1 DOF each" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003715_robio.2014.7090476-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003715_robio.2014.7090476-Figure6-1.png", + "caption": "Fig. 6. Orientation planning and projection", + "texts": [ + " For the first constraint, the axis of the tool should be in the normal direction of the surface at the desired point, or in other words, the tool axis should direct to the center point of the sphere. To satisfy this constraint, according to the kinematics, the Z axis direction of the tool coordinate frame should be \u2192 z = (xp \u2212 xo\u2032 , yp \u2212 yo\u2032 , zp \u2212 zo\u2032), where (xp, yp, zp) and (xo\u2032 , yo\u2032 , zo\u2032) are the coordinates of the TCP and the sphere center in the base frame of the robot, respectively, refer to Fig.5. For the second orientation constraint, suppose that the tool axis is to keep a constant angle \u03b8 with the Z axis of the object frame, refer to Fig.6(a). Let P (x\u2032, y\u2032, z\u2032) denote the coordinates of TCP in the object frame. Then the coordinates of the intersecting point of the tool axis and the Z axis of the object frame can be easily found as (0, 0, \u221a x\u20322 + y\u20322 \u2217 cot\u03b8 + z\u2032). The Z axis orientation of the tool coordinate frame is thus (in the object frame) \u2192 z = ( \u2212x\u2032,\u2212y\u2032, \u221a x\u20322 + y\u20322 \u2217 cot\u03b8 ) . Since x\u2032 = xp \u2212 xo\u2032 , y \u2032 = yp \u2212 yo\u2032 , and z\u2032 = zp \u2212 zo\u2032 , we get \u2192 z = ( xo\u2032 \u2212 xp, yo\u2032 \u2212 yp, \u221a (xo\u2032 \u2212 xp)2 + (yo\u2032 \u2212 yp)2 \u2217 cot\u03b8 ) . In a similar manner, if the tool axis is to keep constant angle \u03b8 with X or Y axis, then its orientation is, respectively, \u2192 z = (\u221a (zo\u2032 \u2212 zp)2 + (yo\u2032 \u2212 yp)2 \u2217 cot\u03b8, yo\u2032 \u2212 yp, zo\u2032 \u2212 zp ) , \u2192 z = ( xo\u2032 \u2212 xp, \u221a (zo\u2032 \u2212 zp)2 + (xo\u2032 \u2212 xp)2 \u2217 cot\u03b8, zo\u2032 \u2212 zp ) ", + " Any vector in this plane and intersecting the Z-axis can be taken as the X-axis of the tool frame. For convenience of calculation and search for directions, get a point (cos \u03b8, sin \u03b8, w) in the plane, where w = z\u2032p \u2212 kx \u2217 cos\u03b8/kz \u2212 ky \u2217 sin\u03b8/kz , \u03b8 \u2208 [0, 360]. Points ( x\u2032 p + sin \u03b8, y\u2032p + cos \u03b8 ) determine a circle and different directions of radii in a plane parallel with the X-Y plane of the object frame. When the circle is projected to the plane perpendicular to the Z axis of the tool frame, its projection is an ellipse, and the relation between direction vectors is kept, as shown in Fig.6(b). Calculating in this way will be simple and accurate. The vector from the trajectory point P to a point on the ellipse, ( cos\u03b8, sin\u03b8, w \u2212 z\u2032p ) , may be taken as the X axis of the tool frame. The Y axis is then determined using right-hand rule by the Z and X axis. All the six parameters to determine the configuration of the tool frame are figured out. \u23a7\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23a8 \u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23a9 x = x\u2032 p + xo\u2032 y = y\u2032p + yo\u2032 z = z\u2032p + zo\u2032\u2192 x = (cos \u03b8, sin \u03b8, w \u2212 z\u2032p)/Lx\u2192 z = (\u2212x\u2032 p, \u2212y\u2032p, \u221a (x2 + y2) \u2217 cot \u03b8 \u2212 z\u2032p)/Lz\u2192 y = ( \u2192 z \u00d7\u2192 x)/Ly (1) where ( x\u2032 p, y \u2032 p, z \u2032 p ) is the trajectory point coordinate in the object frame, and (xo\u2032 , yo\u2032 , zo\u2032) is object center coordinate in the robot base frame" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002979_s11249-015-0466-9-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002979_s11249-015-0466-9-Figure3-1.png", + "caption": "Fig. 3 Cross-section view of idealized stationary heat pipe disk", + "texts": [ + " Figure 2c shows the polished front end (rubbing face), which is lapped to 1\u20132 helium light band. Three thermocouple holes are drilled through the sidewall blocks; two in the front end for measuring the face temperature (one near the inner diameter and another close to the outer diameter) and the third hole for measuring the rear end temperature. The rotating disk is a conventional disk, without a heat pipe, also made of 17-4 PH and heattreated to the same hardness of 45 Rockwell C, see Fig. 2d. As shown in Fig. 3, during the operation, water inside the wick under the rubbing face absorbs friction heating and vaporizes. Then, vapor flows through the near-vacuum space inside the heat pipe housing and condenses at the sidewall and the bottom of the disk. Finally, water flows back to the area under the rubbing face through the wick. 3.2 Rotating Disk with Built-In Heat Pipe The making of heat pipe rotating disk follows the same procedure as that of stationary disk. Figure 4 shows the design of the heat pipe rotating disk, and Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003721_detc2014-34297-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003721_detc2014-34297-Figure3-1.png", + "caption": "FIGURE 3. Serially connected kinematic chain corresponding to the not-direct sum of two subalgebras sO1 + sO2 .", + "texts": [ + " Not-direct sum of subalgebras. In this case dim(A1 +A2)< dim(A1)+dim(A2) (2) It should be noted that, in the synthesis of legs of parallel platforms, is undesirable to have passive or redundant degrees of freedom. Hence, in order to obtain a serial chain that forms a leg of a parallel platform is necessary to eliminate redundant or passive kinematic pairs. Therefore, the number of kinematic pairs must be equal to dim(A1 +A2), see definition 2. On this regard, consider the serially connected kinematic chains shown in Figure 3. In this case, the not-direct sum of two subalgebras is given by sO1 + sO2 , where sO1 , represents the subalgebra associated with the spherical displacements around the point O1 while sO2 , represents the subalgebra associated with the spherical displacements around the point O2. Figure 3(a), shows the serially connected kinematic chain where the redundant kinematic pair is represented by V, note that any of the kinematic pairs of the chain sO1 and sO2 can be redundant. It can be proved that the dimension of the screw system associated 4 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use with the serially connected chain is 5, furthermore, the ordered screw system have locally constant rank, see definition 2 and corollary 1. Figure 3(b), shows the serially connected kinematic chain where the redundant kinematic pair has been removed. It can be proved that the dimension of the screw system associated with the serially connected chain is 5, furthermore, the ordered screw system have locally constant rank, see proposition 2. Consider the serially connected kinematic chains shown in Figure 2, where the ordered screw system of locally constant rank is generated by the direct sum t\u22a5u\u03021 \u2295 sO. Simply changing the direction of the first revolute pair of the subalgebra sO, could be considered like a serially connected kinematic chains generated by the not-direct sum gu\u03021 +sO" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001423_aim.2015.7222672-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001423_aim.2015.7222672-Figure1-1.png", + "caption": "Fig. 1. The VTAV (a)Top View (b)Side View", + "texts": [ + " Aircrafts such as Bell Boeing V-22 Osprey, Bell X-22 and Curtiss-Wright X-19 have served well in many military applications utilizing the thrust vectoring capability [6][7] . The Vectored Thrust Aerial Vehicle (VTAV) discussed in this paper introduces thrust vectoring capability to a tri-rotor platform [8]. The VTAV platform consists of three ducted fans, and the rear two ducted fans have the capability to change the direction of generated thrust by turning around an axis common to both fans, the a2a3 axis, as shown in Fig.1. This provides the VTAV an additional control input compared to the quadrotor with one less propeller, enabling H.Jayakody and J.Katupitiya are with School of Mechanical and Manufacturing Engineering, University of New South Wales, Sydney, Australia. 1hiranya.jayakody@student.unsw.edu.au 2jay.Katupitiya@unsw.edu.au it to perform zero roll and pitch flight in space. The design also provides additional resilience against wind disturbances. Both linear and nonlinear control methodologies have been implemented to autonomously control UAVs [9][10][11]", + " (4) Based on the work in [17], vector f(x) follows as, f(x) = \u2212gs\u0398 \u2212 (wq \u2212 vr) gc\u0398s\u03a6 \u2212 (ur \u2212 pw) gc\u0398c\u03a6 \u2212 (pv \u2212 uq) [(Izz \u2212 Iyy)qr] /Ixx [(Ixx \u2212 Izz)pr] /Iyy [(Iyy \u2212 Ixx)pq] /Izz (5) with g representing the gravity and s, c standing for sine and cosine functions respectively. The terms \u03a6,\u0398 and \u03a8 represent the roll, pitch and yaw angles of the vehicle, respectively. The vector g(x) can be represented as, g(x) = diag [ 1 m , 0, 1 m , f Ixx , e Iyy , f Izz ] (6) where, parameters e and f denote the lengths from thrusters to the center of gravity of the VTAV as shown in Fig. 1. The aerodynamic drag forces and their resulting moments are caused by the external wind disturbance at a given time. Therefore, the disturbance vector d(x) can be derived as, d(x) = FAxb + \u22113 j=1 FDjxb FAyb + \u22113 j=1 FDjyb FAzb + \u22113 j=1 FDjzb [ \u22113 j=1 (MDjxb ) ] 1 Ixx [ \u22113 j=1 (MDjyb ) ] 1 Iyy [ \u22113 j=1 (MDjzb) ] 1 Izz . (7) The terms FA, FDj and MDj represent the aerodynamic drag forces, ram drag forces and moments due to these drag forces in each direction of the VTAV\u2019s body coordinate system" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003944_icamechs.2013.6681812-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003944_icamechs.2013.6681812-Figure2-1.png", + "caption": "Fig. 2 Sketch of PxPyPzRz limb", + "texts": [], + "surrounding_texts": [ + "The Singularity Research of A Class of 4-DOFs Parallel Mechanism and Configuration Analysis\nYe Ji Department of Mechanical and Electrical Engineering\nLuoyang Institute of Science and Technology, , No.90 Wangcheng Road, Luoyang, 471023, China Corresponding author Email: ji2000ye@126.com\nAbstract\u20144-SPS parallel mechanism(PM) adding a constraint driven limb, a new kind of parallel mechanism which possesses four degrees of freedom(DOFs) is built. These mechanisms are composed of fixed and moving platform by means of four parallel driving-limbs and one constraint driven-limb. Moving character is analysed by using screw theory and the different DOFs of PMs are calculated by revised Kutzbach-Gr\u00fcbler formula. The moving platform of this mechanism may have two translational and two rotational (2T2R) DOFs, 1T3R DOFs and 3T1R DOFs. The special position of PM leads to determinant of constraint screws being equal to zero by line geometry method, that is to say, this class of PM is clear at singular configuration. This analysis lays a foundation for further research on kinematics and dynamics.\nKeywords \u2014 parallel mechanism, singular configuration, screw theory, line geometry\nI. INTRODUCTION 2-DOFs, 3DOFs, 4-DOFs and 5-DOFs PMs are knows as spatial imperfect-DOF PMs and 4-DOFs PMs which may be used for 4-axis parallel machine tool and oscillating screen of harvester etc. are few studied. Above all, PMs of 4-DOFs have widespread prospect in application.\nSingularity is ubiquitous for mechanisms. Singular position can be segmented into boundary singularity, configuration singularity and structured singularity by velocity constraint equation [1]. Based on the forward and inverse kinematics, Raffaele puts forward positive singularity and inverse singularity [2]. Terminal singularity and driver singularity are proposed by O\u2019Brien [3]. Singularity is divided into limp singularity and platform singularity by Zlatanov [4]. To sum up, there are many means of classification according to different research ways. Merlet discovered some singular configurations of Stewart PM by Grassmann Geometry [5]. The same method was used to search the singularity of a three-leg 6-DOFs PM by Monsarrat [6]. Kanaan researched a kind of lower mobility PM by Grassmann\u2013Cayley algebra and discovered the singular configuration [7]. Alon researched the singularity of 3-DOFs CaPaMan PM by Line Geometry [8].\nBased on configuration evolution, parallel mechanism possessing 4-driving-limbs are acquired through Stewart parallel mechanism getting rid of two driving limbs. According to the screw theory and Kutzbach-Gr\u00fcbler criterion, this class of spatial PMs has four degrees of freedom whereas the prismatic joints are usually chosen as the active joints. Base on the screw theory, by constraint screws constituting determinant,\nthis paper analyzes kinematic property and find out some singular positions of spatial PM.\nII. DESCRIBTION OF 4-SPS PM As it is shown in Fig.1, this PM includes four SPS limbs and one driving-limb. The symbol of Pi(i=1,2\u20264) stands for prismatic pair and Si (i=1,2\u20264) shows spherical pair. The platforms are connected by spherical pairs and the prismatic joint of Pi, and the prismatic joint of Pi may move along limp. The platform of A1A2A3A4 may be in motion and the platform of B1B2B3B4 is fixed. The coordinates of {O}\u2014Oxyz is located at the center of fixed platform, meanwhile the point of O is the geometric center of fixed platform and the coordinates of {O'}\u2014O'x'y'z' is located at the center of movement platform, meanwhile the point of O' is the geometric center of movement platform. A driven-limb restricts the DOFs of movement platform on O'O-line.\nP1 P2\nP3 P4\nx'\ny'z'\nO'\nx\ny\nz\nO B1 B2\nA1 A2\nB3B4\nA4 A3\nFig.1 4-SPS PM configuration\nIII. CONFIGURATION EVOLUTION OF 4-SPS PM\nA. configuration evolution of 3T1R DOFs PM SPS-limb that has seven kinematic screws is widely used for PM. The constrained screw of this limb is nonexistent and six linearly independent kinematic screws is existent, so driven-limb need restrict the DOFs of movement platform by two constrained screws.\nAssuming that x-rotation DOF and y-rotation DOF of movement platform are restricted, the kinematic screws of driven-limp may be given by\nProceedings of the 2013 International Conference on Advanced Mechatronic Systems, Luoyang, China, September 25-27, 2013\n978-1-4799-2519-3/13/$31.00 \u00a92013 IEEE", + "1 (0 0 0;1 0 0)$ 2 (0 0 0;0 1 0)$ 3 (0 0 0;0 0 1)$ 4 (0 0 1;0 0 0)$\nConstraint screws may be obtained as follows r 1 (0 0 0;1 0 0)$ r 2 (0 0 0;0 1 0)$\nwhere 1 r$ denotes the constraint moment along the x-axis and\n2 r$ denotes the constraint moment along the y-axis about the {O}-coordinate, so the topological structure of driven-limb is as follows\nThe PM of 3T1Rz DOFs can be expressed 4-SPS/PxPyPzRz and the PM of 3T1Rx and the PM of 3T1Ry DOFs are obtained in a similar way, so 4-SPS/PxPyPzRx is belonged to PM of 3T1Rx DOFs and 4-SPS/PxPyPzRy is belonged to PM of 3T1Ry DOFs.\nB. configuration evolution of 2T2R DOFs PM If the y-movement DOF and Z-rotation DOF of movement\nplatform are restricted, the kinematic screws of driven-limp may be given by\n1 (1 0 0;0 0 0)$ 2 (0 1 0;0 0 0)$ 3 (0 0 0;0 0 1)$ 4 (0 0 0;1 0 0)$\nConstraint screws may be obtained as follows r 1 (0 1 0;0 0 0)$ r 2 (0 0 0;0 0 1)$\nwhere 1 r$ denotes the constraint force along the y-axis and 2 r$ denotes the constraint moment along the z-axis about the {O}coordinate, so the topological structure of driven-limb is as follows\nThe PM of 2Txz2Rxy DOFs can be expressed 4- SPS/PxPzUxy and the other PMs of 2T2R DOFs are obtained in a similar way, so 4-SPS/PxPyUxy is belonged to PM of 2Txy2Rxy DOFs, 4-SPS/PyPzUxy is belonged to PM of 2Tyz2Rxy DOFs, 4- SPS/PxPzUxz is belonged to PM of 2Txz2Rxz DOFs, et al.\nC. configuration evolution of 1T3R DOFs PM Last, the x-rotation DOF and y-rotation DOF of movement\nplatform are restricted, so the kinematic screws of driven-limp may be given by\n1 (0 0 0;0 0 1)$ 2 (1 0 0;0 0 0)$ 3 (0 1 0;0 0 0)$ 4 (0 0 1;0 0 0)$\nConstraint screws may be obtained as follows r 1 (0 1 0;0 0 0)$ r 2 (1 0 0;0 0 0)$\nwhere 1 r$ denotes the constraint force along the y-axis and 2 r$ denotes the constraint moment along the x-axis about the {O}coordinate, so the topological structure of driven-limb is as follows\nThe PM of 1Tz3R DOFs can be expressed 4-SPS/PzS and the PM of 1Tx3R and the PM of 1Ty3R DOFs are obtained in a similar way, so 4-SPS/PxS is belonged to PM of 1Tx3R DOFs and 4-SPS/PyS is belonged to PM of 1Ty3R DOFs.\nIV. PROVING OF DOFS CALCULATION The revised Kutzbach-Gr\u00fcbler formula for spatial\nmechanisms is as follows\n978-1-4799-2519-3/13/$31.00 \u00a92013 IEEE", + "( 1) 1 g M d n g f vii \n(1)\nwhere M is DOFs of mechanism; d(d=6-\u03bb) is the order of mechanism; \u03bb is the general constraint; n is the number of part; g is the number of kinematic pair; fi is DOFs of ith kinematic pair; v is the number of overconstrained except general constraint; \u03be is the passive DOFs of mechanism.\nThe screw coordinates of SPS-chain include seven kinematic screws, and six screws are linearly independent and none constraint screw which is inverse with all the kinematic screws, so \u03bb=0, d=6, in addition, there is none overconstrained, that is to say, v=0. There is a passive DOF revoluting around its axes of SPS-chain, so \u03be=4. PMs of 3T1R DOFs are calculated from Eq.(1) as follows\n6 (13 16 1) 32 4 4M \nPMs of 2T2R DOFs are calculated from Eq.(1) as follows\n6 (12 15 1) 32 4 4M \nPMs of 1T3R DOFs are calculated from Eq.(1) as follows\n6 (11 14 1) 32 4 4M \nV. SINGULAR CONFIGURATION ANALYSIS Singularity of spatial mechanism that may seriously influence the kinematic property is an inherent property. Based on line geometry, locked all the driving-pairs so that\n' 1 r$ , ' 2 r$ , ' 3 r$ and ' 4 r$ are showed, if 0S (S shows matrix of all\nthe constraint screws), singular configuration will generate. Assuming that the length of the side of moving platform is 2a and the length of the side of fixed platform is 2b, next two kinds of singularity should be discussed. For example, the constraint screws of 2Txz2Rxy DOFs PM is given by\nr 1 (0 1 0;0 0 0)$ r 2 (0 0 0;0 0 1)$\nCase 1: there isn\u2019t point of four driving-limps\u2019 intersection\nIf four driving-limps are parallel, in other words, '\n1 r$ , ' 2 r$ , ' 3 r$ and ' 4 r$ are parallel. Because screw linear correlation has nothing to do with the choice of the coordinates, suppose constraint force screws ( '\n1 r$ , ' 2 r$ , ' 3 r$ and ' 4 r$ ) are parallel to z-axis. Now calling\n' ' ' ' 1 2 1 2 3 4( ; ; ; ; ; )r r r r r rS $ $ $ $ $ $\nwe have\n0 1 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 1 0 0 0 1 0 0 0 1 0 a a a a a a a a S\nIn this case four driving-limps are locked whereas DOF of moving platform isn\u2019t equal to zero, so this mechanism is located at singular configuration.\nCase 2: The point of four driving-limps\u2019 intersection is existent and the {O}-coordinate is translated to B1-point in order to be easy to calculate, now S is obtained as follows\n31 33\n41 43 45\n51 52 53 54 56\n61 62 63 64 65 66\n0 1 0 0 0 0 0 0 0 0 1\n0 0 0 0 0 0 0\n0\na\ns s s s s s s s s s s s s s s s\n S \uff082\uff09\nThree singular configurations may be existent.\n1) Assuming that four driving-limps intersect at only one point\nIf the point of intersection is (x0,0,z0), in others words, four constrain screws( '\n1 r$ , ' 2 r$ , ' 3 r$ and ' 4 r$ ) intersect at one\npoint, then we have\n0S\nwhere the element of Eq.(2) is as follows\n0 31 2 2\n0 0\nxs x z \n0 33 2 2\n0 0\nzs x z \n0 41 2 2\n0 0( )\nx as x a z 0 43 2 2 0 0( ) zs x a z \n0 45 2 2\n0 0( )\nazs x a z 0 51 2 2 2 0 0 xs x a z \n52 2 2 2 0 0 as x a z \n0 53 2 2 2\n0 0\nzs x a z \n0 54 2 2 2\n0 0\nazs x a z \n0 56 2 2 2\n0 0\naxs x a z \n0 61 2 2 2\n0 0( )\nx a s\nx a a z\n \n 62 2 2 2\n0 0( )\nas x a a z \n978-1-4799-2519-3/13/$31.00 \u00a92013 IEEE" + ] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.28-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.28-1.png", + "caption": "Fig. 2.28. Humanoid manipulator", + "texts": [ + " In particular, consider the configuration where the last joint is so that \u03b17 = \u03c0/2. To simplify, the kinematic structure allowing the articulation of the robot\u2019s head in Fig. 1.33. The torso can be modelled as an anthropomorphic arm (three DOFs), for a total of seventeen DOFs. Further, a connecting device exists between the end-effector of the anthropomorphic torso and the base frames of the two manipulators. Such device permits keeping the \u2018chest\u2019 of the humanoid manipulator always orthogonal to the ground. With reference to Fig. 2.28, this device is represented by a further joint, located at the end of the torso. Hence, the corresponding parameter \u03d14 does not constitute a DOF, yet it varies so as to compensate Joints 2 and 3 rotations of the anthropomorphic torso. To compute the direct kinematics function, it is possible to resort to a DH parameters table for each of the two tree kinematic structures, which can be identified from the base of the manipulator to each of the two end-effectors. Similarly to the case of mounting a spherical wrist onto an anthropomorphic arm, this implies the change of some rows of the transformation matrices of those manipulators, described in the previous sections, constituting the torso and the arms. Alternatively, it is possible to consider intermediate transformation matrices between the relevant structures. In detail, as illustrated in Fig. 2.28, if t denotes the frame attached to the torso, r and l the base frames, respectively, of the right arm and the left arm, and rh and lh the frames attached to the two hands (end-effectors), it is possible to compute for the right arm and the left arm, respectively: T 0 rh = T 0 3 T 3 t T t rT r rh (2.78) T 0 lh = T 0 3 T 3 t T t lT l lh (2.79) where the matrix T 3 t describes the transformation imposed by the motion of Joint 4 (dashed line in Fig. 2.28), located at the end-effector of the torso. Frame 4 coincides with Frame t in Fig. 2.27. In view of the property of parameter \u03d14, it is \u03d14 = \u2212\u03d12 \u2212 \u03d13, and thus T 3 t = \u23a1 \u23a2\u23a3 c23 s23 0 0 \u2212s23 c23 0 0 0 0 1 0 0 0 0 1 \u23a4 \u23a5\u23a6 . The matrix T 0 3 is given by (2.66), whereas the matrices T tr and T tl relating the torso end-effector frame to the base frames of the two manipulators have constant values. With reference to Fig. 2.28, the elements of these matrices depend on the angle \u03b2 and on the distances between the origin of Frame t and the origins of Frames r and l. Finally, the expressions of the matrices T rrh and T llh must be computed by considering the change in the seventh row of the DH parameters table of the DLR manipulator, so as to account for the different kinematic structure of the wrist (see Problem 2.14). As described in the previous sections, the direct kinematics equation of a manipulator allows the position and orientation of the end-effector frame to be expressed as a function of the joint variables with respect to the base frame", + " By applying the rules for inverting a block-partitioned matrix, prove that matrix A1 0 is given by (2.45). 2.11. Find the direct kinematics equation of the four-link closed-chain planar arm in Fig. 2.34, where the two links connected by the prismatic joint are orthogonal to each other. 2.12. Find the direct kinematics equation for the cylindrical arm in Fig. 2.35. 2.13. Find the direct kinematics equation for the SCARA manipulator in Fig. 2.36. 2.14. Find the complete direct kinematics equation for the humanoid manipulator in Fig. 2.28. 2.15. For the set of minimal representations of orientation \u03c6, define the sum operation in terms of the composition of rotations. By means of an example, show that the commutative property does not hold for that operation. 2.16. Consider the elementary rotations about coordinate axes given by infinitesimal angles. Show that the rotation resulting from any two elementary rotations does not depend on the order of rotations. [Hint : for an infinitesimal angle d\u03c6, approximate cos (d\u03c6) \u2248 1 and sin (d\u03c6) \u2248 d\u03c6 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003970_amr.779-780.664-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003970_amr.779-780.664-Figure4-1.png", + "caption": "Fig. 4 Placement of an ultrasonic sensor", + "texts": [ + " 3, the angle between the temperature sensor and the horizontal plan is 0\u00b0, and the sensor is directly inserted in the inner ring. In particular, the sensor probe should be placed in the raceway groove. Ultrasonic sensor. In addition to rolling element broken and raceway wear, crack in the raceway is also a critical factor to affect the service life of a slewing bearing. With a ultrasonic sensor, a crack can be promptly detected and accurately located. Thus\uff0can ultrasonic sensor embedded model (USM) is built. As shown in Fig. 4, the ultrasonic sensor is placed in the bottom of the inner ring through threaded connection. Fatigue life analysis of the smart slewing bearing Since the actual load a slewing bearing bears is much lower than its load capacity, the strength of a slewing bearing is usually enough. However, fatigue wear is inevitably to occur during the 20-year usage. When the fatigue wear level is to a certain extent, the slewing bearing is then required to be replaced by a new one [8]. Since the structure of the smart slewing bearing proposed in this paper is changed, its fatigue life must be calculated to ensure that it can still meet the demand service life. To conduct a fatigue life analysis by MSC.FATIGUE, three inputs are required, that is, geometry properties, material fatigue properties and a fatigue load spectrum. The flowchart of the fatigue life calculation of a 1.5MW smart yaw slewing bearing is shown in Fig. 5. Geometry properties. As stated above, the installation approaches of the sensors are shown in Fig. 2-Fig. 4. According to the research of Wang [9], the maximum dimension of the eddy current sensor mounting groove is 40mm\u00d720mm\u00d75mm, the temperature sensor mounting aperture is 10mm, and the ultrasonic sensor mounting aperture is 8mm. Besides, the maximum contact load between the rolling element and the raceway is 131.5KN when a 1.5MW slewing bearing withstanding its extreme load. The maximum stresses and deformation amounts of the raceways are calculated and shown in Table 1 respectively, which could be used as the material property parameters" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002845_s11249-011-9811-9-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002845_s11249-011-9811-9-Figure9-1.png", + "caption": "Fig. 9 Forces applied to the ball", + "texts": [ + " Figure 8 shows the effect of the initial gap size on the critical displacement. Even the data have slight fluctuation, they clearly indicate that the larger the initial gap, the smaller the critical displacement is. The size of the initial gap relates to the resistance of the leakage of entrapped oil. Hence, the dimple collapses earlier and the critical displacement is shorter for those dimples with larger initial gaps. Numerical analyses were carried out to simulate the impact, i.e. the formation of the dimple, and the subsequent lateral dimple movement. Figure 9 illustrates the mathematical model used in the simulation. The governing equations include the Reynolds equation, the elasticity equation, the viscosity\u2013pressure (Roelands) and the density\u2013pressure (Dowson and Higginson) equations. Numerical solutions were obtained with the multigrid method. During the impact or the entrainment process, the rigid separation h00 between the ball and the plate is determined by Newton\u2019s second Law. h00\u00f0t\u00de \u00bc h00\u00f0t Dt\u00de \u00fe v\u00f0t Dt\u00de Dt \u00fe a\u00f0t Dt\u00de Dt2 2 \u00f01\u00de a\u00f0t Dt\u00de \u00bc w\u00f0t Dt\u00de W mg m \u00bc R p\u00f0t Dt\u00dedxdy W mg m \u00f02\u00de The film thickness can be obtained by the elasticity equation as, h\u00f0x; y\u00de \u00bc h00 \u00fe x2 Rx \u00fe y2 Ry \u00fe 2 pE0 ZZ p\u00f0x0; y0\u00dedx0dy0 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi \u00f0x x0\u00de2 q \u00fe ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi \u00f0y y0\u00de2 q \u00f03\u00de The Reynolds equation is, o ox \u00f0qh3 g op ox \u00de \u00fe o oy \u00f0qh3 g op oy \u00de \u00bc 12 o\u00f0qh\u00de ot \u00fe 12 o\u00f0queh\u00de ox : \u00f04\u00de An experiment of HVI650 was simulated with the same operating conditions" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002480_2015-01-0680-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002480_2015-01-0680-Figure11-1.png", + "caption": "Figure 11. Piston assembly", + "texts": [ + " If the width of loading occupies half of the entire loading period, the loading period is 400 ms when the rotation wheel is rotating at 300 rpm, at which the test rig rotates. Figure 10 shows the load on the piston assembly and rotation angle of the small end of the connecting rod versus the angle of the crank shaft. connecting rod versus the angle of the crank shaft The connecting rod, piston and piston pin are production components. The piston pin is a hollow cylinder made of steel, the outer radius is 10 mm and inner radius is 5.5 mm; the surface of the piston pin is polished to Ra=0.08 um. The piston is made of aluminum alloy. The piston assembly is shown in Figure 11. All the bench tests were run under non-lubricated conditions. During the tests, floating pins were found to rotate in the piston bore. However, the pin's rotational motion became slower and slower until it stopped in the pin bore due to the increasing friction force caused by surface damage on the pin bore surface. After the pin stopped in the bore, the piston assembly was disassembled from the rig. The piston bore surface was observed and compared with scuffed bore surfaces run in a fired engine" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003500_s12205-013-0643-z-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003500_s12205-013-0643-z-Figure1-1.png", + "caption": "Fig. 1. Schematic Diagram of Cube Type Microbial Fuel Cell: (a) Batch Microbial Fuel Cell Reactor, (b)Placement of Electrodes in each Window", + "texts": [ + " The oxygen reduction catalyst Effect of Multiwall Carbon Nanotube Contained in the Exfoliated Graphite Anode on the Power Production and Internal Resistance of Microbial Fuel Cells Vol. 00, No. 0 / 000 0000 \u2212 3 \u2212 was mixed with Nafion solution (30%wt., Sigma-Aldrich Co., St. Louis, USA) as a binder, and the catalyst slurry was loaded on the other side surface of the primitive MWCNT electrode by screen-printing to complete the cathode. A cube type MFC, effective volume 972 mL, with window (4 cm \u00d7 4 cm) on four vertical walls, was constructed with polymethyl methacrylate (PMMA) (Fig. 1(a)). The cathode was installed on the air facing side of each window, and four different anodes including the control were fitted on the inside of each cathode together with a non-woven polypropylene sheet as a separator (Fig. 1(b)). Each anode was connected to its partner cathode to make an electric circuit using copper wire, and 200ohms of external resistance was inserted into the circuit. Anaerobic sludge taken from S sewage treatment plant was used for seed sludge. Artificial wastewater containing acetate (1,000 mg/L), phosphate buffer (50 mM), minerals (12.5 mL/L) and vitamin solution (12.5 mL/L) was prepared following the method of Song et al. (2012). For the start-up of the MFC reactor, artificial wastewater was filled up to 70% of the effective volume, and anaerobic seed sludge was added to the remaining volume of the MFC" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003678_1.4024704-Figure12-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003678_1.4024704-Figure12-1.png", + "caption": "Fig. 12 x-direction deformation (in meters) of the workpiece. Welding away from fixed face AA0B0B.", + "texts": [], + "surrounding_texts": [ + "The GTA welding process and the various transport processes involved are discussed in detail in Part I [2]. The mathematical model can be divided into two parts: (a) weld pool dynamics modeling and (b) structural analysis modeling. In the weld pool dynamics modeling, the melting/solidification problem is handled using the enthalpy-porosity formulation. The molten metal flow in the weld pool is obtained using the governing equations of continuity, momentum and energy, based on the assumption of incompressible laminar flow. The Navier\u2013Stokes (N\u2013S) momentum equation takes into account the mushy zone through the momentum sink term, and includes the electromagnetic (Lorentz) force as a body force term. The Lorentz force is determined using the current continuity equation in association with the steady state version of the Maxwell\u2019s equation in the domain of the workpiece for the current density and magnetic flux. The structural analysis model is developed based on isotropic material behavior. The elastic response is handled using the isotropic Hooke\u2019s law with temperature dependent Young\u2019s modulus and Poisson\u2019s ratio. For the inelastic response or plasticity, incompressible plastic deformation is assumed with rate-independent plastic flow and vonMises yield criterion. The yield strength is considered as a function of temperature only. Also, the bilinear isotropic hardening model is employed to consider the material strain-hardening behavior. The mathematical models for both weld pool dynamics and structural analysis have been discussed in extensive detail in Part I and hence is not represented here. However, it is to be noted that the analysis in this study ignores the influences from the arc pressure and a flat weld pool surface is assumed. These assumptions are reasonable for the present study and discussed in detail in Part I of the present study. Also, the boundary and initial conditions used in the mathematical model are described in detail in Part I." + ] + }, + { + "image_filename": "designv11_84_0003494_s11661-014-2462-3-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003494_s11661-014-2462-3-Figure2-1.png", + "caption": "Fig. 2\u2014Photograph of the built microtensile testing system.", + "texts": [ + " A Quanta 3D FEG SEM FIB with a Ga+ ion source operated at 30 kV under perpendicular ion impact was carefully used to cut the polished samples into microtensile pieces with 13 lm gauge length. These were finally thinned down to a cross section of 2 lm by 3 lm. Since the samples were machined from polycrystalline bulk material they are single crystals microtensile samples. 2. Micro-mechanical testing Themicrotensile sampleswere subsequently testedwith in situ observation of the microstructures from the scanning electron microscope (SEM). The construction of the microtesting system (Figure 2) used in this study involves a piezo-electric drive used to apply load to the sample through a linear air bearing. Loadsweremeasured using aminiature load cell with an overall load capacity of 0.5 N and a resolution of 0.001 N. Application of load at a load displacement step size of 40 nm gives a corresponding voltage output used to determine the loadwith a conversion relationship, 100 mV = 1 g. The microtensile samples were tested in situ by pulling to fracture through a load cell attached to a piezoelectric actuator and simultaneously observing the microstructures" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.33-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.33-1.png", + "caption": "Fig. 2.33. The four configurations of an anthropomorphic arm compatible with a given wrist position", + "texts": [ + "102), as pWx = \u00b1c1 \u221a p2 Wx + p2 Wy pWy = \u00b1s1 \u221a p2 Wx + p2 Wy which, once solved, gives the two solutions: \u03d11,I = Atan2(pWy, pWx) (2.109) \u03d11,II = Atan2(\u2212pWy,\u2212pWx). (2.110) Notice that (2.110) gives18 \u03d11,II = {Atan2(pWy, pWx) \u2212 \u03c0 pWy \u2265 0 Atan2(pWy, pWx) + \u03c0 pWy < 0. Atan2(y,\u2212x) = { \u03c0 \u2212 Atan2(y, x) y \u2265 0 \u2212\u03c0 \u2212 Atan2(y, x) y < 0. As can be recognized, there exist four solutions according to the values of \u03d13 in (2.100), (2.101), \u03d12 in (2.105)\u2013(2.108) and \u03d11 in (2.109), (2.110): (\u03d11,I, \u03d12,I, \u03d13,I) (\u03d11,I, \u03d12,III, \u03d13,II) (\u03d11,II, \u03d12,II, \u03d13,I) (\u03d11,II, \u03d12,IV, \u03d13,II), which are illustrated in Fig. 2.33: shoulder\u2013right/elbow\u2013up, shoulder\u2013left/elbow\u2013 up, shoulder\u2013right/elbow\u2013down, shoulder\u2013left/elbow\u2013down; obviously, the forearm orientation is different for the two pairs of solutions. Notice finally how it is possible to find the solutions only if at least pWx = 0 or pWy = 0. In the case pWx = pWy = 0, an infinity of solutions is obtained, since it is possible to determine the joint variables \u03d12 and \u03d13 independently of the value of \u03d11; in the following, it will be seen that the arm in such configuration is kinematically singular (see Problem 2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002681_amr.328-330.186-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002681_amr.328-330.186-Figure5-1.png", + "caption": "Fig. 5 Coordinate systems of internal gear couple Fig. 6 Models of Internal Gear Couple", + "texts": [ + " 7, meshing equation is derived from Willis theorem as follows [3]: t t t t t t x y X x Y y N N \u2212 \u2212 = . (8) Here, ( tX , tY ) are coordinates of the instantaneous center I in tS , tx N and ty N are the projections of the common normal vector of profiles at the contact point whose coordinates are ( tx , ty ). With Eq. 7 and Eq. 8, the mathematical formula of the tooth profile of the generated external gear can be derived. Tooth Profile of Conjugated Internal Gear. The relationship of coordinate systems for movement of internal gear couple are shown in Fig. 5. Coordinate systems 1S and fS have been defined as above stated. Coordinate system 2S ( 2x , 2y ) is rigidly connected to the internal gear and rotates with the internal gear. The origin of 2S coincides with the center of the internal gear. When the external gear rotates 1\u03d5 angle, the internal gear has a rotation of 2\u03d5 angle and the two angles have a relationship as follow: 1 2 1 2 r r \u03d5 \u03d5= . (9) Here, 1r is pitch circle radius of external gear and 2r is pitch circle radius of internal gear. The matrix of coordinate transformation from 1S to 2S is 1 2 1 2 2 21 1 2 1 2 2 cos( ) sin( ) cos sin( ) cos( ) sin 0 0 1 e M e \u03d5 \u03d5 \u03d5 \u03d5 \u03d5 \u03d5 \u03d5 \u03d5 \u03d5 \u03d5 \u2212 \u2212 \u2212 = \u2212 \u2212 \u2212 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002913_2011-01-2238-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002913_2011-01-2238-Figure3-1.png", + "caption": "Figure 3. Universal joint secondary couple loading", + "texts": [ + " Secondary couples are widely thought to be a second order, or twice per revolution, disturbance. However, the second order disturbance is to the bearings of the connected shafts and their housings where the housings are stationary. The secondary couple loading on the bearings is unreversed and the direction is about an axis perpendicular to the joint angle. To understand the effects of secondary couples on the rotating driveline, a careful study of the forces from the kinetics of a universal joint is necessary. From the top freebody-diagrams of Fig. 3 in the Appendix, a stress element on the top of the \u2018input torque\u2019 shaft will be in compression from the secondary couple. As the shaft system rotates 90\u00b0, depicted by the bottom free-body diagrams of Fig. 3 in the Appendix, the same element on the \u2018input torque\u2019 shaft has no loading from the secondary couple. As the shaft system rotates another 90\u00b0, the top free-body-diagrams again apply. The element will be on the bottom and in tension from secondary couple loading. As the shaft system rotates yet another 90\u00b0, the bottom free-body-diagrams again apply and the element has no loading from the secondary couple. This study of the element shows that the secondary couple disturbance to the \u2018input torque\u2019 shaft is first order and that it is a fully reversed couple loading" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002629_iembs.2011.6090460-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002629_iembs.2011.6090460-Figure1-1.png", + "caption": "Fig. 1 Tread-Walk 2", + "texts": [ + " of Robotics & Design for Innovative Healthcare (Panasonic), Graduate School of Medicine, Osaka University, Osaka, Japan and Faculty of Science and Engineering, Waseda University, Tokyo, Japan. (e-mail: takecando@gmail.com). Yo Kobayashi and Masakatsu G. Fujie are with the Faculty of Science and Engineering, Waseda University, Tokyo, Japan (e-mail: you-k@fuji.waseda.jp, mgfujie@waseda.jp). we have been developing a new vehicle called \u201cTread-Walk 2 (TW-2)\u201d, which is controlled by the walking movement (Fig. 1) [3], [4]. TW-2 estimates the user\u2019s desired walking velocity from the user\u2019s anteroposterior force and drives the treadmill belt. Also, TW-2 amplifies the walking velocity to the driving wheel to expand the user\u2019s range of travel. Therefore, TW-2 uses active treadmill velocity control, which allows the user to change velocity by changing the walking movement, similar to the way velocity is changed in natural walking. In other words, TW-2 has more intuitive mobility control. In order to control treadmill belt, TW-2 estimates the user\u2019s anteroposterior force acting on the split belt of the treadmill through the current value of the DC motor that drives the treadmill belt (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001429_9781118886397.ch14-Figure14.3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001429_9781118886397.ch14-Figure14.3-1.png", + "caption": "Figure 14.3 Horizontal conductor motion during through-fault. (a) end view; (b) side view; and (c) plan view (Source: EPRI (2006)).", + "texts": [ + " Even though such currents on the compact line may be less than the maximum fault currents attainable on the system, they may be sufficient to be determining in the selection of phase-to-phase spacing or in establishing the need for insulating spacers. The motion of conductors subjected to electromagnetic forces is similar to that of weighted, stretched strings, with the complication that the string is usually a compound conductor, such as an aluminum conductor, steel reinforced (ACSR). Relatively simple analyses of conductor motion of both vertically and horizontally spaced conductors can be shown to give results well within line design accuracy requirements. Figure 14.3 illustrates the conductor configuration used as a basis for calculations. It is assumed that the forces to which each catenary span of the conductor is subjected will cause the span to swing in a plane, as shown in Figure 14.3(a). The plan projection of each catenary is again a catenary (Figure 14.3(b) and (c)). This assumption, supported by experimental results, simplifies the calculation technique. The most severe fault is phase to phase on adjacent phases, which impresses a cyclic separating electromagnetic force. Since all spans of a line contributing to a through-fault will behave similarly, the net pole-top force along the span, and therefore motion, will be zero. Consequently, each span can be assumed to be rigidly terminated. The true shape of each conductor is a catenary, as shown in Figure 14" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002639_amr.314-316.653-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002639_amr.314-316.653-Figure3-1.png", + "caption": "Fig 3 The direction of the X axle Fig 4 The direction of the Y axle", + "texts": [ + " The mesh divided is shown as Fig2: Fig1 The solid model Fig2 The mesh model The connection rod does plane sport, but calculate behaviour is select the two most loaded position, the edformation of the connection rod is self relative deformation in the strength analyse and the both ends of the connection rod is loaded, so we choose the centre position of the line of the connection rod\u2019s big and small end hole as the fixed restraint, which restrict the displacement and rotation of the rigid body and calculate the deformed quantity which is realtive to the fixed restraint surface [4] . In the most compression condition, the most stress of the connection rod emerged in the unilateral internal surface of the rod\u2019s small end anear big end. In the most tension stress condition, the most stress emerged in the connected arc of the small end and the rod body. So we applied stress to the connection rod\u2019s small end, the emulator results is shown as Fig3-Fig8 : The results are mapped path and the result graph of the path is shown as Fig 9: From the figure of simulation result, we can see that the most distortion appeared in the compressed position of the connection rod and piston pin in the most compression case, where is closed to the most stress position. And the least diatortion appeared in the contact surface of the connection rod and bolt, which is gradually getting decrescence from the small end to big end. This is similar to the distortion trend of the stretching case, and they are all decided by structure from the connection rod\u2019s small end to big end which is gradully getting thick from weak [5] " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002322_1.4029294-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002322_1.4029294-Figure4-1.png", + "caption": "Fig. 4 A two-body system that includes a redundancy in the B matrix when CoM data are only gathered in the X\u2013Y plane", + "texts": [ + " Identical columns of B cause infinite workable solutions to the corresponding rows in the SESC vector, S. If a unique solution for S is desired, the redundant columns of B should be eliminated such that all columns are linearly independent. After eliminating unused rows and redundant columns, the reduced version of B is denoted as D 2 Rv n, where v is the dimension of the CoM data collected for each configuration and n is the number of unique columns that remain. A two-link spatial chain is analyzed herein as a simple example of modifying B. As shown in Fig. 4, the first joint rotates about the x axis, which is perpendicular to the page, and the second joint rotates about the y axis in their local reference frames. The CoM of the system is C \u00bc m1 M Ax\u00f0h1\u00dec1 \u00fe m2 M \u00f0Ax\u00f0h1\u00ded1 \u00fe Ax\u00f0h1\u00deAy\u00f0h2\u00dec2\u00de \u00bc \u00bdAx\u00f0h1\u00de Ax\u00f0h1\u00deAy\u00f0h2\u00de S (14) where S \u00bc m1 M c1 \u00fe m2 M d2 m2 M c2 8><>: 9>=>; \u00bc fS1 \u2026 S6gT (15) Expanding Eq. (14) Cx Cy Cz 8><>: 9>=>; \u00bc 1 0 0 ch2 0 sh2 0 ch1 sh1 sh1sh2 ch1 sh1ch2 0 sh1 ch1 ch1sh2 sh1 ch1ch2 264 375 S1 .. . S6 8><>: 9>=>; (16) where shi and chi denote sin hi and cos hi, respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002845_s11249-011-9811-9-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002845_s11249-011-9811-9-Figure1-1.png", + "caption": "Fig. 1 Schematic diagram of a experimental apparatus and b test scheme", + "texts": [ + " This study aims at finding out the effect of parameters such as speed, dimple depth, load, initial gap, and viscosity, on the moving range of a dimple within an EHL contact under pure rolling conditions. Experiments were carried out using a conventional optical EHL test rig. To understand and explain the experimental results, numerical analyses were completed and qualitative comparisons were provided. Furthermore, this study gives more insight into the transportation of impact dimple across the EHL contact, even though plentiful research works about squeeze and impact EHL have been carried out. The test apparatus is shown in Fig. 1a. A glass block is inlaid in an aluminium disc. The loaded side of the glass block is coated with a Cr film of about 20 nm in thickness. The steel ball is 25.4 mm in diameter. The magnitude or stroke of the impact is controlled by a servo motor, such that the desirable depth or size of an impact dimple can be obtained. The glass block is driven by the rotating steel ball in the present experiments to obtain pure rolling. Once the impact with a prescribed depth is formed (Step I), rotation of the ball is then activated (Step II)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001876_tmag.2015.2443017-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001876_tmag.2015.2443017-Figure1-1.png", + "caption": "Fig. 1. (a) Schematic and (b) finite-element model of the head\u2013disk interface for the case of asperity contact between a TFC slider and a disk [3], [4].", + "texts": [ + "2443017 the size of disk asperities is not uniform, and contact between a TFC slider and the same disk asperity can occur many times during the life of a disk drive. In this paper, the effect of asperity height and asperity diameter on scratches in the read/write shields is investigated, and the maximum temperature at the reader location is determined. In addition, plastic deformation and the temperature rise for repeated contacts between the TFC slider and the disk asperity are also investigated. In Fig. 1, a model of the head\u2013disk interface for the case of asperity contact between the slider and the disk is shown together with a finite-element model of contact between a 0018-9464 \u00a9 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. TFC slider and a disk asperity [3], [4]. As shown in Fig. 1(a), the asperity is modeled as a spherical cap on the top of a cylinder. The asperity height is h, and the asperity diameter is d . The disk and the asperity move with the same circumferential velocity Vx . The radius of the thermal protrusion is chosen to be 5 mm. The spacing between the disk surface and the bottom of the TFC slider is defined as FH. The location of the read and write elements is indicated in Fig. 1(b) by dashed lines. In our calculations, the effect of the read and write elements is neglected, because their widths are only a few nanometers. The materials of the asperity and the disk are chosen to be Al2O3 and NiP, respectively. The material of the read and write shields is assumed to be NiFe. The thermoelastic-plastic properties of Al2O3, NiP, and NiFe are summarized in Table I. The initial temperature of the disk and the asperity is assumed to be 300 K, while the initial temperature of the read and write shields is assumed to be 335 K [1]", + " This is because the contact pressure increases when the slider contacts the asperities with smaller diameters. From Fig. 4(b), we observe that the scratch width increases with increasing asperity diameter. Fig. 5 shows the maximum temperature Tmax at the reader location versus the diameter of asperity d , assuming h = 10 nm, FH = 2 nm, Vx = 20 m/s, and \u00b5 = 0.2. It can be seen in Fig. 5 that the value of Tmax increases with the asperity diameter. When the contact between asperity and read shields starts, the asperity is still in partial contact with the alumina region I [i.e., Al2O3 #1 in Fig. 1(b)]. During this partial contact, a certain amount of heat will be transferred from the alumina region I to the conductive read shields. The duration of the partial contact increases with an increase in the diameter of asperity. Consequently, the asperity with a larger diameter leads to a higher temperature at the read element [3]. Fig. 6 shows the scratch dimensions in the read/write shields versus the number of repeated contact cycles, assuming h = 15 nm, d = 300 nm, FH = 10 nm, Vx = 20 m/s, and \u00b5 = 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002230_1.4897827-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002230_1.4897827-Figure4-1.png", + "caption": "FIGURE 4. Dynamical model layout. The small circles on both sides of the spring represent the gear tooth mass.", + "texts": [], + "surrounding_texts": [ + "For the execution of the dynamical simulation a single-stage, symmetric involute spur gear reducer with a gear ratio of two, with no errors or modifications in the geometry was assumed as an example. However, the algorithms developed for this study are generic and any combination of gear sets can be modeled in the same way. The design variables that were assumed for the current simulations are gathered in the following table. The time-varying gear tooth stiffness for the assumed design variables was calculated as mentioned earlier by means of MatLab software [17] and the obtained results are shown in the following figures plotted about the angle of revolution of the pinion gear for a single meshing gear pair along the length of contact and for a constantly alternating number of gear pairs in contact during the operation of the reducer. of the pinion gear ( 1)." + ] + }, + { + "image_filename": "designv11_84_0003538_amm.325-326.870-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003538_amm.325-326.870-Figure3-1.png", + "caption": "Fig. 3. Coordinate systems used for the generated bevel gear", + "texts": [ + " In figure 2 is presented the coordinate system where OX1Y1Z1 is the coordinate system of the face gear and OXFYFZF is the coordinate system fixed at a flank. The parametrical equations of the face gear\u2019s flanks are: = \u22c5\u22c5\u2212\u22c5\u2212= \u22c5\u22c5\u2212\u22c5= \u03a3 vvuz tgvuvuy tgvuvux aa aa st ),( cossin),( sincos),( :)( 1 1 1 \u03c8\u03b1\u03c8 \u03c8\u03b1\u03c8 (1) = \u22c5\u22c5+\u22c5= \u22c5\u22c5\u2212\u22c5= \u03a3 vvuz tgvuvuy tgvuvux aa aa dr ),( cossin),( sincos),( :)( 1 1 1 \u03c8\u03b1\u03c8 \u03c8\u03b1\u03c8 (2) The used parameters are: \u03c8a the suitable angle of the quarter of the pitch, \u03b1 the pressure angle, u and v are liear parameters. In figure 3 is presented the relative movements between the face gear and the bevel gear, [6]. After the transformations, according to figure 5, the transfer matrix is obtained: \u22c5+\u22c5\u22c5\u2212= \u22c5+\u22c5\u22c5= \u22c5+\u22c5\u22c5= \u22c5+\u22c5\u22c5\u2212= \u22c5 \u22c5\u2212 \u2212\u22c5\u2212\u22c5 = 12122,3 12121,3 12122,2 12121,2 22,31,3 22,21,2 11 cossinsinsincos sinsincossincos coscossinsinsin sincoscossinsin coscos cossin sinsincoscoscos \u03d5\u03d5\u03d5\u03b4\u03d5 \u03d5\u03d5\u03d5\u03b4\u03d5 \u03d5\u03d5\u03d5\u03b4\u03d5 \u03d5\u03d5\u03d5\u03b4\u03d5 \u03b4\u03d5 \u03b4\u03d5 \u03b4\u03d5\u03b4\u03d5\u03b4 m m m m mm mmM T (3) Therefore the parametrical equations are: \u22c5= 1 ),( ),( ),( 1 ),,,( ),,,( ),,,( 1 1 1 212 212 212 vuZ vuY vuX M vuZ vuY vuX T\u03d5\u03d5 \u03d5\u03d5 \u03d5\u03d5 (4) Expression (4) represents the equations of the surface family in the bevel gears system, which envelope curve represents the tooth flanks of the generated gear" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001715_978-81-322-1859-3_15-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001715_978-81-322-1859-3_15-Figure2-1.png", + "caption": "Fig. 2 Ball burnishing tool 1 adapter cover, 2 ball: diameter 10 mm, 3 adapter, 4 spring seat, 5 spring, 6 casing, 7 shank, 8 allen screw", + "texts": [ + " Fuzzy model is considered as the most effective model for optimization of machining parameters for economical machining processes (Liu 2004). Basak and Goktas (2009) used a fuzzy model to achieve the optimum burnishing parameters for nonferrous components. The present research paper deals with the use of an adaptive neuro-fuzzy inference system (ANFIS) to model dry ball burnishing process and effectively employ genetic algorithm (GA) to search the optimal solution on the response surfaces modeled by neuro-fuzzy inference system. Figure 2 represents the 3-D drawing of ball burnishing tool that has been designed and fabricated for the present research work. A 10-mm-diameter ball made up of high carbon high chromium material has been used in the burnishing tool. Table 1 shows the ball burnishing process parameters considered in the present study based upon the results of pilot experiments and specifications of available lathe machine. Digital tachometer model DM 6234-P, Mitutoyo, has been used to measure the rpm. The feed of tool has been measured using a vernier depth gauge along with digital stopwatch" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002412_20140824-6-za-1003.01624-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002412_20140824-6-za-1003.01624-Figure1-1.png", + "caption": "Figure 1. (a) Prototype of the diver tracking platform developed in LABUST and (b) actuator configuration.", + "texts": [ + " Autonomous marine surface vessels are mostly being developed for science (e.g. exploration and observation, environmental data gathering and sampling, Caccia et al. [2007]), bathymetric mapping (e.g. Manley [1997]), defense (e.g. minecountermeasures Djapic and Nad [2010]), and general robotics research. A concise overview of some developed prototype vessels can be found in Manley [2008]. The Laboratory for Underwater Systems and Technologies (LABUST) at the University of Zagreb, Croatia have developed an overactuated autonomous surface marine platform (shown in Fig. 1) capable of omnidirectional motion while keeping desired heading. The main purpose of this platform is to serve as a diver tracking unit. LABUST has, in cooperation with their partners who are professional divers, demonstrated that cooperation between autonomous marine vehicles and divers is the most \u22c6 This work is supported by the Business Innovation Agency of the Republic of Croatia (BICRO) through the Proof of Concept programme, and by European Commission under the FP7\u2013ICT project \u201dCADDY \u2013 Cognitive Autonomous Diving Buddy\u201d under Grant Agreement Number: 611373", + " A step further in the development of the diver dedicated system is introducing an autonomous underwater vehicle capable of interacting with the diver. This step will be taken within the European FP7 project \u201dCADDY - Cognitive Autonomous Diving Buddy\u201d (611373). The autonomous surface platform, that carries the international flag marking underwater activity, is overactuated with 4 thrusters forming the \u201dX\u201d configuration. This configuration enables motion in the horizontal plane under any orientation. The current version of the platform is 0.35m high, 0.707m wide and long, as it is shown in Fig. 1, and it weighs approximately 25kg. The control computer (isolated from environmental disturbances inside the platform hull) executes control and guidance tasks (dynamic positioning, path following, diver following) and all the data processing. The complete control architecture is developed in the ROS environment. The research that involves the use of the described platform has been divided into the following phases: First phase: Simulations A number of simulation experiments where elementary control and dynamic positioning algorithms were tested under realistic simulation conditions", + " x\u0307 y\u0307 \u03c8\u0307 = [ cos\u03c8 \u2212 sin\u03c8 0 sin\u03c8 cos\u03c8 0 0 0 1 ] \ufe38 \ufe37\ufe37 \ufe38 R(\u03c8) [ u v r ] (2) Additional equation in the kinematic model is \u03c8\u0307 = r. The platform is overactuated, i.e. it can move in any direction in the horizontal plane by modifying the surge and sway speed, while attaining arbitrary orientation. Actuator allocation The actuator allocation matrix \u03a6 gives relation between the forces exerted by thrusters \u03c4i = [ \u03c41 \u03c42 \u03c43 \u03c44 ] T and the forces and moments \u03c4 acting on the rigid body. Actuator configuration of the autonomous surface platform for diver tracking is given in Fig. 1(b) where \u03b4 = 45\u25e6. The allocation matrix is given with (3). \u03c4 = [ cos 45\u25e6 cos 45\u25e6 \u2212 cos 45\u25e6 \u2212 cos 45\u25e6 sin 45\u25e6 \u2212 sin 45\u25e6 sin 45\u25e6 \u2212 sin 45\u25e6 D \u2212D \u2212D D ] \ufe38 \ufe37\ufe37 \ufe38 \u03a6 \u03c4i (3) For the low\u2013level, speed controller we choose a PI controller in the form \u03c4 = KP\u03bd (\u03bd\u2217 \u2212 \u03bd\u0303) +KI\u03bd \u222b (\u03bd\u2217 \u2212 \u03bd\u0303) dt+ \u03c4F (4) where \u03bd\u2217 = [ u\u2217 v\u2217 r\u2217 ] T are the desired linear and angular speeds of the platform, KP\u03bd = diag (KPu, KPv, KPr ) and KI\u03bd = diag (KIu, KIv, KIr ) are diagonal matrices with proportional and integral gains for individual degrees of freedom, respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001950_robio.2011.6181573-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001950_robio.2011.6181573-Figure4-1.png", + "caption": "Fig. 4. 4(a) is a square with all 3 acting forces. Red and green axises show the two axises in 2D. 4(b) is GWS in 3D of this grasp. The red and green vectors represent the forces axis, whereas the blue vector represents the torque axis. These three axises were also used as task wrench in our test. It is shown that these wrenches all have a intersection with GWS in 4(b) and 4(c). GWS in 4(d) is another GWS built from f1 and f2 without f3. A task wrench (\u22120.408, 0.408, 0.816), the gray line, can be performed only in certain region between the upper and lower bound.", + "texts": [ + " In the new coordinate system, we build a ray that has an opposite direction with the ray in original coordinate system, namely w\u2032 t = (0.408,\u22120.408,\u22120.816). So that a new linear programming problem is formulated. Matrix A is composed of vertexes of GWS\u2032\u2032. Its maximal feasible solution is (\u22123.78,\u22122.8,\u22121.89). We can get the intersection point (0.096,\u22120.096,\u22120.193). This is the point o\u2032\u2032 = (\u22120.408, 0.408, 0.816) in original coordinate system. After comparing o\u2032 and o\u2032\u2032 with task wrench wt, we have upper bound 1.35 and lower bound 1.1 of this grasp under this task wrench. The given task wrench is shown as a gray line in Fig. 4(d). In this paper, a novel algorithm has been introduced to compute the grasps for a specific task. The task was described by a TWS, which consists of the wrenches acting on the object. Ray intersection of the vertexes of the TWS with hyperplanes of GWS was solved as linear programming problem to find the scale factor, with which the TWS just fits in the GWS. The algorithm can also accept single task wrench for evaluation. Furthermore, both force-closure and non force-closure grasps are supported" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001961_icra.2014.6907673-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001961_icra.2014.6907673-Figure7-1.png", + "caption": "Fig. 7. Position constraint", + "texts": [ + " We deal with such limit as an inequality constraint, and it is derived as g(u\u0304) = F\u0304 ix 2 + F\u0304 iy 2 \u2212 (mg sin\u03b1)2 (10) Here, suffix i represents i th estimate in evaluation interval, \u03b1 [rad] is permissible maximum tilt angle of the helicopter. The relation between the external force and the tilt angle of the helicopter is shown in Fig.6. We introduce such input constraints to criterion as the penalty function Pu(u\u0304(t+ \u03c4)), and it could be obtained as Pu(u\u0304) = (max{0, g(u\u0304)})2. (11) Fourth and fifth term in (7) are state constraints for collision avoidance. In order to avoid the collision, the prohibited area is established around each helicopter (Fig.7). The inequality constraints could be obtained as gx(x\u0304) = r2 \u2212 { (x\u0304ij \u2212 x\u0304ik) 2 + (y\u0304ij \u2212 y\u0304ik) 2 } (12) Here, suffix j\uff0ck represent j th and k th helicopter, r is the radius of no entry area. To avoid the collision certainly, We introduce the constraints to criterion as the barrier function as follows: B(x\u0304) = 1 gx(x\u0304)2 (13) Using these constraints, collision avoidance may be achieved. However, the trajectory for the avoidance could not be optimized, and undesirable trajectory may be generated" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002362_i2mtc.2014.6860523-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002362_i2mtc.2014.6860523-Figure1-1.png", + "caption": "Fig. 1: Physical Setup of Testing Platform", + "texts": [ + " The multi-sensor measurements are meant to be correlated with the failures of the monitored BLDCM and bearing. The motor under test is a 17.5 turn BLDCM manufactured with hall sensor by Turnigy TrackStar. Moreover, the platform was conceived for non-invasive measurement techniques allowing quick motor replacement without unnecessary sensor reconfiguration. This ensures the extensibility of the platform to more complex testing scenarios. This setup was also designed to be used with minimal customization and easy to use. As shown in Figure 1, there is a force load cell which is in charge of measuring the perpendicular force applied on the shaft of motor for bearing testing. This testing takes place in the upcoming experiment. The main goal is to detect signatures on the signals that intensify over time (non-stationary). In this case, the force applied varies (which is the one applied perpendicular to the shaft of the motor). Figure 3 shows the main components of this testing platform and Figure 4 presents our data processing scheme for fault identification, which is described in the upcoming sections", + " The maximum sampling rate supported by the data acquisition modules for each of the insulated voltage channels is 51000 samples/s. Hence, for plotting analysis and convenience the sampling rate for the experiments was defined to be 50000 samples/s., this while having the motor rotating at 2000 RPM. A bearing test was performed by sampling data at the mentioned rotating speed and sampling rate. The bearing attached to the BLDCM was led to fail after increasing the perpendicular force on the gradually up to 460 Newtons. This force was measured by the load cell shown in Figure 1. Figure 6 shows the main measurements performed on each of the 18 sampling channels of 4.5 hours or 510 minutes of testing leading to the bearing failure. The rms of the perpendicular force value reflects how the perpendicular force on the bearings was increased from 360 N. to 464 N. Then at the end, after approximately 480 minutes of testing, the bearing failed and the speed shown by the Hall sensor channel decreased from 33.3 to 18 revolutions per second: decreasing the torque, sound pressure and vibration measured by the accelerometer; but still increasing the temperature and current on each phase" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002221_j.proeng.2011.08.084-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002221_j.proeng.2011.08.084-Figure1-1.png", + "caption": "Figure 1 Geometric relationship between vehicle kinetic parameters All the vectors are in terms of point C; CPO\u03b2 = \u2220 is the vehicle sideslip angle; v is the velocity at point C; K is the curvature of vehicle track at point C. The new non-holonomic constraint equation is:", + "texts": [ + " If the sideslip angle is not considered, the vehicle kinetic status can be expressed as . . . cos sin x v y v Kv \u03c8 \u03c8 \u03c8 \u23ab= \u23aa \u23aa = \u23ac \u23aa = \u23aa \u23ad (3) Where [ ], , Tx y \u03c8 stands for the position and direction vector of vehicle; ,x y are the coordinates of vehicle position; \u03c8 is the direction or the yaw angle (angle between X axle and longitude symmetrical line); K is the curvature of vehicle track; v is the vehicle speed. Usually, vehicle position is determined by the center of mass, as is shown by C in Figure1. Unless the steer angles of front wheel and rear wheel are in the same value but opposite direction, when the vehicle is steering around point P, PC line is not perpendicular with the vehicle longitude symmetric line, which means the velocity at C is not satisfied in Equation(3). This is because in Equation(3), sideslip angle is not considered. So, here Equation(3) is modified to take the sideslip angle into account: ( ) ( ) . . . cos sin x v y v Kv \u03c8 \u03b2 \u03c8 \u03b2 \u03c8 \u23ab= + \u23aa \u23aa = + \u23ac \u23aa \u23aa= \u23ad (4) ( ) ( ) . " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001388_047134608x.w4531.pub2-Figure24-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001388_047134608x.w4531.pub2-Figure24-1.png", + "caption": "Figure 24. Loci of the localized flux density in a three-phase induction motor stator core assembled from nonoriented silicon iron: (upper) fundamental (50Hz); (lower) third harmonic (150Hz) (core back flux density of 1.0T) (45).", + "texts": [ + " Stator cores are generally assembled from nonoriented steel unless the stator has such a large diameter that segmented cores can be assembled from grain-oriented steel to minimize losses. In small stator cores, the flux can be assumed to be almost wholly in the plane of laminations, so the distribution can be solved by 2D analysis, and in case of isotropic nonoriented steel, close agreement is usually found between computed and measured distributions. If grain-oriented steel is considered, its high anisotropymust be adequately catered for. Figure 24 shows typical flux distributions measured in the stator or a three-phase induction motor (45). The fundamental and third harmonic flux density loci are separated for convenience and both contain large components of rotational flux behind the teeth and slots. At the core back, the flux is mainly low in magnitude and circumferential in direction, whereas in the teeth it is high in magnitude and radial in direction. The loss distribution in a motor core can be easily related to the flux distribution" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002460_ichqp.2014.6842802-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002460_ichqp.2014.6842802-Figure1-1.png", + "caption": "Figure 1. PMSG cross section.", + "texts": [ + " In small applications the aggregate is very light, which means that the diameter of the generator should be increased to maximize the moment of inertia according to equation (1). 2 2 1 RmIc \u22c5\u22c5= (1) where Ic is the moment of inertia for a thin disc, m is the mass of the disc and R is the radius of the disc [5]. Considering that the magnets will have the same area and the same thickness, generators with round-shaped magnets will have a bigger diameter in comparison with the generators that will use trapezoidal magnets in the rotor construction. In figure 1 is presented the cross section of the model. The three phases of the stator are disposed as above, with three coils connected in series on each phase. Coreless stator implies that the magnetic flux on each phase is zero when the magnets are in the center of the coils, phase C in figure 1. The flux path is represented with green color in figure 2. II. 2D MAGNETIC FLUX SIMULATION The model of the generator has circular coils. Because of this, the analytical determination of the magnetic flux through the stator can be quite complicated. For this reason the magnetic flux was determined from the 2D simulation of the generator. The simulation was performed in Quick Field program [6]. For this analysis we have considered three pairs of poles of the generator and the distances: the distance between two magnets of a pair of poles is 15 mm and the distance between two poles situated on the same disc is equal with 17" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002923_gt2013-95585-Figure12-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002923_gt2013-95585-Figure12-1.png", + "caption": "Figure 12. COMPARISON OF AIR-RIDING AND CONTACTING LEAVES", + "texts": [ + "org/about-asme/terms-of-use where FDh is the hydrodynamic drag force obtained from the corresponding look-up table and \u00b5 is the coefficient of friction. However if FS, FA, and FLh fulfil the inequality in Eqn. (11), sufficient lift exists and air-riding will take place and an iterative process is required to determine the equilibrium air-riding gap G. To find this one must consider how the forces FS and FA change as a consequence of air-riding, which is a different condition than the low speed operation for which the empirical model was previously developed. From Fig. 12, which gives a frontal view of both an air-riding and non air-riding seal, it can be seen that it is most appropriate to calculate FS based on R = RR +G and FA based on R=RR. This is the case as the force due to leaf mechanical stiffness is a direct function of leaf tip position. However the aerodynamic force is a function of flow through the seal, which is predominantly a function of the radial distance between coverplate bore and rotor surface [10]. Consequently evaluating FA based on RR is more appropriate" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002278_amr.939.283-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002278_amr.939.283-Figure2-1.png", + "caption": "Fig. 2 Differential element used in slab analysis of cylinder compression forming process", + "texts": [ + " To minimize P , the optimum b can be determined by differentiating the formula of ypP \u03c3/ with respect to b and equating the derivative to zero and solving it. In analyzing the cylinder compression forming process using the UBM method, the present study utilizes the formulae derived by Avitzur [2,3] and Tsai et al. [5] (see Table 1). yp P \u03c3 2 22 2/3 12 , 133 ]1)1[( 1 3 2 o o b o yp H Rb B e b h mR B B P = \u2212 +\u2212+= \u03c3 b )3/)(3/2()/( 3/4 mHR m b oo + = yp\u03c3 1+ = n yp \u03c3 \u03c3 Slab Method. Assuming a static equilibrium condition in the r-direction (i.e., 0=\u03a3 rF ), the force balance on the differential element shown in Fig. 2 can be expressed as 02 2 sin2))(( =\u2212\u2212\u2212++ drrd d hdrhrdhddrrd rrr \u03b8\u03c4 \u03b8 \u03c3\u03b8\u03c3\u03b8\u03c3\u03c3 \u03b8 (1) Simplifying, the governing equation is obtained as hdr d r \u03c4\u03c3 2 = (2) Given a constant shear frictional stress of mk=\u03c4 , and a boundary condition of 0=r\u03c3 at Rr = , the compression force can be estimated in accordance with the von Mises yield criterion as follows: ) 3 2 (2 yp h Rmk RP \u03c3\u03c0 +\u00d7= (3) In analyzing the cylinder compression forming process using the Slab Method (SM), the present study utilizes the formulae derived by Hosford and Caddell[1] and Tsai et al" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002479_amm.668-669.729-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002479_amm.668-669.729-Figure5-1.png", + "caption": "Fig. 5. A cross section velocity contours of fan The left for the original fan and right for the optimized fan", + "texts": [ + "4(b) shows that the export static pressure of the optimized fan under 40% ~ 110% load conditions increase and is slightly lower than the original fan under 110% ~ 130% load conditions. But export static pressure fluctuations are relatively stable within the scope of the variable condition. The optimized fan is strong ability to variable conditions. Results analysis of the rated conditions In order to study the internal gas flow of the optimized fan volute, choose A cross section to analyze the static pressure fluctuation in the volute exit, the result is shown in Fig. 5. The volute exit velocity of left Fig. 5 is bigger than right Fig. 5, which suggests that the optimized fan volute exit of acceleration process is relatively flat. Therefore, the volute exit pressure fluctuations are relatively small. Analysis from the angle of volute shape suggests that the radius of curvature at the exit of the volute of the optimized fan is relatively small; a cross section of flow area near the volute outlet is bigger than the original fan. As a result, the optimized fan volute exit fluid resistance is small; the location of the fluid acceleration position relative to the original fan speeds up sooner" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure1.46-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure1.46-1.png", + "caption": "Fig. 1.46 Extension of the part in green outline", + "texts": [ + " Next, it was to create an elliptic shape to the top of the part with an elliptical extrusion, as shown in Fig. 1.44. The sketching and extrude tools were reused. For the next step, five rounds were applied as shown in Fig. 1.45, this is in order to not generate injury and discomfort to the patient and generate more similar forms to anatomical ones. Only the rounding tool was used here. 1 Comparative Study of Interferometry and Finite Element Analysis \u2026 29 Almost to finish, an extension was added to the same piece, as shown in Fig. 1.46 with a thickness of 4 mm, equal to that of the part. This was done so that the piece is close to that of the prosthesis. Sketching and extrusion tools were used. Finally, a rounding was generated to the outline of the piece to soften the plate, Fig. 1.47, and leave it ready for assembly with the prosthesis, see Fig. 1.48. The assembly was imported in STL format for 3D printing as shown in the results. 30 J. A. Beltr\u00e1n-Fern\u00e1ndez et al. Once the jaw was scanned, and in STL format available, PTC Creo Parametric was used to convert the part, by exporting the wrap-type level 10, to a solid piece" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003608_isciii.2011.6069753-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003608_isciii.2011.6069753-Figure4-1.png", + "caption": "Fig. 4. The results of the field computations", + "texts": [], + "surrounding_texts": [ + "HAVING DIFFERENT FAULTY CONDITIONS All the winding faults of a SRM cause unsymmetrical field distribution inside the machine. The best way to emphasize these changes is to perform a precise numeric field analysis of the SRM. The main data of the simulated SRM are: i.) Rated power 350 W ii.) Rated voltage 300 V iii.) Rated current 6 A iv.) Rated speed 600 1/min v.) Number of stator poles 8 vi.) Number of rotor poles 6 The cross section of the motor together with its pole's notations is given in Fig. 2. The numeric field computations were carried out by using the FEM based Flux 2D program package produced by Cedrat (France) [10], [11]. The simulations were performed for the healthy machine and for the following three winding fault conditions: i.) coil A having 20% of turns shorted ii.) coil A having 50% of turns shorted iii.) coil A having all its turns shorted. The most significant results are given in Figs. 3 and 4, where the flux lines obtained via field computations are shown for the A stator pole being aligned, half-aligned and unaligned Rare\u015f Terec et al. \u2022 On the Usefulness of Numeric Field Computations in the Study of the Switched Reluctance Motor's Winding Faults - 118 - relatively to the rotor poles, respectively for all the four conditions of the SRM in study. As it can be clearly seen in these figures the symmetry of the magnetic flux distribution is more and more lost as the severity of the simulated faults is increasing. The magnetic flux in pole A versus the angular displacement is plotted in Fig. 5. As it can bee seen in Fig. 5 as the number of shorted turns in the coils is increased the magnetic flux in pole A is decreasing. Of course the lower flux in one of the motor's pole will have influence also on the torque development capability of the machine. The plots of the torque developed by the SRM versus the angular displacement for different machine conditions are given in Fig. 6. The mean value of the developed torque during a displacement from the aligned to the unaligned position in the case of the healthy SRM is 3.41 N\u00b7m. When 20% of the turns - 119 - ISCIII 2011\u2022 5th International Symposium on Computational Intelligence and Intelligent Informatics \u2022 September 15-17, 2011, Floriana, Malta of one coil are shorted the torque development capability of the motor is reduced by 15% to 2.92 N\u00b7m. When only half of A coil's turns are working the developed torque of the SRM is 65% of the healthy machine's one (2.22 N\u00b7m). As also the testing of the flux differential winding fault detector given in Fig. 1 was proposed, the variation of the emf induced in the 100 turns search coil wound round pole A was plotted, too (see Fig. 7). As it was expected also this quantity is strongly influenced of the SRM's faults. The numeric field computations performed permitted also the computation of the voltage differences between the electromotive forces induced in two search coils from opposite poles (A and A'). This voltage difference is practically the one sensed by the flux differential detector. In Fig. 8 the input of the detector is plotted versus the angular position of the SRM for three conditions of the winding from pole A. All the results in Fig. 9 confirm the sensing capability of the detector in study. The voltages of hundreds of mV at the input of the operational amplifier are enough for an effective detection of the winding faults." + ] + }, + { + "image_filename": "designv11_84_0001715_978-81-322-1859-3_15-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001715_978-81-322-1859-3_15-Figure1-1.png", + "caption": "Fig. 1 Dry ball burnishing process", + "texts": [ + "eywords Burnishing process Neural network Adaptive neuro-fuzzy interface system Genetic algorithm Optimization Ball burnishing is one of the surface finishing processes used for producing surface with very high-quality texture (Fig. 1). Grinding, honing, and lapping are other processes, which are also used to produce fine surface, but these processes are chip J. Singh (&) P. S. Bilga Department of Mechanical Engineering, Guru Nanak Dev Engineering College, Ludhiana 141006 Punjab, India e-mail: er.jogindersing@gmail.com P. S. Bilga e-mail: psbilga@gndec.ac.in S. S. Khangura et al. (eds.), Proceedings of the International Conference on Research and Innovations in Mechanical Engineering, Lecture Notes in Mechanical Engineering, DOI: 10" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001236_chicc.2015.7260737-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001236_chicc.2015.7260737-Figure1-1.png", + "caption": "Fig. 1: Geometrical relationship between neighbor agents", + "texts": [ + " The relative position measurements are assumed available as piji(t) = pij(t)\u2212pii(t) where pij(t) is the j-th position measured from the i-th body coordinate frame. The ith local body frame is denoted as i\u03a3 for a convenience. The topology for relative position measurements is of directed one. The orientation is updated by the following control law \u03b8\u0307i(t) = u\u03b8i(t). The difference in orientations between two neighboring nodes is assumed computable by way of exchanging the relative bearing measurements \u03b4ji, where \u03b4ji is the j-th direction measured at the i-th body frame, which is depicted in Fig. 1. The difference in orientations between two neighboring nodes is denoted as \u03b8ji = \u03b8j \u2212 \u03b8i. It is shown in [8] that \u03b8ji = PV(\u03b8j \u2212 \u03b8i) = PV(\u03b4ji \u2212 \u03b4ij + \u03c0) (1) where PV(\u03b8j \u2212 \u03b8i) [(\u03b8j \u2212 \u03b8i + \u03c0)mode2\u03c0]\u2212 \u03c0. Thus, it is assumed that \u03b8ji is available to the i-th agent, when agents j are neighbor nodes of i, i.e., j \u2208 Ni. It is noticeable that the topology for orientation difference measurements is of directed one, while the relative angle sensing should be done in bidirectional way. The main goal of this paper is to achieve a desired formation among agents with respect to the leading reference node" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure19.13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure19.13-1.png", + "caption": "Fig. 19.13. The capturing approach of the floating, non-cooperative target system: (1) Close approach phase, (2) capture of the target by chaser with the aid of manipulator, and (3) pulling along guide elements of Docking Units and fixation with the help of locks of Docking Units", + "texts": [], + "surrounding_texts": [ + "A German-Russian space robotics project is going on: the TECSAS (TEChnology SAtellite for demonstration and verification of Space systems) project aims at the in-orbit qualification of the key robotics elements (both hardware and software) for advanced space maintenance and servicing system, especially w.r.t. docking and robot-based capturing procedures. It is planned, in close cooperation with the Russian Babakin Space Center, to perform such sensitive operations as rendezvous and close approach maneuvers, which will be necessary for further servicing activities. The Russian Multi-Purposes Orbital Boost Platform, as the base module for the mission, provides the insertion into initial parking orbit, as well as supports rendezvous and docking maneuvers. For docking and capturing operations we prefer to use our own robotics means (manipulator, controller, etc.), including MARCO to provide teleoperation and supervisory control from ground. The entire mission will be performed utilizing the following steps: far rendezvous, close approach, inspection fly around, formation flight, capture, stabilization and calibration of the compound as well as compound flight maneuver, manipulation of the target, active ground control via telepresence, passive ground control during autonomous operations (monitoring), and controlled de-orbiting of the compound. For the capturing of the floating target satellite, the control modes will be applied as developed in the ESS study and verified during the ETS-VII mission. Even a high-fidelity telepresence mode as demonstrated within ROKVISS could support the critical phases of the capturing process. After capturing, the manipulator can be used as an active damping system for compound 340 D. Reintsema, K. Landzettel, and G. Hirzinger stabilization issues: since the platform, including the robot system, and the captured target system build one compound system, the dynamic behavior of the complete system can be influenced by moving the robot. Also for de-orbiting maneuvers the manipulator can be used as a support system: it can be considered as a passive link, building the mechanical interface between the chaser and the target system. The geometry of the compound system can be influenced by the robots attitude. Additionally, the robot can be considered as an active link, controlling the thrusters vector pointing to the common center of mass." + ] + }, + { + "image_filename": "designv11_84_0001979_amm.284-287.461-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001979_amm.284-287.461-Figure3-1.png", + "caption": "Fig. 3 The configuration and generalized coordinates for gear pair", + "texts": [ + " No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 142.103.160.110, University of British Columbia, Kelowna, Canada-14/07/15,06:35:22) which the fixed reference frame, ZYX , is used to describe the system motion. Five degrees of freedom , , , , WV are considered at each nodal point of the shaft, where V and W are lateral displacements along Y and Z directions, respectively, and are rotational displacements, is the torsional displacement, and the axial translational vibration is neglected. Fig. 3 shows the configuration and generalized coordinates for the pair of gears [15], which are mounted on the shafts as shown in Fig.1. The teeth are regarded as flexible cantilever beams and are deformed by both the bending and shear. The mass, transverse mass moment of inertia, polar mass moment of inertia of the driving gear and driven gear are, respectively, 1dm , 1dDI , 1dpI , and 2dm , 2dDI , 2dpI . The motion of the gear set can be defined with ten generalized coordinates, which are the same as the node coordinates of two shafts, where the gears are mounted" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001808_s0263034613000608-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001808_s0263034613000608-Figure4-1.png", + "caption": "Fig. 4. (Color online) Schematic showing near-field region and far-field region in the plume generated from molten pool surface during laser irradiation. Dark spots inside the plume indicates high diffusivity of Al vapors compared to the remaining Mg vapors. Liquid droplets of molten materials are not presented.", + "texts": [ + " The ratio of Mg:Al was 10:1 in the base composition of AZ91DMg alloy, and vaporization rate of Mg was calculated as almost two orders of magnitude greater than that of Al according to Zhao and Debroy (2001) and He et al. (2004). Therefore, the combined effect of different vaporization rate and original chemical composition resulted in the great ratio of Mg:Al in the deposited plume, as shown in Figure 3 and Table 1. Distribution of vaporized Mg and Al species in the plume is further explained in Figure 4. At the bottom of plume, which was indicated as near-field region, density of vaporized species was very high and local vapor pressure exceeded ambient pressure when the plume generated from molten pool surface (Steen, 2003; Yoo et al., 2000). According to Table 1, height of the near-field region was estimated as 800 \u03bcm. When the plume expanded into surrounding environment, vapor pressure decreased significantly. Simultaneously, pressure difference between the core region and the atmosphere would lead to a far-field region at the top of the plume (Cieslak & Fuerschbach, 1988; Yoo et al" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003380_icisa.2014.6847401-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003380_icisa.2014.6847401-Figure11-1.png", + "caption": "Figure 11. Impending Hit view console IHVC, a,b and c are operator choice views (d): autonomous IHVC console shows zone around blue bar where near hit is detected.", + "texts": [], + "surrounding_texts": [ + "An autonomous virtual visual perception technique with on-line hit susceptibility assessment has been developed. Modeling of precise shape of real workspace object is of utmost importance and necessitates use of accurate CAD for high fidelity STS formation. The method is efficient and amenable to automated setup from CAD models without any shape constraint. By doing away with search of \u2018hit\u2019 it achieves bounded deterministic computation load that is essential in real time working . References [1] J. K. Mukherjee et. al. \u201cRemote managed Teleprobing for nuclear Applications \u201d Proc, Int. conf.-on Peaceful uses of Atomic Energy PEACEFUL ATOM (nuclear instrumentpp154-169)Sept29-Oct.3, 2009 New Delhi [2] J. K. Mukherjee et al, \u2018Machine Intelligence Tools for supporting metrics on active structural elements \u201d. Int.. Conference- SMiRT 21, New Delhi, Nov. 6-21 2011 [3] J. E. Speich, \u201cMethod for Simultaneously Increasing Transparency and Stability in Bilateral Telemanipulation\u201d. IEEE Int. Con. Robo. Auto., pp. 2671-76, April 2000. [4] G. R. Hopkinson, A. Mohammadzadeh, \u201cRadiation Effects In (CCD) Imagers& CMOS Active Pixel Sensors\u201d J. High Speed Elect and Sys., Vol.14, Issue 2, June 04, ISSN: 0129-1564 [5] Oren Tropp et al \u2018A fast triangle to triangle intersection test for collision detection\u2019 Comp. Anim. Virtual Worlds ,Wiley Inter Science (www.interscience.wiley.com). DOI: 10.1002/cav.115 [6] J.D. Foley et al. Computer Graphics: Principles and Practice (2nd Ed.), Addison\u2013Wesley 1990 ISBN-10: 0201848406 [7] Peter J Kovach Inside Direct3D. Microsoft Press.(2000) ISBN-13: 978-0735606135 [8] \u2018Mental Vision: Computer Graphics Platform for V.R. Science Science and Education\u2019\u00e9cole Polyt. F\u00e9d\u00e9rale De Lausanne [9] Spong, Huthinson, Vidysagar ; \u2018Robot modeling and Control\u2019 Wiley, 2006. \u201cForward and inverse kinematics\u201d page 65-100 [10] M. Lin et al Collision detection between geometric models: A survey. In Proc. of IMA Con.- Math. of Surfaces, 1998. http://www.cs.unc.edu/ dm/collision.html. [11] Y. Zhou and S. Suri. Analysis of a bounding box heuristic for object intersection. Jour. ACM, 46 no. 6:833\u2013857, Nov. 1999. [12] Y. Koren, et-al, Potential Field Methods and Their Inherent Limitations Proc. IEEE Conf. on Robotics & Auto., California, April 7-12, 1991, pp. 1398-1404 [13] S . Payandeh and Z. Stanisic.\u201dOn Application of Virtual Fixtures.\u201d X Int. Sym. On Hap. & Teleoperator Systs., 2002 [14] J.K. Mukherjee, \u2018Synthesized Transduction for Proximity Sensing in Tele-operated Systems\u2019 Intl. Conf. on Intelligent sensing ICST2012 , Dec2012 , Kolkata India. Pg 595-599 [14] J.K. Mukherjee, \"AI Based Tele-Operation Support Through Voxel Based Workspace Modeling and Automated 3D Robot Path Determination\", IEEE Conf - Convergent Tech. AsiaPacific Region, vol. 4 of 4, Oct.2003, pg.305-3092007. [15] Kaufman, A., Shimony, E., \u201c3D Scan-Conversion Algorithms for Voxel Based Graphics\u201d, Proceedings of the 1986 workshop on Interactive 3D graphics, pp. 45-75, 1987. [16] H. Samel, \u201cDesign and analysis of spatial data structures in Computer graphics, Image processing and GIS\u201d, Addison Wesley, Reading M. A. 1990" + ] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure2.11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure2.11-1.png", + "caption": "Fig. 2.11 Distribution of the maximum principal stress from the mandible FEA test", + "texts": [ + " Surface displacement points were obtained by GOM and compared with those of the actual experimental photo-elastic test. Interestingly, the maximum displacement occurred at the point of bone reabsorption, just where the fracture of the model was produced, see Fig. 2.9. Figure 2.10 shows the graph of time versus deformation percentage and the most critical point yields a maximum value of 14.402% in respect of the Y-axis. The FEA of the mandible reported a maximum stress at the mandible ramus, see Fig. 2.11, and the maximum von Mises stress occurred mainly at the condylar area of the mandibular ramus. At the mandibular osteotomy, see Fig. 2.12, the stress concentration occurred in areas of decreased bone geometry, and also in nearby areas to the simulated bone cut. In reference to the generated stresses with the traditional fixation without implant, and those with fixed screws to the sagittal osteotomy, there was higher stress at the more distant screw to the mandible ramus, localized at the mandibular body where the mandible thickness is significantly reduced, see Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001574_978-3-642-21111-9_5-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001574_978-3-642-21111-9_5-Figure1-1.png", + "caption": "Fig. 1. free-floating space manipulator system", + "texts": [ + ") and choose the controller u = KP2Lx\u03033 + \u2212L\u03a9\u0302m(z\u03021, z\u03022, z\u03023, x1, qd, q\u0307d, q\u0308d, \u02d9\u0302z3, z\u03022, z\u03023, uf) (24) and we employ the same idea of \u03a9(.) in Eqn(11).The stability proof relies on choosing the Lyapunov function as [10] V = 1 2 [rT x\u0303T 3 ] [ M L ] [ r x\u03033 ] + 1 2 tr [ \u0398\u0303T \u0398\u0303T m ] [ \u039b1 \u039b2 ] [ \u0398\u0303 \u0398\u0303m ] (25) and fuzzy adaptive tunning algorithms are \u0398\u0307 = \u039b1yrT \u2212 Kw\u039b1 \u2225 \u2225 r x\u03033 \u2225 \u2225\u0398 \u0398\u0307m = \u039b2ymx\u0303T 3 \u2212 Kw\u039b2 \u2225 \u2225 r x\u03033 \u2225 \u2225\u0398m (26) To validate the proposed scheme, simulation of motion along desired joint trajectories were performed with a planar two-link free-floating space manipulator driven by two dc motors(Fig.1). m0 is the base mass(Kg), m1, m2 are the link masses(Kg), 0, 1, 2 are the link lengths(m). The nominal values of the parameters are 0 = 1, 1 = 2, 2 = 2, and m0 = 400, m1 = 4.0, m2 = 8.0, the nominal parameters of joint motors for space manipulator are L = diag(0.04, 0.04), R = diag(70, 70), KB = diag(0.145, 0.145), KT = diag(1.38, 1.38). However, the actual parameters of the manipulator and its driving joint motors are assumed to be m0 = 300, m1 = 6.0, m2 = 6.0, and L = diag(0.05, 0.05), the other parameters are assumed to be same as the nominal parameters" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001096_978-81-322-1035-1_30-Figure30.1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001096_978-81-322-1035-1_30-Figure30.1-1.png", + "caption": "Fig. 30.1 Aircraft axes, forces and moments", + "texts": [ + " In order to reduce the complexity of this mathematical modeling problem, some simplifying assumptions may be applied. First, it is assumed that the aircraft motion consists of small deviations from its equilibrium flight condition. Second, it is assumed that the motion of the airplane can be analyzed by separating the equations into two groups [2]: \u2022 Longitudinal Equations: composed by X-force, Z-force and pitching moment equations. \u2022 Lateral Equations: composed by Y-force, yawing and rolling moment\u2019s equations. Figure 30.1 shows the above mentioned aircraft reference axes, forces and moments. The longitudinal-directional motion of an aircraft disturbed from its equilibrium state is pitching. The longitudinal-directional motion, specifically the pitch motion, is the movement intended to be simulated on the proposed test platform as demonstration example. To simplify the studied case, it was chosen to address specifically the pitch motion in this work. Even though this simplifies the dynamic model equations, the platform functionality and characteristics, are very well demonstrated" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003883_icar.2011.6088592-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003883_icar.2011.6088592-Figure3-1.png", + "caption": "Fig. 3. Foot trajectory planning of a bipedal robot.", + "texts": [ + " Based on the de nition of falling, the angle parameter \u03c6 is directly related to the falling of the robot system and it is considerable as a factor to identify the current state of falling. Table I shows some possible states in terms of falling for the robot system. In addition, the yaw angle parameter \u03b1 can provide us the direction of falling. Since the falling phenomenon may cause the imbalance of walking robots, it is valuable to consider the concept of falling as a measure for an optimal footstep planning. The foot trajectory planning problem considering in this paper can be depicted as Fig. 3 and its goal is actually to determine the optimal foot location supporting for the robot to follow the prescribed trajectory at the COM space. This approach is meaningful for the task-based mobile manipulation of bipeds or humanoid robots, and it is different from the conventional ZMP-based methods [6][7]. Our approach can be classi ed as the following two strategies. One is to determine the next footstep location of the swing leg which is actually moving for recovering the imbalance in a bipedal walking", + " Accordingly, if the goal position of the swing foot is determined at each starting moment of walking, the intermediate location from the starting position to the goal position is made by the cubic trajectory planning method given by (9). There exists a preferable walking style for bipedal robots to achieve a speci c mobile manipulation task. It is highly dependent on the desired behavior of the body. So, we predetermined an exemplary motion trajectory at the COM space for a bipedal manipulation as shown in Fig. 3 and then tried to plan the foot trajectory supporting for the desired COM trajectory. Especially, for the motion trajectory, we considered a human-like sequential walking style as follows: 1) START. 2) Homing action for the initial upright posture standing at the state of dual support(DS). 3) Shift the body to right at the state of DS for the shift time tshift. 4) Move the left foot by keeping the right foot support(RS) for T \u2212 tshift, where T is the time period for one step motion. 5) Shift the body to left at the state of DS for tshift" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002136_s1068798x14070090-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002136_s1068798x14070090-Figure1-1.png", + "caption": "Fig. 1. Experimental apparatus in [2]: (a) basic beam; (b) its analog.", + "texts": [ + " At the same time, the authors assumed that the contact pliability may be disregarded if the surface quality of the rod is no worse than class 5 (with microprojections of mean height less than 2.5\u20135.0 \u03bcm) and with a work ing load no greater than 0.5\u20130.6 of the tensile force. In the present work, we consider in more detail the influence of contact pliability with different surface quality of the contact surfaces on the load distribution within a complex screw joint under a central force and a skew torque, when the flanges are flexurally rigid. The rigidity of flanged joints tightened by screws was measured experimentally in [2]. Specifically, the flexure f of the basic beams (Fig. 1a) and the flexure fa of their analogs (Fig. 1b) was measured under a load Fr = 1.2 kN. The basic beam consists of two parts tightened by four M8 screws; the tightening force on each screw is Fti = 7 kN; the total length L = 450 mm; the distance from the force Fr to the butt joint is l0 = 280 mm; the flange diameter D = 110 mm; the rod diameter dro = 20 mm; and the diameter of the circle where the screws lie is Dscr = 80 mm. In other words, the distance between the screws lscr = 57 mm. Struc tures with flanges of different thickness are tested: h = 10, 16, and 25 mm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003727_2015-01-1524-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003727_2015-01-1524-Figure4-1.png", + "caption": "Figure 4. 6- DOF Tire Test Rig", + "texts": [], + "surrounding_texts": [ + "As we know, curvature factor dominate tire lateral force characteristics between linear region and saturation region. Compared to tire cornering stiffness factor and friction coefficient factor, it is less important. So in this paper, a curvature model used in UniTire [4] is applied. The parameters are estimated by tire lateral force data under low-load condition. (27) Where, Ey is the tire lateral force curvature factor, PEy1 , PEy2 are curvature parameters." + ] + }, + { + "image_filename": "designv11_84_0001195_978-81-322-1656-8_29-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001195_978-81-322-1656-8_29-Figure1-1.png", + "caption": "Fig. 1 Schematic views of two typical GFBs a Leaf type GFB, b Bump type GFB [10]", + "texts": [ + "1007/978-81-322-1656-8_29, Springer India 2014 331 h Film thickness (m) ho Initial bump height (m) hmin Minimum film thickness (m) H Non-dimensional film thickness Hmin Non-dimensional minimum film thickness i, j Grid location in circumferential and axial directions of FDM mesh Kf Bump foil structural stiffness per unit area \u00f0N/m2\u00de lo Half bump length (m) ls Length of segment between the bumps (m) L Bearing length (m) m Number of divisions along j direction of FDM mesh n Number of divisions along i direction of FDM mesh O Centre of bearing O0 Centre of journal p Hydrodynamic pressure in gas film \u00f0N/m2\u00de pa Atmospheric pressure in gas film \u00f0N/m2\u00de P Arithmetic mean pressure along bearing length \u00f0N/m2\u00de P Non-dimensional hydrodynamic pressure p Non-dimensional arithmetic mean pressure along bearing length R Radius of journal (m) s Bump foil pitch (m) S Compliance number: pa CKf t Time (s) tb Bump foil thickness (m) tt Top foil thickness (m) wt Top foil transverse deflection (m) W Non-dimensional top foil transverse deflection W0 Steady state load carrying capacity (N) W0 Non-dimensional steady state load carrying capacity x; y; z Coordinate system on the plane of bearing Z Non-dimensional axial coordinate of bearing: z R a Compliance of the bump foil \u00f0m3=N\u00de : 1 Kf e Eccentricity ratio K Bearing number: 6lx pa R C 2 l Gas viscosity \u00f0N - s/m2\u00de / Attitude angle \u00f0rad) h Angular coordinate of bearing (rad): x R ho Half bump angle (rad) t Poisson\u2019s ratio s Non-dimensional time: xt x Rotor angular velocity \u00f0rad/s) Dh;DZ Non-dimensional mesh size of FDM mesh Gas foil bearings (GFBs) are compliant surface, self-acting hydrodynamic bearings typically constructed from several layers of sheet metal foils which support a rotor by any combination of bending membrane or elastic foundation effects. Gas foil bearing technology has made significant progress over the last 40 years and fulfills most of the requirements of novel oil-free turbo-machinery by increasing tenfold their reliability in comparison to rolling elements bearings [1, 2]. GFBs compliant surface provides bearing structural stiffness and comes in several configurations such as the leaf type (Fig. 1a), bump type (Fig. 1b), and tape type, among others. The underlying compliant structure (bumps) provides a tenable structural stiffness source [3, 4] resulting in a larger film thickness than the rigid wall bearings [5, 6] enabling high-speed operation and larger load capacity including tolerance for shaft misalignment [2]. Also damping of coulomb type arises due to the relative motion between the bumps and the top foil, and between the bumps and the bearing wall [7]. Foil Bearings generally operates with ambient air" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002460_ichqp.2014.6842802-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002460_ichqp.2014.6842802-Figure3-1.png", + "caption": "Figure 3. Magnetic flux density.", + "texts": [ + " The simulation was performed in Quick Field program [6]. For this analysis we have considered three pairs of poles of the generator and the distances: the distance between two magnets of a pair of poles is 15 mm and the distance between two poles situated on the same disc is equal with 17.5 mm. These dimensions are the dimensions of real generator. In the construction of the generator model we have chosen round-shaped magnets with diameter \u03c6=18 mm and thickness 5 mm. The magnets are made from NdFeB35. The selected area in figure 3 represents the area of the stator that was analyzed. For the selected area we obtained the next average values for magnetic flux density: \u2022 Baverage = 0.3876 T; \u2022 \u03c6 = 89.537 degrees; \u2022 Baverage X = 0.0031 T; \u2022 Baverage Y = 0.3875 T, where \u03c6 is the angle between Baverage and the X axis (tangent to the magnet surface), Baverage is the average magnetic flux density, and Baverage X and Baverage Y are the average values for magnetic flux density on X and Y axis. This average value is the value obtained considering the harmonic content" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003377_icra.2013.6630735-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003377_icra.2013.6630735-Figure4-1.png", + "caption": "Fig. 4. Data set I (left) and II (right): The lines/ curves between the blue Location Areas refer to trajectories of lifted objects (red) and trajectories of pushed objects (green). The robot\u2019s base positions are symbolized in yellow.", + "texts": [ + " The final points px do not need to stay on a circle, since they can be placed arbitrarily within each sphere Sx. In the case of a pushed object, we have to ensure, that each point stays at the original height above the plane on which the object is pushed. Therefore, the sphere Sx is reduced to a circle at the desired height above the plane. Our experiments are based on a object manipulation scenario. The used data consists of two data sets, which were extracted in [2]. Both data sets are illustrated in Fig. 4. A. Implementation The implementation is done in C/C++. The stochastic approach for global minimization was presented in [15]. We use the implementation by Oliver Ruepp [16]. The manipulator consists of three rotational joints perpendicular to each other (D = 3). All links have a length of 300 mm (DH-Parameter: d1, d2, a3) The possible change of a position on the path is limited to the corresponding sphere Sx. We allow each joint d to change its position partially within the sphere. Hence, the contributed part of a joint is limited to rx/D" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003351_amr.712-715.1420-FigureI-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003351_amr.712-715.1420-FigureI-1.png", + "caption": "Fig1 Institution diagram of the pumping unit Fig 2 Agency simplified model of the pumping unit L1 is a crank length, L2 is the length of the link, L3 is the length of the arm of the amusement Pumping after, L4 is the linear distance from the support center of the walking beam to the center of the speed reducer output shaft [4] .", + "texts": [], + "surrounding_texts": [ + "The primary members of the beam pumping unit is shown in Figure 1, 1 - donkey head, 2 - beam, 3 - beam support ,4- bracket, 5 - Tour beam assembly ,6 - connecting rod assembly, 7- crank device, and the simplified model is shown in Figure 2, All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 128.210.126.199, Purdue University Libraries, West Lafayette, USA-24/05/15,04:10:28)" + ] + }, + { + "image_filename": "designv11_84_0003021_0954406215582015-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003021_0954406215582015-Figure1-1.png", + "caption": "Figure 1. Rolling element bearing structure and load distribution.", + "texts": [ + " Envelope spectrum analysis, LDA, and CHMM have been applied in many fields, so they are briefly introduced here. Analysis of rolling element bearing vibration is illustrated in the next section, and the subsequent section explains the envelope spectral energy. LDA and CHMM are presented in the following sections, respectively. \u2018\u2018Bearing faults diagnosis based on LDA and CHMM\u2019\u2019 section illustrates how to identify the bearing defects based on bearing envelope spectral energy and CHMM. Rolling element bearing vibration analysis Figure 1 shows the structure and load zone of the rolling element bearing. D and d are the pitch diameter and the ball diameter, respectively, is the bearing contact angle, and Z is the number of rolling bodies. Vi, Vc, and Vo are linear velocity of inner race, cage, and outer race, respectively. As can be seen from the figure, there are five characteristic frequencies of the bearing elements. When fault occurs on a particular bearing element, it will produce successive impulses which may cause resonances in bearings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure6.18-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure6.18-1.png", + "caption": "Fig. 6.18 Original piece previous to the study", + "texts": [ + " The program performs a pre-verification where we observe if the necessary information is complete or some adjustments are needed, see Fig. 6.15. 10. A previous visualization of the piece is generated, see Fig. 6.16. 11. Once the parameters have been established a fast test is run to generate the iterations. It is important to understand that the processing is made in the cloud and hence it has to be uploaded. choose which option is optimal, see Fig. 6.17. As we have previously mentioned, the design requirements were plotted in Table 6.1 by Martinez, and so the outcome of that is a tailored device (Fig. 6.18) which is subject of improvements that is why the aim of this paper is to find that the proposed piece (arch-support, proposal 2) would offer a well balance between its weight and structural stiffness. The distractor requires a left and right version of the above-mentioned piece, and both parts represent 40.33% of the total weight of the assembly, expressed in a mass with a Ti6A1-4 V alloy (403.7 gm, depicted in Table 6.2), this piece has the largest mass in respect with the other components and because of that, it is the component that will be processed in the generative design study" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003420_j.mspro.2014.06.203-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003420_j.mspro.2014.06.203-Figure2-1.png", + "caption": "Fig. 2. Calculation schemes of two stages of pit formation : \u0430) shear stage, b) opening stage; \u0412 \u2013 counterbody motion direction; f, fc \u2013coefficient of friction between contact bodies and crack edges, respectively ; Lsl, Lst, Lop \u2013 regions of sliding with friction, sticking, opening respectively.", + "texts": [ + " The distinctive features of the typical defects formation in rolling bodies (such as pitting and spalling) are studied depending on the operating characteristics of rolling couple and characteristics of fatigue crack growth resistance of materials. Now let us consider the case of unidirectional rolling with sliding. The contact influence of the counterbody is modeled by the repeated translational motion of Hertzian normal stresses (with tangential component reflecting the action of sliding friction forces) along the edge of the half plane (Fig. 2). The damaged body is a follower. In the contact region these forces take the form 2 0 2 0 xxapxp , xfpxq ; (4) outside this region they are zero; where 2a is the length of the contact region, p0 is the maximum pressure in the center of the contact region. Pitting appears practically on the whole rolling couples surface. The depth and periodicity of rails surface grinding as well as their contact durability in engineering practice are determined by the crumbling pits depth and the number of cycles before their appearance", + "Way (Way (1935)) expressed the assumption that pitting appears only in the presence of lubricant during rolling bodies contact. Assuming this postulate, the paper considers the growth of edge cracks in rolling bodies under boundary lubrication. Let us consider that a cracked body is a follower. Properly in such a body the shear edge cracks are oriented in the direction of counterbody movement, thus assisting the penetration of the lubricant into the crack under rolling. At the same time the lubricant decreases the friction between the crack edges at the stage of shear (Fig. 2a). Then, under rolling the lubricant can divide the cracks edges and wedge them. Later the transition from mode II fracture to mixed mode I+II fracture occurs. The wedging affect of the lubricant (or other liquid) on the crack edges is modelled by the normal pressure distributed uniformly (Panasyuk et al (1995)) or linearly (Datsyshyn et al (2011)) over the crack length (Fig. 2b). So, let us assume that the beginning (the first stage) of pit formation is the propagation of the rectilinear and inclined edge crack on its elongation by transversal shear mechanism. This assumption was verified by the experimental and theoretical calculations. Using SIE solutions to the contact problem for a semi-plane with an edge crack, which edges can get in contact (sliding with friction, sticking, opening) under the influence of moving Hertzian loading, the crack edges contact map are constructed and also SIF KII and KII range are investigated" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002157_detc2013-12233-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002157_detc2013-12233-Figure7-1.png", + "caption": "Fig. 7. COORDINATE SYSTEMS FOR CUTTING EDGES FOR (A) STRAIGHT-LINED BLADE EDGE (B) CIRCULAR BLADE", + "texts": [ + " As shown in Fig. 6, the cutter head has 0z blade groups. Each blade group has inner and outside blades for cutting left and right flanks of gear, respectively. b\u03b1 is the pressure angle, h\u03b1 is the hook angle, and i\u03b2 is the initial setting angle. 0\u03b4 is the offset angle to let the normal of plane T align the cutting direction. 3 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 02/01/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use EDGE. As shown in Fig. 7, the profile of cutting edge can be straight-lined or circular ( ( )l lr ) with a circular-arc tip fillet ( ( )f lr ). The later is applied to profile crowning. The position vector of the blade edge may be specified using homogeneous coordinates as [ ]( )= ( ) 0 ( ) 1 T l l lu x u z ur (5) For the straight-lined cutter edge, ( )( ) 2 21 1 ( ) ( ) 1 1 2 2 ( ) ( cos )( ) sin ( ) cos ( ) sin fl l cf fl b l f l b l cf f x u x ux u u z u u z u z u \u03c1\u03b1 \u03b1 \u03c1 \u23a7 = \u00b1 \u2212\u23a7 = \u00b1\u23aa \u23aa \u23a8 \u23a8 = = +\u23aa \u23aa\u23a9 \u23a9 (6) where tan tan( / 4 / 2)cf r b f b cf r f x h z h \u03b1 \u03c1 \u03c0 \u03b1 \u03c1 = + \u2212\u23a7\u23aa \u23a8 = \u2212\u23aa\u23a9 For the circular cutter edge, ( ) 1 1 ( ) 1 1 ( ) 2 2 ( ) 2 2 ( ) ( cos( )) ( ) sin( ) ( ) ( / cos ) ( ) sin l l cl c b l l cl c b f l cf f c c f l cf f x u x u z u z u x u x u z u z u \u03c1 \u03b1 \u03c1 \u03b1 \u03c1 \u03c1 \u03c1 \u03c1 \u23a7 = \u00b1 + \u2212 \u23aa = \u2212 \u2212\u23aa \u23a8 = \u00b1 \u2212\u23aa \u23aa = +\u23a9 (7) where 2 2 cos( ), sin( ) ( cos ( ) ( sin( )) ) cl c b cl c b cf c b c f r f c b cf r f x z x h z h \u03c1 \u03b1 \u03c1 \u03b1 \u03c1 \u03b1 \u03c1 \u03c1 \u03c1 \u03c1 \u03b1 \u03c1 = \u2212 =\u23a7 \u23aa\u23aa = \u2212 \u2212 + \u2212 \u2212 \u2212\u23a8 \u23aa = \u2212\u23aa\u23a9 f\u03c1 and c\u03c1 are the fillet radius and the curvature radius of blade, respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure28.3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure28.3-1.png", + "caption": "Fig. 28.3. Arm angle \u03b8 definition", + "texts": [ + "1) where p = [px py pz]T is the position vector of the end effector and Q = [\u03b5T \u03b7]T is the unit quaternion representing the orientation of the end effector in which \u03b7 and \u03b5 denote scalar and vector part. Using unit quaternions to represent orientation errors allows global parametrization of orientation not suffering from representation singularities [13]. The arm angle \u03b8 proposed in [11, 12] represents the orientation of the arm plane determined by the centers of the shoulder s, elbow e, and wrist w, see Fig. 28.3. It is a kinematic function of the joint angle vector q, which gives a measure of the following physical mobility: if we hold the shoulder s, the wristw, and the end-effector t in fixed positions, the elbow e is still free to swivel about the axis from the shoulder s to wrist w. The arm angle on the circle can be defined by an interior angle between the planes se0w and sew. The reference position e0 of elbow is chosen such that e0 is on the plane which is spanned by s, w and the other shoulder position s\u2032 of the dual arm manipulator so that \u03b8 is equal to zero when e = e0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure2.6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure2.6-1.png", + "caption": "Fig. 2.6. Admittance control scheme", + "texts": [], + "surrounding_texts": [ + "An increase of the master admittance decreases the deviation between the haptic interface closed loop dynamics and the target impedance rendered by the relation between q and hr. One has, however, to consider, that the minimum closed loop mass and inertia of the haptic interface is bounded by stability (see Sec. 2.5). A common worst case scenario is free space motion of the teleoperator producing hr = 0 for arbitrary device motions. In that case the minimum mass and inertia has to be solely provided by the master admittance. The implementation of the ViSHaRD10 master admittance is illustrated in Fig. 2.7. The rendering of the inertia and rotational damping is based upon the well known Euler\u2019s dynamical equations of rotation. The indices B and E indicate the corresponding quantity to be defined respective the coordinate system {B} and {E} defined in Fig. 2.4. The rotation matrices EBR and B ER map vectors from the base to the end-effector coordinate system and vice versa. The virtual mass, inertia, translational, and rotational damping is defined by the matrices Mtrans, Mrot,KD,trans, and KD,rot, respectively. Possible extensions to this master dynamics are e. g. the implementation of virtual spring forces to constrain the workspace of the haptic interface and the teleoperator." + ] + }, + { + "image_filename": "designv11_84_0003721_detc2014-34297-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003721_detc2014-34297-Figure6-1.png", + "caption": "FIGURE 6. Kinematic chains corresponding to cases from 1 to 6 of Table 1.", + "texts": [ + "org/about-asme/terms-of-use S1,u\u03021 \u2295S2,u\u03022 The kinematic chain that is shown in Figure 5, is generated from the direct sum of two subspaces S1,u\u03021\u2295S2,u\u03022 , \u2014which is obtained from the serial chain of the form h1,1 P,u\u03021,1,p1 \u2212h1,2 Q,u\u03021,2,p2 \u2212h1,3 R,u\u03021,3,p3 \u2212 h2,1 M,u\u03022,1,p4 \u2212h2,2 N,u\u03022,2,p5 \u2014, and represents a screw systems of locally constant rank with additional constraints. The results of this section are shown in Table 1. This table shows, if there are, additional conditions to each of the subspaces S1,u\u03021 and S2,u\u03022 , the subspaces resulting after applying the relationships as well as the result of the direct sum of subspaces. Figure 6 shows the cases from 1 to 6 of Table 1. In each Figures 6(a), 6(b) and 6(c), the first three kinematic pairs generate in a different way the subalgebra gu\u03021 . Figure 6(a) shows especifically the cases 1 and 4; here the unit vectors u\u03022,1 and u\u03022,2, are parallel. Similarly, Figure 6(b) shows specifically the cases 2 and 5. Finally, Figure 6(c) shows specifically the cases 3 and 6. Figure 7 shows the cases from 7 to 12 of Table 1. In each Figure 7(a), 7(b) and 7(c), the first three kinematic pairs generate in a different way the subalgebra yu\u03021,p. Figure 7(a) shows specifically the cases 7 and 10; here the unit vectors u\u03022,1 and u\u03022,2, are parallel. Similarly, Figure 7(b) shows specifically the cases 8 and 11. Finally, Figure 7(c) shows specifically the cases 9 and 12. Figure 8 shows the cases from 13 to 18 of Table 1. In each Figure of 8(a), 8(b) and 8(c), the three first kinematic pairs generate in a different way the subspace S1,u\u03021 < xu\u03021 ", + " Figure 8(a) shows specificaly the cases 13 and 16; here the unit vectors u\u03022,1 and u\u03022,2, are parallel. Similarly, Figure 8(b) shows specifically the cases 1 and 17. Finally, Figure 8(c) shows specifically the cases 15 and 18. Figures 6(a), 7(a) and 8(a), show different ways to generate the subspace S2,u\u03022 , when p4, p5 6= \u221e and u\u03022,1||u\u03022,2. Similarly, Figures 6(b), 7(b) and 8(b), show different ways to generate the subalgebra cu\u03022 , when u\u03022,1||u\u03022,2. Finally, Figures 6(c), 7(c) and 8(c), show different ways to generate the subspace S2,u\u03022 , in Figure 6(c), S2,u\u03022 < xu\u03022 , in Figure 7(c), S2,u\u03022 < gu\u03022 , while in Figure 8(c), S2,u\u03022 < yu\u03022,p. It should be mentioned that, both the Table 1 and Figures from 6 to 8, show only the kinematic chains where the first three kinematic pairs generate the subspace S1u\u03021 , while the last two kinematic pairs generate the subspace S2u\u03022 . However, besides the permissible permutations of the kinematic pairs that form the kinematic chains S1,u\u03021 and S2,u\u03022 proved in the propositions 4 and 5, the following proposition shows that many other permutations of the kinematic pairs are permissible" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001950_robio.2011.6181573-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001950_robio.2011.6181573-Figure1-1.png", + "caption": "Fig. 1. 1(a) shows computation of task-oriented quality measurement of force-closure GWS and single task wrench with ray shooting. 1(b) shows force-closure GWS and TWS, the quality is determined by scale factor of TWS in GWS, which in this case is determined by scale factor of v1.", + "texts": [ + " After solving the linear programming problem as reviewed above, the hyperplane Hk that has an intersection with the ray is found. The scale factor of the ray intersects with the hyperplane Hk can be computed by \u03c1 = \u2212ck nk\u00b7wt . As the origin is contained in a force-closure GWS, a ray from origin towards arbitrary direction has always an intersection with the convex hull. Thus force-closure grasp is feasible for any task wrench, however with different qualities, which is represented by length of line segment of the ray inside the GWS, the value of \u03c1. As shown in Fig. 1(a), wt1 and wt2 are two single task wrenches from the origin O. Their intersections with the GWS are computed to obtain the task-oriented qualities of this grasp of these two tasks respectively. The circle represents the largest sphere inside the GWS. The radius of the sphere is smaller than the lengths of both ray intersections. This way, quality for individual task wrenches can be evaluated. The GWS of a non force-closure grasp does not contain the origin point. There are three cases if a ray-shooting in the direction of a given task wrench is performed: \u2022 There is no intersection", + " Desired quality is how well a TWS can fit into a given GWS, such that maximize \u03c1, subject to \u03c1 \u00b7 TWS \u2282 GWS (14) The vertexes vi of TWS are used to compute the desired scale factor. A ray starts from origin through the point vi, presenting task wrench wti , results in an intersection point with scale factor \u03c1i. By a force-closure grasp, we can always get a value of \u03c1i solving corresponding ray shooting problem as introduced in Sec. III-A. The minimum of this value min(\u03c1i) is the weakest task wrench that the grasp can perform, with which the TWS just fits in the GWS. Fig. 1(b) gives an example of the algorithm for TWS. Each vertex of the TWS are used to form a ray from origin, which has an intersection vi with the GWS. The multiple between the length from origin to vi and the length from origin to the corresponding vertex is used as the scale factor of this vertex. The vertex, where the minimum of scale factor is found, determines the quality of the grasp for this TWS. In the case shown in Fig. 1(b), grasp quality is decided by v1, which has the minimum normalized scale factor. Although a non force-closure grasp can not resist against disturbance force from all the directions, it can also apply forces and torques in certain region onto the object. To evaluate a grasp quality for a given TWS in a non force-closure grasp, we can compute \u03c1min and \u03c1max for each wrench wti by solving corresponding ray shooting problem introduced in III-B. If there is any ray without intersection with the GWS, the task can not be performed" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure19.7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure19.7-1.png", + "caption": "Fig. 19.7. The influence of the satellite attitude control mode on the path described by the robot end-effector - the same joint motion is carried out by a robot with a fixed base (left), an attitude controlled robot (middle) and a free-floating robot (right)", + "texts": [ + " The free-floating mode of operation is of interest for space robots not only for the reason that attitude control fuel may be saved, it will also be of importance during repair missions, when the servicing satellite is very close or in contact to the target satellite: any action of the attitude control system of either of the two satellites during this phase could lead to a collision and thus to a potential damage on the two spacecrafts. As long as the tasks, performed by the robot, are described in robot-fixed coordinates, the fact that the satellite position remains uncontrolled has no influence. If, however, the task is described in relation to an orbit-fixed co-ordinate system, as for example, capturing of a defect satellite, the satellites motion has to be taken into account (see Fig. 19.7). 332 D. Reintsema, K. Landzettel, and G. Hirzinger Space represents an extremely harsh operational environment for robotics technology, that limits the on-orbit life and performance capability of space robotics technologies, in particular, the electronics components due to radiation surroundings. The break for (low-cost) intelligent space robotics technology was the absence of innovative, high-performance and survivable electronics space components. Even if radiation-hardened (rad-hard) circuit versions are available, their tolerance levels are not always compatible with the space requirements" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003729_detc2011-48019-Figure12-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003729_detc2011-48019-Figure12-1.png", + "caption": "FIGURE 12: ADDITIONAL MEMBERS.", + "texts": [ + " Therefore, the developed flexible multibody vehicle model is considered to be effective for the vehicle handling simulation. 3 Copyright c\u00a9 2011 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/30/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use In order to evaluate the handling characteristics, a quasisteady state cornering test was conducted with the racing kart. Two kinds of the kart frame were prepared to evaluate the effect of the frame stiffness to handling characteristics. As shown in Fig. 12, the frame equipped with additional members such as front bumper, rear bumper, side bumper, and extra seat stay was compared to the frame without those additional members. In this test, the steering was kept fixed to +60 and -60 deg, which means the turning direction was opposite, and the running speed was increased from 5 to 20 km/h by little and little. The running test result shown in Fig. 13 consists of lateral bending strain at point (b) and yaw rate to the running speed. From the Fig. 13 (a), it can be observed that the strain and the yaw rate results in right turn and left turn were asymmetric, which is 4 Copyright c\u00a9 2011 by ASME Downloaded From: http://proceedings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003713_mmar.2014.6957388-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003713_mmar.2014.6957388-Figure4-1.png", + "caption": "Fig. 4. Two Rotor Aero-dynamical System - parts scheme.", + "texts": [ + " The laboratory setup consists of the mechanical unit with power supply and interface to a PC and the dedicated RTDAC/USB2 I/O board configured in the Xilinx technology. The software operates in real time under MS Windows XP/7 32-bit using MATLAB R2009/10,11, Simulink and RTW toolboxes. Real-time is supported by the RT-CON toolbox from INTECO. Control experiments are programmed and executed in real-time in the MATLAB/Simulink environment. The real-life installation is presented in Fig.3, and the scheme of the system is presented on Fig.4. According to the TRAS instruction manual the equations describing the motion of the system can be written as follows: d\u2126v dt = lmFv(\u03c9m)\u2212 \u2126vkv + Uhkhv \u2212 a1\u2126vabs(\u03c9v) Jv \u00b7 \u00b7 \u00b7 + g((A\u2212B)cos\u03b1v \u2212 Csin\u03b1v) Jv \u00b7 \u00b7 \u00b7 \u2212 1 2\u2126 2 h(A+B + C)sin2\u03b1vUhkhv Jv (27) d\u03b1v dt = \u2126v (28) dKh dt = Mh Jh = ltFh(\u03c9t)cos\u03b1v \u2212 \u2126hkh + Uvkvh Dsin2\u03b1v + Ecos2\u03b1v + F \u00b7 \u00b7 \u00b7 \u2212 a2\u2126habs(\u03c9h) Dsin2\u03b1v + Ecos2\u03b1v + F (29) d\u03b1h dt = \u2126h, \u2126h = Kh Jh(\u03b1v) , (30) and two equations describing the motion of rotors: Ih d\u03c9h dt = Uh \u2212H\u22121 h (\u03c9h) (31) and Iv d\u03c9v dt = Uv \u2212H\u22121 v (\u03c9v) (32) where: \u2126v - angular velocity (pitch velocity) of TRAS beam [rad/s]; \u2126h - angular velocity (azimuth velocity) of TRAS beam [rad/s]; \u03c9v - rotational speed of main rotor [rad/s]; \u03c9h - rotational speed of tail rotor [rad/s] Kh - horizontal angular momentum [N m s]; Mh - horizontal turning torque [ Nm]; Ih - moment of inertia of the main rotor" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003442_icar.2013.6766514-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003442_icar.2013.6766514-Figure3-1.png", + "caption": "Fig. 3. Definition of the plane frame P with respect to the base frame B in the general case (a) and of the ellipse frame E with respect to P.", + "texts": [ + " There are different possible ways to represent the geometry of such a path. The ellipse is defined in three steps. The plane in 3D space in which the ellipse lies is defined by its normal vector n and the plane\u2019s normal distance d from the origin of the base frame B in which it is represented (see Fig. 2). The normal vector n additionally defines the sense of rotation of the movement on the ellipse. The pose of the ellipse within the plane is defined by the position xe of the ellipse centre and the ellipse\u2019s orientation angle \u03c6e (see Fig. 3(b)). The shape of the ellipse is represented by its semiminor ellipse axis length b and its eccentricity \u03b5. The eccentricity of an ellipse expresses the relation between the lengths of the major and minor axis. It is defined as \u03b5 = \u221a 1\u2212 (b/a)2. The translation component of the path is represented by the translation vector t, which corresponds to one full ellipse orbit. The start point of the path lies in the ellipse plane, on the ellipse. The polar angle \u03b80 in the ellipse frame E determines where on the ellipse the path starts", + " These are a general base frame B, a plane frame P to represent the ellipse pose within the ellipse plane and an ellipse frame E to represent the ellipse itself. We use the following conventions: The z-direction of the plane frame P is the plane\u2019s normal direction. The x-axis of P is parallel to the intersection line of the ellipse plane with the xy-plane of the base frame B and the y-direction is chosen such that yp values in the plane frame grow for increasing zb values in the base frame, if the z-component nz of the plane\u2019s normal vector is not negative in B. The origin is the normal projection of the base frame origin onto the plane (see Fig. 3(a)). The ellipse frame E has its origin in the ellipse centre. The x-axis and y-axis lie in the direction of the semimajor and semiminor axis, respectively. The z-axis is parallel to the one of the plane frame (see Fig. 3(b)). Minuscule superscript indices indicate in which frame a geometric object is represented. For example, Xb would be object X in the base frame B. To determine the path geometry from given movement directions at the beginning and the end of the path, d and e, respectively, it is more convenient if the path begins and ends parallel to the ellipse plane and not, as in a general Squeezed Screw path, at an angle to it. Therefore, in a Modified Squeezed Screw path, the progression of the translation component is not totally linear to the progression of the ellipse component" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001275_acc.2015.7171875-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001275_acc.2015.7171875-Figure4-1.png", + "caption": "Fig. 4. The 3D helicopter prototype", + "texts": [ + " (32) Case 2: When all the relative degrees ri j \u2265 2 \u03d5\u0304d(t) = \u03d5(t)\u2212\u2126 \u22121( \u02d9\u0304xI(t)+\u039bx\u0304I(t)), (33) x\u0304I(t) = K\u2020 I (t)(u(t)\u2212K\u03c9u(t)\u03c9u(t)), (34) \u02d9\u0304xI(t) =\u2212S\u2020 I (t)(F(t)+S\u03c9u(t)\u03c9\u0307u(t)), (35) where F(t)\u2208Rm and S(t), [S\u03c9u SI ]\u2208Rm\u00d7na , S\u03c9u \u2208Rm\u00d7nm, SI \u2208Rm\u00d7p are defined as F(t), R\u0307\u22121(t)BT a P(t)xa(t)+R\u22121(t)BT a (Pa(t)+Q(t))xa(t), (36) S(t), R\u22121(t)BT a (2(P(t)xa(t)x\u2020 a(t)\u2212 xa(t)x\u2020 a(t)P(t)) \u2212P(t)+2x\u2020 a(t)P(t)xa(t)Ina), (37) Then the control law (24) along with the saturation (29), based on the predictor (11), and the reference system given by (2) and (31), guarantees that i) The tracking error e(t) , y \u2212 ym(t) are uniformly ultimately bounded. ii) |ui(t)|\u2264umaxi and |u(t)\u2212u(t\u2212\u2206t)|\u2264\u2206umax, i = 1, ...,m. Figure 1 presents the control framework, and Fig. 2 and Fig. 3 illustrate the predictor and the controller structures, respectively. The controller performance is studied by considering the real time implementation on the Quanser 3-DOF helicopter depicted in Fig. 4. The helicopter body is mounted at the end of an arm and is free to rotate around the arm (pitch). The arm is free to rotate around the y-axis (elevation) and z-axis (travel) at the pivot point O. Two DC motors with attached propellers generate driving forces for the helicopter. Hence, the system has 3 outputs, i.e. the pitch \u03c6(t), the elevation \u03b8(t), the travel \u03c8(t) angles, all of which are measured via optical encoders, and has 2 control inputs v(t) = [vf(t), vb(t)]T where vf, vb are the voltages applied to the front and the back motor respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003116_amr.837.699-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003116_amr.837.699-Figure2-1.png", + "caption": "Figure 2. Tangential stress variation along the coordinate system axes", + "texts": [ + " In order to evaluate the tangential stress distribution over the entire cross section, it is first necessary to determine the orthogonal tangential stresses, as: 2 2 2 2 2 1 ( ) t xy t M R y G z z I y z ; 2 2 2 2 2 2 2 1t xz t M R y G R y y I y z y z . (14) Along the y axis, at 0z , the xy annuls, so that the stress vector reduces to its xz component, parallel to the y axis, given by: 2 2 t xz t M R R y I y . (15) Stress variations along the y and z axes are illustrated in Fig. 2. It can be noticed that xz annuls for a value of the y abscissa that satisfies 2 3 2 0Ry y R . This coordinate indicates the torsion center position, which is no longer the same as the cross section weight center. Maximum stress is reached at the intersection between the y axis and the circle of radius : max 2 1 2 t t M R I R . (16) For 0 , the maximum stress becomes max 2 t tM R I , which is twice the value obtained for a full circular cross section. The tangential stress corresponding to the intersection between the z axis and the circle of radius R, where 2y R , is significantly smaller: 2 1 2 1 4 t xz t M R I R " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003462_2015-01-0615-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003462_2015-01-0615-Figure1-1.png", + "caption": "Figure 1. 4 DOF half vibration model of truck", + "texts": [ + " A corresponding 2 DOF rigid vibration model is also built. The frequency domain analysis method is utilized to analyze truck ride comfort, and two models are simulated to explain the influence of body flexibility on truck ride comfort. The body flexibility is modeled as an elastic Euler-Bernoulli equal section beam with both free ends. Bilateral symmetry of automotive quality and the same road excitation of left and right sides are assumed, and a 4 DOF half vibration model of truck is built in Figure 1. The parameters in Figure 1 are given in Table 1. The DOFs of the model include vertical displacement and pitch angle displacement of body, the first and second order bending displacements of body. Bending Vibration of Elastic Beam of Equal Section with Both Free Ends The body is assumed to be an elastic Euler-Bernoulli beam of equal section. The bending vibration equation of elastic beam of equal section is as following [9] (1) Where EI is bending stiffness, \u03c1 is mass per unit length, f(x,t) is distributed load. When the elastic beam does the free vibration, distributed load f(x,t) is equal to zero" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003471_detc2011-48055-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003471_detc2011-48055-Figure6-1.png", + "caption": "FIGURE 6. Calyx FEM model of the example ideal gear pair.", + "texts": [ + "org/about-asme/terms-of-use above three sub-sections yields nb(t) = n(\u2212t \u2212 T 2 + b\u2212\u2206c tan\u03b1 p T ) (9) Equation (9) is the back-side tooth number variation function for an arbitrary anti-backlash gear or a general external gear pair with 2b nominal backlash and \u2206c center distance change. NUMERICAL VERIFICATION OF BACK-SIDE MESH STIFFNESS The drive-side and back-side mesh stiffness relationship in (5) is the key to the general back-side mesh stiffness in (8) or back-side tooth variation function in (9). Therefore, equation (5) is verified first. The verification is achieved with a Calyx [12] finite element model of an ideal external gear pair (Figure 6). Calyx has very precise tracking of tooth contact for precise tooth geometry. In the simulations that follow Calyx tracks the contact for specified gear kinematics under unloaded conditions. The gear parameters are listed in Table 1. Figure 7 shows the in-contact tooth number variations at the drive and back sides by tracking the numbers of gear teeth along the drive- and back-side lines of action using the above Calyx model. At t = 0, the pitch point at the drive side of the driving gear is in contact", + " O\u2032\u2032 in Figure 7 and Figure 9 is the point on the drive-side tooth number variation function at t = T 2 . Figure 9 shows that the drive-side tooth number variation function before point O \u2032\u2032 is symmetrical to the back-side gear tooth number variation function after point O\u2032. This provides numerical validation of equation (5). The next step is to verify the back-side mesh stiffness varying function for the gear pairs with 2b nominal backlash in equation (7). To do so, the tooth thickness of the gear pair in Figure 6 is reduced by 10% such that b p = 0.05. The tracking results of the drive-side and back-side tooth number variations from Calyx are shown in Figure 10. O\u0304 in Figure 10 is the point on the drive-side gear tooth number variation function at t = \u2212 T 2 + b p T . The drive-side gear tooth number variation function before point O\u0304 is symmetrical to the back-side gear tooth number variation function after point O\u2032. Thus, the results in Figure 10 agree with equation (7). In order to verify the general back-side tooth variation function in equation (9), the center distance between the two gears in Figure 6 is reduced such that \u2206c tan\u03b1 p = 0.025. The tracking results for the drive-side and back-side tooth number variation functions are shown in Figure 11. O\u0303 in Figure 11 is the point on the drive-side gear tooth number variation function at t = \u2212 T 2 + b\u2212\u2206c tan\u03b1 p T . The drive-side gear tooth number varia- tion function before point O\u0303 is symmetrical to the back-side gear tooth number variation function after point O\u0303. Comparing with Figure 10, the phase lag of the back-side tooth number variation function in Figure 11 is reduced by 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001113_gt2015-43561-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001113_gt2015-43561-Figure3-1.png", + "caption": "Figure 3: Diagram showing a typical IGB shrouded crown gear. Arrows show the direction of the air flow through the system.", + "texts": [ + " The shroud covering a Trent-style IGB crown gear contains slots to allow oil and air to exit the gear rear chamber and travel into the front chamber from which it can be scavenged. Comparison of Figure 1 and Figure 2 shows that an aeroengine rear chamber is quite constrained in size compared to that utilised in the CFD models where attention was focussed on the gear-shroud system. Also, and perhaps more importantly, the CFD modelling has no coupling between outlet flow and inlet flow. Consequently, any swirl induced in the chamber in front of the gear by the gear rotation is not propagated through to the shroud inlet. Figure 3 shows a stylised representation of an IGB crown gear illustrating how it is likely that swirl at shroud exit slots would affect the condition of the air drawn into the gear at shroud inlet. 2 Copyright \u00a9 2015 by Rolls-Royce plc Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 12/24/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use The work in this paper explores the notion that the environmental conditions found within the IGB must be considered in order to correctly understand and model the flow through the shrouded gear" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003411_amm.318.59-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003411_amm.318.59-Figure2-1.png", + "caption": "Fig. 2 Finite element model of gearbox", + "texts": [ + " The 3-D geometric model of marine gearbox system is showed in Fig.1 1 housing, 2 input gear, 3 input shaft, 4 bearing 1, 5 output shaft, 6 bearing 3,7 output gear, 8 coupling, 9 bearing 3, 10 bearing 4 Fig.1 3-D geometric model of marine gearbox system The simulation effect of the stress wave by linear element is better than that by the secondary element [6]. There were 736,460 hexahedral elements, 80428 tetrahedral elements and 908,164 nodes in the gearbox system model. The gearbox assembly finite element mesh model is shown in Fig. 2. Engineering The contact pairs among helical gear teeth and bearings were defined as dynamic contact\uff0cthe binding constraints (tie) were adopted to simulate interference fit between the inner ring of bearing and the shaft, the bearing cages were simulated by connection element (connector). The initial velocity field of each components were predefined, the constant speed v=500r/min was defined on the reference point of input shaft coupling, the load torque Tout=3842N\u00b7m was applied to the reference point of output shaft coupling, when t=0~1ms, the torque increased from 0 to Tout, then remained constant, the simulation time was 10ms" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002301_s0140525x13000617-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002301_s0140525x13000617-Figure3-1.png", + "caption": "Figure 3. The experiment by Holbrook and Burt de Perera (2009). (a) Line diagram of the choice maze (taken from the original article, with permission), in which fish were released into a tunnel and had to choose one of the exit arms to obtain a reward. (b) The conflict experiment of that study, in which the fish were trained with the arms at an angle, so that one arm pointed up and to the left and the other arm pointed down and to the right. During testing, the arms were rotated about the axis of the entrance tunnel so that one arm had the same vertical orientation but a different left-right position, and one had the same horizontal (left-right) position but a different vertical level. (c) Analysis of the choices showed that the fish greatly preferred the arm having the same vertical level. Redrawn from the original.", + "texts": [ + " Interestingly, although fish can process information about depth, Holbrook and Burt de Perera (2009) found that they appear to separate this from the processing of horizontal information. Whereas banded tetras learned the vertical and horizontal components of a trajectory equally quickly, the fish tended to use the two components independently, suggesting a separation either during learning, storage, or recall, or at the time of use of the information. When the two dimensions were in conflict, the fish preferred the vertical dimension (Fig. 3), possibly due to the hydrostatic pressure gradient. Several behavioural studies support the use of elevation information in judging the vertical position of a goal. For instance, rufous hummingbirds can distinguish flowers on the basis of their relative heights above ground (Henderson et al. 2001; 2006), and stingless bees can communicate the elevation as well as the distance and direction to a food source (Nieh & Roubik 1998; Nieh et al. 2003). Interestingly, honeybees do not appear to communicate elevation (Dacke & Srinivasan 2007; Esch & Burns 1995)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002268_detc2014-34898-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002268_detc2014-34898-Figure1-1.png", + "caption": "FIGURE 1. COMPOUND GEARBOX OF A WIND TURBINE [10]", + "texts": [ + " Like any other rotating machinery [3, 4], transmissions are subjected to internal and external vibrations, which compromise their performance, and can increase the cost and time of maintenance [5]. Internal vibrations can come from unbalanced on misaligned components, surface failures of the gear teeth, or damaged bearings. External sources of vibration can come from wind transients [6], voltage sags [7], or wave loading [8] in the case of offshore turbines. It is a natural development build a model of the wind turbine drivetrain (Fig. 1) in a multidisciplinary platform, that can account for the aforementioned effects. There exist several platforms like Simulink, Dymola and SimulationX, that can deal with this kind of simulation; however, the development of custom elements in these platforms are usually difficult because they are closed source. Alternatives like OpenModelica allow these modifications, but cannot simulate complex multibody systems with the required detail. One free software, and thus open application that can be used for multibody dynamics in a multidisciplinary view is MBDyn1 [9]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000947_978-3-540-73958-6_5-Figure5.9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000947_978-3-540-73958-6_5-Figure5.9-1.png", + "caption": "Fig. 5.9. Rotor disk for integration", + "texts": [ + " Due to wind and cyclic pitch, lift and drag change during one cycle. For control purposes, these forces and the resulting torques can be combined to mean values. MARVIN has a main rotor speed of approximately 1300 r p m and two blades, which gives an effective frequency of 40Hz. That means it is not reasonable to increase the controlling frequency of the used controller beyond this frequency using just mean values of the forces. For determination of mean values of forces, all components of each rotor element have to be integrated according to Fig. 5.9. The blades are assumed to be non-twisted and of constant chord from radius R1 to R2. For calculation of the force dF of one blade element the relative air speed u at this element is needed. It can be divided into two parts: vertical component uz and horizontal perpendicular to the blade ur. The third component, parallel to the blade, is not relevant for calculated forces. ur = \u03c9Mr \u2212 cos(\u03b1)vwy + sin(\u03b1)vwx (5.15) uz = vwz (5.16) Figure 5.10 shows the relevant relationships of angles \u03c6 and forces" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002700_chicc.2014.6896331-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002700_chicc.2014.6896331-Figure3-1.png", + "caption": "Fig. 3: The integrated SVPWM", + "texts": [ + " In one segment SVPWM mode, the actual voltage space only works in the basic vector and zero vector as shown in Fig. 2(d). As analyzed above, the actual linear modulation region loses 2 limt in the seven segments SVPWM mode, namely, the actual linear modulation region is 0, 1 2 limr tu , wherein, ru is marked as the given reference voltage. In order to expand the linear modulation region, we synthesize all the above mentioned SVPWM modes in Fig. 2. The synthesized linear modulation region is 0, 1 limr tu , as shown in Fig. 3. In Fig. 3, the new linear modulation region is an inscribe circle ( (1 )0 r limu t ), and the actual over-modulation region is outside the inscribe circle ( (1 ) 2 / 1.732lim rt u ). It is able to switch smoothly between the seven segments PWM mode and five segments PWM mode in new linear modulation region. When 0 2 limt t , the seven segments SVPWM mode is implemented; when 0 2 ,lim limt t t five segments SVPWM mode is performed. As some regions are unable to reach directly in the actual over-modulation region 0 limt t , so we present the four optimized over-modulation strategies to convert the voltage vector into five segments boundary and three segments or one segment PWM voltage space as shown in Fig", + " Marked ,lim db pwlt t t wherein, / ,db db dbsT T Tt is the dead time; / ,pwl pwl pwlsT T Tt is the minimum limitation of narrow pulse width. The concrete analysis is as follows: 1) Provided that 0 2 ,limt t directly output ,mu namely, ,p mu u at this moment, 1 1 2 2, ,t t t t thus 0u . 2) Provided that 0 2lim limt t t , directly output mu , at this moment, 1 1 2 2, ,t t t t thus 0u . 3) Provided that 02lim limt t t , the modified voltage mu is transformed into the practical voltage pu by an equal proportion compression method according to Fig. 3(a), at this moment, 1 1 1 2 1 ,lim t t t t t 2 2 1 2 t t t t Thus 1 2( ) (1 ) 2 3 .lim dc s t t t u U T 4) Provided that 00 2,limt t the modified voltage mu is transformed into the practical voltage pu by an equal proportion stretch method according to Fig. 3(b), at this moment, 1 2 1 2 1 2 1 2 , ,t t t t t t t t thus 1 2( ) 1 2 3 .dc s t t u U T 5) Provided that 0 1 2 2 10 2 1 2 1,t t t t t the modified voltage mu is transformed into the practical voltage pu by an equal proportion compression method according to Fig. 3(c), at this moment, 1 2 1 2 1 2 1 2 ,t t t t t t t t . Provided that 1 ,limt t thus 1 20, 1;t t Provided that 2 ,limt t thus 1 21, 0t t . So 1 2( ) 1 2 3 .dc s t t u U T 6) Provided that 0 1 2 1 20 2 1,t t t t t the modified voltage mu is transformed into the output ahead adjacent basic vector kv according to Fig. 3(d), at this moment, 1 21, 0,t t thus 1 2( ) 1 2 3 .dc s t t u U T 7) Provided that 0 1 2 2 10 2 1t t t t t , the modified voltage mu is transformed into the output latter adjacent basic vector 1kv according to Fig. 3(d), at this moment, 1 20, 1,t t 2 1t , thus 1 2( ) 1 2 3 .dc s t t u U T Taking all these above steps into account, the function 1 2 0( ), ,F t t t regarding to 1 2 0, ,t t t is obtained as follows: 0 1 2 1 2 0 0 1 2 0 0 ( ) (1 ) 2( , , ) 2 3 ( ) 1 2 2 3 dc s dc s lim lim lim lim lim t t t t t u F t t t U t t tT t t U t tT In order to verify the correctness and advantages of the proposed flux-weakening control strategy, this paper implements a series of comparative experiments based on Fig. 1 and Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001933_amr.1030-1032.1167-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001933_amr.1030-1032.1167-Figure2-1.png", + "caption": "Fig. 2 Diagram of force bearing model Fig. 3 Maximum contact stress point", + "texts": [ + " Before carrying out finite element analysis, removing the mounting hole, chamfering of outer ring and inner ring as well as the bearing retainer. In order to facilitate the selection, we build four contact points of each rolling body [11,12] . Introduction of each bearing model into the finite element analysis software Ansys Workbench, then taking static analysis. Number of mesh elements is 100736 and number of node is 188707. Applying a fixed constraint on the outer ring bottom surface. Applying axial load Fa=103KN, radial load Fr=103KN and bending moment M=590.3KN, M. The load state is show in fig.2. Probe analysis stress of primary contact poins and secondary contact points. From fig.3 we can see the maximum contact stress point does not occur in the theory of probe point but nearby the theoretical analysis point, as shown in figure3.This is because inner ring, outer rings and rolling bodies under combined loads, deformation change the location of the contact point. This is consistent with the conclusion of paper[3]. 0.00E+00 2.00E+07 4.00E+07 6.00E+07 1 3 5 7 9 11131517192123252729 Primary 1\uff08Pa\uff09 Primary 1\uff08\u2026 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001740_s11018-015-0773-4-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001740_s11018-015-0773-4-Figure1-1.png", + "caption": "Fig. 1. Simplifi ed physical model of transducer of triaxial accelerometer.", + "texts": [ + "1007/s11018-015-0773-4 MEMS accelerometer is based on transformation of an incremental capacitance signal into an electrical signal formed by comb-shaped moveable and fi xed micromechanical plates. An MMA7331L capacitance-type triaxial MEMS accelerometer mounted in each submodule and forming an analog nine-degree block of micromechanical accelerometers is used in an inertial measurement module. Three of the nine sensitivity axes of each accelerometer are collinear. Such an accelerometer possesses a differential capacitor structure (Fig. 1). Each of the sets of fi xed capacitor plates of the microcapacitor is connected in parallel within the circuit chip. Once there is an acceleration along the X axis, the two fl oating ledges L1 and L2 approach one of the sets of fi xed columns Vx1, Vx2 and recede from the other set Vx3, Vx4. As a result, as the platforms of the two ledges L1 and L2 move relative to the two columns Cx1 and Cx2, the areas are no longer identical and the capacitances between the moveable and each of the fi xed capacitor plates change" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001567_icma.2015.7237783-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001567_icma.2015.7237783-Figure11-1.png", + "caption": "Fig. 11 Coordinates of neck mechanism", + "texts": [ + " The upper and lower supporting plates are connected by a fixed supporting shaft and two RSSR space linkage mechanism. As performer part, the upper plate has 2 DOFs: pitch and roll. Its revolving angle are 3 and 4 respectively. This mechanism is driven by two motors, which are installed symmetrically. Its revolving angle are 1and 2 respectively. Similarly, it is an inverse kinematics problems to deduce 1 and 4 with the expression of 2 and 3. The coordinates systems {0}, {1}, {2}, {3} are established as it is showed in Figure 11. Coordinate {0}, {1}, {2} are fixed on the ground. The origin of coordinate {1} and {2} are centres of motor rockers, whose revolving angles are 1, 2 respectively. The origin of {0} is centre of the cardan joint which connects upper plate and supporting shaft. Coordinate {3} are fixed on upper plate and revolve with it. Coordinate {3} revolves around x axis and y axis in reference to coordinate {0}. Its revolving angles are 3 and 4. 2 1 3 4 2 3 4 3 3 4 2 4 3 4 1 1 2 4 1 3 4 (25 cos 35.79 sin 10" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003643_s0894-9166(15)60013-1-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003643_s0894-9166(15)60013-1-Figure1-1.png", + "caption": "Fig. 1. Contact model of two elastic solids with regular surfaces.", + "texts": [ + "com/lzq537@aliyun.com \u22c6\u22c6 Project supported by the National Basic Research Program of China (973 Program)(No. 2009CB724406). II. THEORETICAL MODEL OF THE NORMAL CONTACT STIFFNESS ON AN ELLIPTICAL AREA Elliptical diffraction fringes were observed when two convex lenses made contact without external force, gradually expanded and became clear under a small load[11]. The normal contact stiffness of two asperities with an elliptical contact area is discussed based on Hertz\u2019s experimental results, as shown in Fig.1. The dashed curves indicate the edge of the original sections, while the solid curves the edge of the sections with maximum elastic deformation \u03b4i. Figure 2 shows a 3-D contact model with \u03b8 denoting the included angle between the principal curvature directions on two surfaces. The initial gap between two solid asperities can be expressed as[2] h = z1 \u2212 z2 = Ax2 + By2 = 1 2R\u2032 x2 + 1 2R\u2032\u2032 y2 (1) where A, B can be written as A = q0b E\u2217e2a2 [K(e) \u2212 E(e)] (2) B = q0b E\u2217e2a2 [ a2 b2 E(e) \u2212 K(e) ] (3) where e is the eccentricity of the elliptical contact area" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002868_s12206-015-0143-9-Figure22-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002868_s12206-015-0143-9-Figure22-1.png", + "caption": "Fig. 22. Schematic diagram for PN guidance.", + "texts": [ + " that centripetal acceleration control is similar to PN guidance and a proportional-derivative controller for pathfollowing guidance [16]. Thus, it is not necessary to use circular path logic for a straight path, and the following PN guidance logic is used in a straight path: N CK l=a V & . (3) For an aircraft traveling with a velocity vector of V and a line of sight (LOS) angle \u03bb to a reference point, CV is the closing velocity and Na is the control force normal to the CV vector. Na is proportional to the rate of change of LOS and the closing speed using the proportional gain value of K. Fig. 22 shows the schematic diagram for PN guidance. In an actual application, PN guidance is applied as: N G cmd N CK l= &V V , (4) ( )1 /N cmd cmd ctany l -= + V V , (5) where Na is altered by N cmdV , the velocity command normal to the closing velocity. The altered schematic and notations are shown in Fig. 23. The LOS is calculated as the heading angle referenced to the northerly direction, and the heading command cmdy is tilted by the arctangent of /N cmd cV V . If the aircraft\u2019s flight path is a series of waypoints, and if the direction of the course is changed by a certain amount of angle, the PN guidance may create an abrupt change in heading angle and side slip, which results in abrupt roll rate changes" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001184_1.3663086-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001184_1.3663086-Figure2-1.png", + "caption": "FIGURE 2. (a) Vibrating model of system, (b) material motion.", + "texts": [ + " THE GOVERNING EQUATIONS OF THE MOTION The equations of motion for the vibrating model of system may be obtained by using Lagrange\u2019s Equation, Here the measurement values are taken from the position, where the components in the same directions of the vertical force applied by the leaf spring on itself and the weight force balance each other. If the mass of the leafs working as spring and the mass of the 3AA connecting rod, the changes in the angle \u03c8 and moment 0M applied on the crank arm by the motor are neglected, the coercive force in the vibration motion can be written as below \u201cFIGURE 2b\u201d [10], then the coercive force becomes 61 Downloaded 24 Sep 2013 to 146.232.129.75. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions 2 ( ) cos , ( 2) ( ), cos sin( ), , , ( ) ( )sin , , ,O t n t F t F t F t M r t F m r m thetotal vibrating mass r thelengthof thecrank = \u03b3 \u03b3 = \u03c0 \u2212 \u03b8 + \u03c8 \u03b3 = \u03b8 + \u03c8 \u0398 = \u03b8 \u2212 \u03c8 \u0398 = \u03c9 = \u03c9 = \u03c9 = = (1) The conveyor empty; writing the equations of motion in terms of nondimensional variables, where c , k are damping coefficient, spring constant of the vibrating conveyor, M is mass of the trough of the conveyor, nt\u03c4 = \u03c9 (the nondimensional time), n\u03c9 is the natural frequency, 2 1n k M\u03c9 = , 1 M\u03b5 = , 2 nc\u03bc = \u03c9 , 2 2 nk\u03b1 = \u03c9 2 3(1 ) sin( ) 2nz z F z z \u2032\u2032 \u2032+ = \u03b5 \u03c9 \u03b8 + \u03c8 \u2212 \u03bc \u2212 \u03b1 (2) Ride mode: writing the equations of motion in terms of nondimensional variables, we will obtain, where m is mass of the conveyed material on the trough of the conveyor, g is acceleration due to gravity, 2 1= +n k m M\u03c9 , = nt\u03c4 \u03c9 (the nondimensional time), 1 ( )M m\u03b5 = + , 2 nb c= \u03c9 , 2 2 nk\u03b1 = \u03c9 , 2 nG mg= \u03c9 2 3(1 ) sin( ) 2 sinnz z F bz z G \u2032\u2032 \u2032+ = \u03b5 \u03c9 \u03b8 + \u03c8 \u2212 \u2212 \u03b1 \u2212 \u03b2 (3) Slide mode: writing the equations of motion in terms of nondimensional variables, we will obtain, where \u03bc the coefficient of friction between the materials and the trough, = nt\u03c4 \u03c9 (the nondimensional time), where 21 [ (sin sin cos )]M m\u03b5 = + \u03b2 \u00b1 \u03bc \u03b2 \u03b2 , 2 nb c= \u03c9 , 2 2 nk\u03b1 = \u03c9 , 2 nG mg= \u03c9 2 3(1 ) sin( ) (sin cos ) 2 \u2032\u2032 \u2032+ = + \u2212 \u00b1 \u2212 \u2212 nZ Z F G bZ Z\u03b5 \u03c9 \u03b8 \u03c8 \u03b2 \u03bc \u03b2 \u03b1 (4) The method of multiple scales is used to obtain approximate analytical solution of Eq" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003719_vppc.2014.7007143-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003719_vppc.2014.7007143-Figure7-1.png", + "caption": "Figure 7. Basic structure of the elastic crank-connecting rod system.", + "texts": [ + " T is a column matrix) matrices L\u0304, A\u0304, B\u0304, C\u0304, N1 and N2 degenerate to scalars and from (8) it follows N1 = N2 = \u02d9\u0304L 2 (9) The mechanical systems with time-varying inertia can always be obtained as the limit case of an extended POG model with masses and elasticities, see the POG model given in (5), when the elasticities tend to zero. The time-varying reduced model (6) can always be obtained from the extended POG model using a time-varying rectangular and congruent POG transformation z = T(t)x which eliminates the elasticities. An example of this method is given in the following section. Consider the elastic crank-connecting rod system shown in Fig. 7. The meaning of the system parameters is shown in Fig. 8. The relation which links the angular position \u03b8(t) of the crank with the translational position xk of point Pk is the following: xk(\u03b8) = R cos \u03b8 + \u221a L2 \u2212 (R sin \u03b8 \u2212 d)2. The velocity x\u0307k can be expressed as a function of the velocity \u03c9 = \u03b8\u0307 as follows: x\u0307k(t)=R [ \u2212 sin \u03b8 \u2212 (R sin \u03b8 \u2212 d) cos \u03b8\u221a L2 \u2212 (R sin \u03b8 \u2212 d)2 ] \u03b8\u0307 =R [ \u2212 sin \u03b8 \u2212 (sin \u03b8 \u2212 \u03b2) cos \u03b8\u221a \u03b12 \u2212 (sin \u03b8 \u2212 \u03b2)2 ] \ufe38 \ufe37\ufe37 \ufe38 H(\u03b8) \u03c9 = H(\u03b8)\u03c9 where function H(\u03b8) and the auxiliary parameters \u03b1 and \u03b2 are defined as follows: H(\u03b8) = \u2202x(\u03b8) \u2202\u03b8 , \u03b1 = L R > 1, \u03b2 = d R < 1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003069_amr.816-817.90-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003069_amr.816-817.90-Figure1-1.png", + "caption": "Fig. 1 Sampling diagram for slab specimens of single crystal superalloy", + "texts": [ + " The size of the single crystal superalloy plate is 130 mm \u00d7 25 mm \u00d7 15 mm. The heat treatment of DD6 single crystal superalloy plates is conducted in a vacuum. The solution heat treatment is 1290 \u00b0C/1h + 1300 \u00b0C/2h + 1315 \u00b0C/4h/AC, and the first aging heat treatment is 1120 \u00b0C/4h/AC, and the second aging heat treatment is 870 \u00b0C/32h/AC. The orientation of single crystal superalloy plates is determined using X-ray diffraction, and the deviation degree between principal stress axis and the [001] orientation is less than 10\u00b0. According to Figure 1, the single crystal superalloy plate after heat treatment is machined into thin-walled slab specimens with the section sizes of 0.5 mm, 0.75 mm and 1.0 mm respectively, and requirements of detailed sizes are shown in Figure 2. The stress rupture lives of thin-walled slab specimens with different section sizes at 980 \u00b0C/250 MPa are measured and fractograph is observed. The stress rupture lives of thin-walled specimens of DD6 single crystal superalloy with different section sizes at 980 \u00b0C/250 MPa are shown in Table 1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001479_carpathiancc.2015.7145141-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001479_carpathiancc.2015.7145141-Figure5-1.png", + "caption": "Fig. 5 Coordinate system in the complex plane.", + "texts": [ + " The controllers were created as a digital in the environment of the signal processor of the dSpace type which is shown in the diagram in Fig. 4. The input and output 2015 16th International Carpathian Control Conference (ICCC) 553 voltage of the DC amplifier is shown also in this figure. The sampling frequency is equal to 5 kHz. IV. CLOSED CONTROL LOOP A variable u in Fig. 4 is a control variable and a variable r is a controlled variable. The controlled variable contains the coordinates of the bearing journal axis while the control variable contains the coordinates of the bushing axis as is shown in Fig. 5. Because both the variables indicate coordinates in the plane, then they can be considered as two component vectors. The same meaning as the vector has a complex variables. The real part of this variable has the meaning of the X-coordinate while the imaginary part has the meaning of the Y-coordinate, therefore r = X + jY. The origin (0, 0) of the coordinate system in the complex plane is situated in the center of the cylindrical bearing bore. There are many ways how to model journal bearings, but this paper prefers a lumped parameter model, which is based on the concept which was developed by Muszynska [10]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001429_9781118886397.ch14-Figure14.10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001429_9781118886397.ch14-Figure14.10-1.png", + "caption": "Figure 14.10 Details of transmission line conductor bundle with spacers (Source: EPRI (2006)).", + "texts": [ + "34) and sin \ud835\udf03 = zt D (14.35) cos \ud835\udf03 = \u221a 1 \u2212 ( zt D )2 (14.36) In the simple case where Fout reduces to zero before maximum swing is reached (i.e., the fault clears), Fspacer = 2SW zt D \u221a 1 \u2212 ( zt D )2 (14.37) Since D = WS2 8H (14.38) the maximum spacer force is Fspacer = 16H zt S \u221a 1 \u2212 ( zt D )2 \u2245 16Hzt S (14.39) This will be the maximum spacer force in both tension and compression for the usual case where the fault has cleared before maximum conductor deflection has occurred. Transmission line spacers (Figure 14.10) are designed to withstand the compressive force on bundled conductors caused by short-circuit forces. Spacer compression may be calculated with the Manuzio formula (Lilien et al., 2000), originally for flexible bus substation design: Pmax = 1.45I\u2032\u2032 \u221a Fst log ( aS dS ) (14.40) where Pmax = compression force on the spacer in N Fst = initial static tension on the conductor bundle in N as = conductor spacing in mm ds = conductor diameter in mm. Tests by Lillien et al. showed that the Manuzio formula underestimated the stress by 50%" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003003_amm.456.549-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003003_amm.456.549-Figure2-1.png", + "caption": "Fig. 2 Ball-on-disc test rig", + "texts": [ + " By gradually increasing the rotation speed of the glass disc, the corresponding interference grey values of the central contact area are continually collected and sent to the computer. With these data, the periodical curve of the central interference grey value with sliding speed is drawn. Compared this curve to theoretical calculation, the interference grade for certain sliding speed can be obtained. The optical interference experiments are conducted in unidirectional lubricating sliding contact using a ball-on-disc configuration (Fig. 2). The ball is made of steel with a diameter of 26 mm. The flat disc is made of glass, and the surface contacted with the ball is deposited with Cr to enhance its reflection. The thickness of deposited Cr is about several nanometers in order to eliminate the influence of interference. The composite Young\u2019s Modulus of the steel ball and glass disc is 9.3\u00d710 8 Pa. A green filter is used to generate the monochromatic light. The weight loaded on the steel ball is 0.68 N. The initial viscosity of the lubricant measured by Digital Viscometer is 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003040_itsc.2014.6958081-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003040_itsc.2014.6958081-Figure1-1.png", + "caption": "Figure 1. An example of a two-dimension decompostion of the acceleration with respect to trajectory in the horizontal plane.", + "texts": [], + "surrounding_texts": [ + "Since the acceleration of a motion can be discomposed in two directions: the tangential direction of the velocity and the normal direction of the velocity, we can decompose the acceleration vector of the vehicle into the direction of travel and the direction which is perpendicular to the direction of travel. As a wheeled vehicle runs in regular conditions on the road for the most of time, its direction of the velocity is constrained by the wheels. This means that the vehicle moves coasting the tangential direction of the wheels and there is no velocity in the normal direction of the wheels unless a leap or a glide happens, which is rare. What can be also inferred from this wheels constrain is that the attitude of the body and the velocity direction change with the same angular velocity. To get the observation of the changing rate of vector enV from the body frame, using Coriolis Theorem, we transform (5) and get: en en eb en dV F V bdt (7) Where eb is the angular velocity of B frame with respect to the E frame. The first term in the right-hand side of (7) is the changing rate of the velocity of the vehicle relative to the E frame observed in the body frame. Because it is in the tangential direction of the wheels, it can be measured by odometer. Also, we get the second term in the right-hand side of (7) which cannot be observed in the body frame. It is in the normal direction of the velocity because it is generated by making a turn, which is similar to the centripetal force. This force changes the direction of the velocity n enV and also the attitude of the body with the same angular velocity eb n because the attitude of the vehicle rotates with the change of travel direction, i.e., the direction ahead of the vehicle rear wheels is the direction of the travel velocity. We have already defined the body frame B. The tangential direction of the wheels is exactly the front direction of the rear wheels, which is y-axis of the body frame; the normal direction of the wheels is in the plane of x-axis and z-axis. Define 1 1 0 0 E , 2 0 1 0 E , 3 0 0 1 E and we have the nonholonomic constrains: 2 b en eny dV V E bdt (8) , b b b eb enenx z VV (9) The velocities of the vehicle in x-axis and z-axis are zero in the body frame. 0b b enx enzV V (10) Where b enyV is the acceleration on y-axis of the body frame; , b enx zV is the normal acceleration in the plane of x-axis and z-axis. b enxV and b enzV are, respectively, the velocities of the vehicle in x-axis and z-axis. III. INTEGRATED ERROR MODEL In the integrated navigation system, the direct and indirect integration methods are often used in the filter. The direct integration estimates the position, velocity, and attitude directly. However, the indirect method which estimates errors states is more popular because the errors are relative small. The nonlinear error equations can be linearized and the dynamics of the error states is relatively slow. In this paper, the indirect method is used to estimate the navigation errors." + ] + }, + { + "image_filename": "designv11_84_0002080_amm.592-594.2179-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002080_amm.592-594.2179-Figure1-1.png", + "caption": "Fig. 1. Structure sketch of a balanced hydraulic motor with planetary gear", + "texts": [ + " The stiffness of the hydraulic spring does effect when the closed cylinder is affected by outside force, which could be described as follows: = = L ( ) (2) From Function (2), when d / d = 0 = L[( ) ] (3) When = L / 2, d / d = 0, so = = (4) Work on current research of a balanced hydraulic motor that focuses on the characteristics of the motor with three planet gears [2]. References of a balanced hydraulic motor with more than three planet gears are hardly found. Theoretically, the larger the number of planet gears K is, the larger the displacement of the motor is. As is shown in Fig. 1, the sealing cavities must be separated from each other in order to form a hydraulic motor. Considering the pressure between sealing cavities and the location of distribution hole, pitch circle radius of gear B should be less than 1/3 arc length between the center points of two neighborhood planet gears. That is, K must satisfy the following equation K< \u00d7 + (5) Where , \u2014Tooth number of gear A and B. From above analysis, it can be seen that the larger the / , is. In order to analyze this kind of motor conveniently, it is called K type balanced hydraulic motor according to the number K of planet gears" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003141_cp.2013.2541-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003141_cp.2013.2541-Figure4-1.png", + "caption": "Figure 4. Model of Automated Bending Machine Cell", + "texts": [ + " The Gross Motion Planning: That determines the transfer motion of the robot. \u201cFine Motion Planning\u201d that determines the motion of the robot when the part is inside the punch-die space, especially the retraction of the part after it is bent. List of Programs for a robot to perform a bending operation Master Reset Inputs/outputs Sub program for different sheet size The figure5 shows in detail the logic of the robot which handles the sheet metal part and also feeds the part into the bending machine where the bending takes place. The figure4 explains the sequence of movements which the robot follows inorder to perform the required task. The program is checked and interfaced with HMI to run in the remote mode and then the production is carried out which has been recorded and there by observed improvements in the throughput which are included in the results. Figure 6 shows the CAD model of the bent sheet which is obtained by using robot to handle and feed during bending. Figure 6. Productivity parts/hr) The graph shows a linear behavior of the productivity in terms of (parts/hr)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001631_ijcnn.2015.7280569-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001631_ijcnn.2015.7280569-Figure1-1.png", + "caption": "Fig. 1. RoUing element ball bearing geometry", + "texts": [ + " These bearing fault frequencies are the functions of the bearing geometry and the running speed. The vibration frequency components for inner-race, outer-race and 978-1-4799-1959-8/15/$31.00 @2015 IEEE ball faults can be calculated using the following expressions [6]. n dball fit = 2\"frm(1 + -d -COSCP) pitch (1) n dball fot = -frm(1- -d -coscp) (2) 2 pitch where frm, dpitch, dball, n, and cP represents the frequency of rotation, pitch diameter, ball diameter, number of balls, and the contact angle as shown in Fig. 1. The vibration analysis technique normally include time domain, frequency domain and time-frequency domain tech niques. In the time domain the individual contributions, e.g. unbalance, to the overall machine vibration are difficult to identify. The most popular method for the bearing fault diagnosis is frequency domain, or spectral analysis [7, 8]. By using fast fourier transform, frequency-domain methods transform the time-domain vibration signals into discrete fre quency components. FFT transforms time-domain vibration signals find into a series of discrete frequency components" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001915_icnsc.2014.6819637-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001915_icnsc.2014.6819637-Figure2-1.png", + "caption": "Fig. 2. The object and coordinate systems", + "texts": [ + " ATmega128 controls the movement of the manipulator through 6 PWM signals, and the driving wheels are controlled by the motor drivers. Wheeled mobile platform Fig. l. The mobile manipulator system Considering the size constraint of the compact mobile manipulator with embedded vision, we consider that the object to be grasped is slender. The accuracy and robustness of the image processing have significant effects on the vision control. In this paper, the object is endowed with a radial symbol with dual outer rings and a dual \"+\" symbol on the top (see Fig. 2), which may satisfy the real-time computation requirement of an embedded processor. The information of radial symbol provides the basis of Bezier path planning, whereas the dual \"+\" symbol is used in vision-based approaching adjustment and grasping operation. Constrained by the field of view of the camera, only a \"+\" symbol is used in the precise grasping. III. CONTROL METHOD A. Vision Information Extraction Initially, the mobile manipulator searches the environment with the rotation of CMOS camera by adjusting the waist joint angle of the manipulator. The recognition of radial symbol is implemented in [8], and the pixel coordinates (u\" vr) of the radial symbol and the diameter dzp+drp are obtained. OcXcYcZc is established as the coordinate system of the CMOS camera, which is shown in Fig. 2. Based on the pinhole model, the position of the radial symbol in OcXcYcZc is \ufffd)c=L*jl(dzp+drp), Xoc=(uo-ur)*Yocifand Zoc=(VO-vr)*\ufffd)(/f, where L is the physical diameter of the symbol, (uo, vo) is the image coordinate of the intersection point of the optical axis centre line and the image plane, and f indicates the focal length in pixel. Different from the rectangle color block in [8], the dual \"+\" symbol is adopted to make the CMOS camera identify and match the object easily and efficiently. Similarly, the distance DiS=DoCl*jldpiXe/ between the object and the CMOS camera can be acquired, where dpixel=Sqrt((UI-U2) 2 +(VI-V2) 2 ), (UI, VI) and (Ub V2) are the pixel coordinates of centers of \"+\" symbols [9], and Dact is the physical distance between centers of\"+\" symbols. The position of the object in OcXcYcZc is X'oc=Di/cosa, Y'oc=Di/sina, Z'oc=Dis*(UO-(UI+U2)/2)/f, where a=e2+e3+e4, and e2, e3 and e4 are pitch angles of shoulder joint, elbow joint and wrist joint, respectively. Next, the visual information is converted into the position information. As shown in Fig. 2, OrXrYrZr is established as the coordinate system of mobile platform. eo and el are angles of the gripper joint and the roll joint of the wrist; es is the angle of waist joint. The coordinate system OrXrYrZr rotates around its z-axis for es, and x-axis for a to achieve the posture of the coordinate system OcXc Y cZc. Therefore, the pose matrix of camera coordinate system in the coordinate system of mobile platform can be obtained by coordinate transformation: -sin es coses o o cosa -sina o 1 sin a cosa (1) Thus we acquire the position of the object in OrXrYrZn which is represented by (Pow Pary, Parz). (2) where (Pom POLY' Pocz) is the position of the object in OcXcYcZc, and the position (Pew Pcry, Pcrz) of the camera in OrXrYrZr is defined as {\ufffdrx. : [lj cos 04 + 12 COS( 04 + 0,) + 13 COS a] sin OJ , \ufffdrJ - [lj cos 04 + 11 COS(04 + OJ) + IJ cos a] cos OJ \ufffdrz =/jsin04 +11sin(04 +OJ)+/Jsina (3) where Ib 12 and 13 are the length of the connecting rods (see Fig. 2). When there is only one \"+\" symbol in the camera's view, the information we rely on is the spoke. As shown in Fig. 3, a template of recognition is introduced. Points marked \"x\" is scanned, and several central points (marked \"0\") with the pixel coordinates (u\" vs)(s=1, ... ,ks) are obtained according to the mark points. On this basis, the slope of the spoke is kl= 7(\" , where kh is determined by (Uh' Vh) and (Uh+ b vh+l)(b=l, ... ,k,-l), Also, the width of the spoke in image is labeled as WI. As mentioned above, the approaching stage of the mobile platform includes two steps: Bezier curve based approaching and vision-based approaching adjustment that is implemented by endeavoring to move toward the center of the object" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003082_cca.2013.6662890-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003082_cca.2013.6662890-Figure3-1.png", + "caption": "Fig. 3. Trajectory with one switch", + "texts": [ + " We can easily discard the trajectory for the case when \u03b8T = \u03b7 as a suboptimal one since we can find a longer, single circular arc of radius R with same entry and exit points as that of the straight line trajectory for this case. Thus, we have to determine the longest trajectory out of the remaining possibilities for the exit states with \u03b8T = \u03b7 . Lemma 2. Given a trajectory with finite number of switches, we can find a longer trajectory with same entry and exit points but without any switches. Proof : Consider any exit point (xT ,yT ) in the first quadrant. Consider a trajectory Q1 with exactly one switch at point 1 and entry at point 4, as shown in Fig. 3. Then, x1 = xT \u2212 2Rsin \u03b8\u0304T cos\u03b7 (30) y1 = yT \u2212 2Rsin \u03b8\u0304T sin\u03b7 (31) where, \u03b8\u0304T = \u03b8T \u2212\u03b7 . Now, the coordinates of the entry points can be obtained as, x4 = x1\u2212 2Rsin \u03b8\u0304 cos ( \u03b7\u2212 \u03b8\u0304T + \u03b8\u0304 2 ) = xT \u2212 2Rsin \u03b8\u0304T cos\u03b7\u2212 2Rsin \u03b8\u0304 cos ( \u03b7\u2212 \u03b8\u0304T + \u03b8\u0304 2 ) (32) y4 = y1\u2212 2Rsin \u03b8\u0304 sin ( \u03b7\u2212 \u03b8\u0304T + \u03b8\u0304 2 ) = yT \u2212 2Rsin \u03b8\u0304T sin\u03b7\u2212 2Rsin \u03b8\u0304 sin ( \u03b7\u2212 \u03b8\u0304T + \u03b8\u0304 2 ) (33) Now, If l is the straight line distance between point 4 and the exit point, then l2 = (yT \u2212 y4) 2 +(xT \u2212 x4) 2. \u21d2 l2 4R2 = sin2 \u03b8\u0304T + sin2 \u03b8\u0304 + 2sin \u03b8\u0304 sin \u03b8\u0304T cos ( \u03b8\u0304T \u2212 \u03b8\u0304 2 ) Since the angles \u03b8\u0304T and \u03b8\u0304 are acute angles, l2 4R2 > sin2 \u03b8\u0304T cos2 \u03b8\u0304 + sin2 \u03b8\u0304 cos2 \u03b8\u0304T + 2sin \u03b8\u0304 sin \u03b8\u0304T cos \u03b8\u0304T cos \u03b8\u0304 (34) \u21d2 l2 4R2 > sin2(\u03b8\u0304 + \u03b8\u0304T ) (35) \u21d2 2Rsin\u22121 ( l 2R ) > 2R(\u03b8\u0304 + \u03b8\u0304T ) (36) This proves that a trajectory without any switches and with the same exit and entry points as Q1, is longer than Q1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003678_1.4024704-Figure21-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003678_1.4024704-Figure21-1.png", + "caption": "Fig. 21 Distribution of von-Mises stress when face CC0D0D is fixed and welding is done toward the fixed face", + "texts": [], + "surrounding_texts": [ + "The present investigation, divided into Parts I and II, presents the mathematical model that incorporates all the physical phenomena involved in a welding process and utilizes it to predict the significance of the various welding parameters and helps to understand the underlying physics in better prediction of weld D/W ratio and welded joint mechanical response. Detailed parametric studies performed in Part II of the present study help to delineate the effects of various welding parameters on the weld D/W ratio. Simulations are run with temperature dependent thermophysical properties. The results have been compared with experimental data. The parameters of particular interest are welding current, welding speed, arc length or electrode gap, electrode angle, and surface active agent content. For each of these parameters, simulations were run for both low ( 40 ppm) and high ( 150 ppm) surface active agent content. The present investigation also predicts the weld solidification mode by analyzing weld pool dynamics simulation results. The CFD results show that with increase in welding speed finer dendritic substructures would be obtained. Variation of weld D/W ratio with heat input per unit length of weld is also calculated. The heat input is varied in two ways \u2013 (a) varying the welding current and (b) varying the welding speed. Changing the heat input per unit length of the weld by varying either the current or speed will influence the important weld pool driving forces differently in a nonlinear manner. The complex interplay of the driving forces determines the final weld D/W ratio. Hence, when analyzing weld D/W ratio plots versus heat input per unit length of weld, it is important that attention be paid as to how the changes in heat input were achieved. Structural analysis of the welded joint highlights the deformation of the welded butt joint under various constraint configurations. The effect of welding direction (toward/away from) the fixed edge/face has also been studied. The von-Mises stress distributions under these constraint configurations have also been highlighted. The coupled field modeling of GTA welding provides a more accurate thermal energy distribution for structural analyses. Through material modeling the effects of microstructure evolution on mechanical behavior has been indirectly incorporated. Thermally induced stress distribution and workpiece deformations 021009-10 / Vol. 6, JUNE 2014 Transactions of the ASME Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 02/19/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use have been reported. The mathematical framework developed here provides us with the tools to study the multiphysics problem of welding. The present model works very well for welding currents 200 A. The predictions from the present model will not agree very well with thicker plates and higher welding currents. Hence, the present model needs to be extended to incorporate arc pressure effects and filler material addition. Also, with higher welding current, surface deformation becomes an issue. Using volume of fluid (VOF) method, the present method can be extended to include the surface depression effects during welding. A future goal is to study stress relief methods on welded joints." + ] + }, + { + "image_filename": "designv11_84_0001185_gt2015-42329-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001185_gt2015-42329-Figure4-1.png", + "caption": "FIGURE 4: Boundary conditions for CFD of the pre-swirl chamber", + "texts": [ + " This is necessary to minimize the pressure load on the rotor, which leads to high axial load on the bearings. Bypass 2 results from the sealing of the traversing casing to the stationary parts. One important feature of the test rig is the continuously adjustable pre-swirl before the seal. There are two mass flow inlets in the pre-swirl chamber to simulate the flow conditions in a real engine with pre-swirl ratio K between 0-0.5. K = 1 A \u222b vtangdA vRotor,max (1) A is defined as a XY-plane in the pre-swirl chamber between the two inlets (see Figure 4 and 7). This position was chosen to cover for all axial flow to the seal and to minimize the influence of the inlet flows. The functionality of the design is based on the mixing of a radial mass flow and a tangential mass flow with a high tangential velocity. The tangential velocity is dependent on the mass flow through the according inlet and limited by choking. In a first step this system is simulated to optimize the dimensions of the pre-swirl chamber. The result is shown in Figure 3. The chamber is rotationally symmetric for a 30 \u00b0 wedge", + " The first four cases a1 to a4 have the same total inlet mass flow and differ only in their mass flow ratio L. L = m\u0307tang m\u0307rad + m\u0307tang = m\u0307tang m\u0307tot (2) These cases are investigated for one specific seal and pressure ratio with changing pre-swirl. The seal defines the total mass flow, but m\u0307tang determines the pre-swirl K. The other two cases b1, b2 have a constant tangential mass flow m\u0307tang to show the influence of the radial mass flow m\u0307rad . The boundary conditions for the simulation are shown in Figure 4. There are two inlets and two outlets in the domain. A mass flow inlet condition with a turbulence intensity I of 5 % is set at both inlets. The turbulence intensity represents a fully developed pipe flow. The outlet was chosen as a 1 bar pressure outlet in order to avoid significant disturbances. The turbulence intensity I was set to 4 Copyright \u00a9 2015 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/30/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use 10 % to consider the influence of the real seal located downstream" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003817_ifsc.2013.6675642-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003817_ifsc.2013.6675642-Figure2-1.png", + "caption": "Fig. 2. Flexible Joint Manipulator System", + "texts": [ + " This feature causes some of the paths come closer to each other while moving away from the other paths. In the following, these attractors are shown for the Lorenz system that is described by the following state equations: ( ) ( ) x y x y x r z y z xy bz \u03c3= \u2212\u23a7 \u23aa = \u2212 \u2212\u23a8 \u23aa = \u2212\u23a9 (1) In this system, it is assumed that 10\u03c3 = and 8 / 3b = constant and by changing r we can observe stable, periodic and chaotic attractors in Fig. 1 III. DESCRIPTION OF THE FLEXIBLE JOINT MANIPULATOR The flexible joint manipulator system considered in this work is shown in Fig. 2, where \u03b8 is the tip angular position and \u03b1 is the deflection angle of the flexible link. The base of the flexible joint manipulator which determines the tip angular position of the flexible link is driven by servomotor, while the flexible link will response based on base movement. The deflection of link will be determined by the flexibility of the spring as their intrinsic physical characteristics [16]. In the following a brief description on the modeling of the flexible joint manipulator system is provided, as a basis of a simulation environment for development and assessment of the nonlinear control" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002036_978-94-007-2069-5_11-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002036_978-94-007-2069-5_11-Figure4-1.png", + "caption": "Fig. 4 Dynamic model", + "texts": [ + " Table 1 lists technical parameters of the gearbox, which are used for subsequent calculations. The gearing on Fig. 1a is a combined six-member mechanism (including frame). On Fig. 3 is release of each member. It should be determine 38 unknown forces and torques, torques Mtk (torsion-bar spring preload) a M2 (load on output shaft). are parameters of equation system. There are 30 equilibrium equations, the remaining eight equations resulting from the geometry of toothing (four meshes, for each one two equations). For the dynamic analysis was created model by Fig. 4. The gear unit is connected by a ball screw with a weight, which is constrained by a ball linear guide. The input of model is known desired kinematics of weight. Subsequently, the reaction forces within the gearbox and input torque course are calculated. Mainly are controlled forces on meshes of pinions. To correct function of gearbox, they must be positive at all events. For dynamic solution is necessary to know geometric characteristics as well as characteristics of mass. Weight mass was chosen of 100 kg" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002923_gt2013-95585-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002923_gt2013-95585-Figure1-1.png", + "caption": "Figure 1. LEAF SEAL SCHEMATIC", + "texts": [ + " [3] in 1994, showed that in a T-700 engine unnecessary compressor discharge seal leakage can cause a specific fuel consumption (sfc) penalty of 3 to 5%. Second, and arguably more important, a seal must retain its performance. Changes in seal characteristic with time cause changes in the secondary air-system flow distribution, which may cause further problems downstream in the system. A seal that potentially meets both these requirements is the leaf seal, first disclosed by Mech [4], and which can be seen as an evolution of the brush seal. A schematic of such a leaf seal shown in Fig. 1. The operation and performance of such a seal was presented by Nakane et al. [5] and Jahn et al. [6]. Both demonstrated that the seal performs better than a comparable labyrinth seal and has the potential to retain performance for longer. This is the case as the low seal stiffness [6] minimises wear during rubbing contact and facilitates air-riding, two effects that ensure seal geometry, and consequently seal performance, is not altered with time. Air-riding was first identified by Nakane et al" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000833_978-3-540-77974-2_13-Figure13.10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000833_978-3-540-77974-2_13-Figure13.10-1.png", + "caption": "Figure 13.10 The C-obstacle of a rotating and translating robot", + "texts": [ + " More precisely, the cross-section of CPi with the plane h : \u03c6 = \u03c60 is equal to Pi \u2295R(0,0,\u03c60). (More precisely, it is a copy of the Minkowski sum placed at height \u03c60.) Now imagine sweeping a horizontal plane upwards through configuration space, starting at \u03c6 = 0 and ending at \u03c6 = 360. At any time during the sweep the cross-section of the plane with CPi is a Minkowski sum. The shape of the Minkowski sum changes continuously: at \u03c6 = \u03c60 the cross-section is Pi \u2295R(0,0,\u03c60), and at \u03c6 = \u03c60 + \u03b5 the cross-section is Pi \u2295R(0,0,\u03c60 + \u03b5). This means that CPi looks like a twisted pillar, as in Figure 13.10. The edges and facets of this twisted pillar, except for the top facet and bottom facet, are curved.300 Section 13.5* MOTION PLANNING WITH ROTATIONS So we know more or less what C-obstacles look like. The free space is the complement of the union of these C-obstacles. Due to the nasty shape of the C-obstacles, the free space is rather complicated: its boundary is no longer polygonal, but curved. Moreover, the combinatorial complexity of the free space can be quadratic for a convex robot, and even cubic for a non-convex robot" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001529_icit.2015.7125077-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001529_icit.2015.7125077-Figure1-1.png", + "caption": "Fig. 1. Linear stage.", + "texts": [ + " Section II shows the position control system of the stage, and Section III explains the PSO algorithm and an implementation method for the control system of positioning stage. The initial placement of the particles is explained in Section IV, and the suitable 978-1-4799-7800-7/15/$31.00 \u00a92015 IEEE 63 placement in consideration of parameter error is discussed in Section V. Finally, conclusions are presented in Section VI. II. LINEAR STAGE The linear stage that is considered in this paper is shown in Fig. 1. The weight of the stage is 4 kg, and its dimension is 270 mm wide by 180 mm long by 15 mm height. The actuator of the stage is a coreless linear synchronous motor. Due to the electromagnetic force generated by coil in the stage and permanent magnets, the stage moves along the linear guide. The position of the stage is detected by using an optical encoder attached on the side of the linear guide. Fig. 2 shows the experimental setup. The position signal of the stage is fed back to a digital signal processor (DSP), which determines control input in accordance with a control law" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003729_detc2011-48019-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003729_detc2011-48019-Figure7-1.png", + "caption": "FIGURE 7: MODE SHAPES OF KART FRAME", + "texts": [ + " The maximum and minimum principal strain values at the measurement point are compared between the experimental and simulation result in Fig. 6 (b). The excellent agreement on the principal strain confirms that the FEM model is representative enough of the actual frame. 2 Copyright c\u00a9 2011 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/30/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use A modal analysis with the FEM model was performed by using ANSYS to evaluate the frequency characteristics of the kart frame. Eight eigenmodes in the frequency range up to 200 Hz are shown in Fig. 7. Next, the frequency response of the actual kart frame was evaluated by hammering test. The experiment was conducted assuming free-free constraint condition by hanging the frame in three points. In this experiment, the frame has been excited with the hammer at point (1) in Fig. 5, and the strain at point (b) was measured. The experimental layout is shown in Fig. 8. The power spectral density of measured strain was shown in Fig. 9. The frequency resonance about 45 Hz and 138 Hz is well observed, which is considered to correspond to the 1st and the 6th mode in Fig. 7. In addition, the peaks at the frequency of the 2nd and the 7th modes can also be found. On the other hands, other eigenmodes could not be found in this test. The hammering position, the sensing position, and the mode shape may effect the observability of the eigenmodes. The FEM frame model has a large number of elastic degree of freedom. This large number of elastic dofs is highly unpractical for multibody simulations environments, causing long CPU time for simulation. In order to reduce the computational effort, model reduction method based on the modal analysis is conducted" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002694_amm.87.200-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002694_amm.87.200-Figure1-1.png", + "caption": "Fig. 1 The physical construction of ultra precision positioning system.", + "texts": [ + " developed equations of the tooth profiles for the gear set and led to the mathematical modeling of the spiral bevel gear [8]. This paper focuses on the modeling and validation of ultra precision positioning system. In the following sections, firstly, two different dynamical models have been developed to describe the system. Then, the comparison between the two models was done after the system identification of unknown parameters. The ultra precision positioning system including inside stage and outside stage is shown in Fig. 1. The outside stage is installed on the liner guide and driven by the lead screw which is fixed on the end of the outside stage. The inside stage is hung by four leaf springs fastened to the other end of the outside stage. As shown in Fig. 1, a piezoelectric ceramic is installed between the two stages and fixed on the outside stage. The other side of the piezoelectric ceramic is free but closely appressed to the inside stage by pre-tightening springs. The positioning system is a macro/micro dual-drive system. The macro positioning means that the outside stage driven by the motor through the lead screw while the micro positioning means the inside stage driven by the piezoelectric ceramic. When the positioning system is in operation, the workpiece is installed on the inside stage" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002336_icsd.2013.6619912-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002336_icsd.2013.6619912-Figure5-1.png", + "caption": "Fig. 5. Number of jumper", + "texts": [ + " The electric field strength on section 1~3 is comparative low for the shielding effect of high voltage wires. As farther and farther 978-1-4673-4461-6/13/$31.00 \u00a92013 IEEE away from the wires, the electric field strength on section 4~6 is gradually increasing. The maximum electric field strength is 2259 V/mm, which appears on cross section 6. The electric field strength on section 7~8 is gradually decreasing for the shielding effect from the line end of suspension string. Subconductors of jumper were numbered as illustrated in Fig. 5 for convenience. When the jumper is on the right of the tower, subconductors is arranged clockwise. Fig. 6 shows the electric field distribution curve along the surface of each subconductor. It can be seen that the overall distribution pattern of all subconductor are basically the same and the maximum electric field strength were located in similar place. Owing to the shielding effect of conductors, the electric field distribution curve is low both at the beginning and the end. In the middle of the curve, with the decrease of the shielding effect of conductors and curvature radius of jumper sag, the electric field strength is high" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002695_s12541-014-0448-0-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002695_s12541-014-0448-0-Figure5-1.png", + "caption": "Fig. 5 (a) Schematic of simulated tool. (b) Experimental tool. (c) Schematic of experimental system", + "texts": [ + " Similar to the calculation for M2 in the previous section, we have the time when contacts and leaves the on the workpiece surface, where can be estimated as (22) where, (23) The traces of can be expressed as (24) \u2205\u2206 lmt\u2013 i2\u03c0+ nT ------------------------------ tM 1 \u2205\u2206 lmt i2\u03c0+ nT --------------------------\u2264 \u2264 tM 1 i[ ] Lwt M1 i tM 1 ij( ) tM 1 i0 tM 1 imax j{ } Lwt M1 i tM 1 ij( ) Lwt M1 i tM 1 ij( ) zw M1 ij ftM 1 ij RT nTtM 1 ij( )sin+= \u03b8w M1 ij nwtM 1 ij = \u23a9 \u23aa \u23a8 \u23aa \u23a7 = tM 2 ij[ ] Lwt M2 i tM 2 ij( ) tM 2 ij tM 2 ij tM 1 ij t\u2206 M+= Lwt M2 i tM 2 ij( ) Lwt M2 i tM 2 ij( ) zw M2 ij f tM 2 ij( ) RT nT tM 2 ij( )sin+= \u03b8w M2 ij nw tM 2 ij( )= \u23a9 \u23aa \u23a8 \u23aa \u23a7 = M 1 k M 2 k M 1 0 M 1 \u2261 M 2 0 M 2 \u2261 M 1 k t M 1 k ij M 1 k Lwt M 1 k i t M 1 k ij \u239d \u23a0 \u239b \u239e t M 1 k ij t M 1 k ij tM 1 ij t\u2206 M 1 k += t\u2206 M 1 k k2\u03c0 /ne nT ---------------= Lwt M 1 k i t M 1 k ij \u239d \u23a0 \u239b \u239e Lwt M 1 k i t M 1 k ij \u239d \u23a0 \u239b \u239e z M 1 k ij f t M 1 k ij \u239d \u23a0 \u239b \u239e RT nT t M 1 k ij \u239d \u23a0 \u239b \u239esin+= \u03b8 M 1 k ij nw t M 1 k ij \u239d \u23a0 \u239b \u239e= \u23a9 \u23aa \u23aa \u23a8 \u23aa \u23aa \u23a7 = The number of marking areas, np, on a single cycle of cutting trace on the workpiece depends on the rotation speeds of the tool and the workpiece, and the number of cutting edges as (27) The angle of each marking area and z-coordinate depends on the rotation speeds of the tool and the workpiece as (28) As shown in Fig. 3(b), the axial length of each marking area depends on the rotation speeds of the tool and the workpiece, the central angle of cutting edge, and the radius of tool as (29) Experiments were conducted to confirm the result of the theoretical approach. The circular tool was used with multiple cutting edges coinciding to the marking lines in the analysis section. The schematic of the tool for modeling and the real tool for experiments are shown in Fig. 5(a) and (b), respectively. Experimental setup was designed using a CNC lathe. Fig. 5(c) shows a schematic of the experimental system for the driven rotary tool. The circular tool is kept by the tool case placed on the tool holder of the CNC lathe and connected to the motor via flexible axis. A servo motor and its driver were used to control the rotational speed as well as rotational direction of the circular tool (clockwise or counter clockwise direction). The workpiece for experimental verification was MC nylon. The experimental procedure carried out as following, Step 1: Surface machining, the actual workpiece is machined by the fixed circular tool to obtain the cylindrical workpiece with the dimension and outside surface quality satisfying design requirements" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003273_time-e.2014.7011622-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003273_time-e.2014.7011622-Figure7-1.png", + "caption": "Figure 7. The example of circular path trajectory and parameters.", + "texts": [ + " Where, vx and vy are the velocity of x, y axis, vf is the velocity of the feed milling. However, the ability to move in a straight line at an angle (\u03b8) in the horizontal plane is limited by the features of the machine. For this angles are in the approximately range 15\u00b0 to 75\u00b0 of each quadrant. Creating a circular path trajectory to determine the motion of a tip of the tool milling machine, there are basically two types of motion, clockwise (G02) and counterclockwise (G03). User commands to specify the value of the starting point, ending point, and radius of a circle. Fig. 7 shows an example of circular path trajectory. Where (x1,y1) is starting point, (x2,y2) is ending point, (x3,y3) is a half of L, L is the displacement from (x1,y1) to (x2,y2), (x4,y4) is the center, and r is the radius of circle. 2 00 2 2 0 a s2s K )s(V )s( ++ = (1) 006056.1035s862961.13s 006056.1035 )s(V )s( 2 a ++ = 970496.1360s678645.38s 164595.1633 )s(V )s( 2 a ++ = 633553.624s495233,34s 633553.624 )s(V )s( 2 a ++ = (2) (3) (4) 2 12 2 12 12 ffy 2 12 2 12 12 ffx )yy()xx( yyvsinvv )yy()xx( xxvcosvv \u2212+\u2212 \u2212== \u2212+\u2212 \u2212== (5) (6) When \u03b2 is the angle that was used as a parameter for selecting the number of points on a circular path, in case the user wants to move the tip of the milling tool just some of the circle, which can be calculated from the cosine rule following" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001440_ever.2015.7112927-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001440_ever.2015.7112927-Figure4-1.png", + "caption": "Fig. 4. The \u2019V\u2019 IPM rotor: main geometric variables.", + "texts": [], + "surrounding_texts": [ + "Any mechanical piece that vibrates in the air produces noise. In electric machines, magnetic vibrations are mainly due to the excitation of the mechanical system by harmonics of air-gap Maxwell stress, also called electromagnetic forces. Our electromagnetic model calculates radial component of Maxwell stress on the average air-gap. These pressures are expressed by Eq.1 [11], where Br and B\u03b8 stand for the radial and tangential components of the air-gap flux density, respectively. \u03b1 is the spatial air-gap discretization, and t is the time discretization. \u03c3r(t, \u03b1) = 1 2\u00b50 [ B2 r (t, \u03b1)\u2212B2 \u03b8 (t, \u03b1) ] (1) The total air-gap flux density, B(t, \u03b1), is the product between the conjugate of the air-gap permeance \u039b(t, \u03b1) and the sum of magnetomotive forces developed by the stator winding fsmm(t, \u03b1), and PM rotor frmm(t, \u03b1) (exponents s and r mean stator and rotor respectively). Using complex notations, we have: B = Br + jB\u03b8 = \u039b\u2217 \u00d7 [Fr mm + Fs mm] (2) where B, \u039b and F are nt \u00d7 na matrix of complex flux density, complex air-gap permeance per area unit and complex magnetomotive forces respectively. na and nt are number of discretization in the space domain (following \u03b1) and in the time domain (following t) respectively. With this formulation, it is possible to apply all the mathematical operations to a single variable, grouping two data." + ] + }, + { + "image_filename": "designv11_84_0002772_amm.658.495-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002772_amm.658.495-Figure3-1.png", + "caption": "Fig. 3. Fix system of reference axes", + "texts": [ + " The analyses of the movement of the kinematic chain arm-forearm-hand was realized with the aid of the direct kinematic, by expressing the coordinates of the contact point between hand and ball, towards the axes system (x\u2019O\u2019y\u2019), attached to the shoulder joint (Figs. 1 and 2) The coordinates of the contact points between hand and ball, expressed as function of the flexion angle of the arm, are: \u03d5\u22c5+\u03d5\u22c5+\u03d5\u22c5=\u2032 \u03d5\u22c5+\u03d5\u22c5+\u03d5\u22c5=\u2032 332211 332211 sinqsinqsinqy cosqcosqcosqx , (1) where: q1, q2, q3 \u2013 are the length of the arm, the forearm and the hand, respectively, 1\u03d5 , 2\u03d5 , 3\u03d5 \u2013 are the flexion angles of the arm, forearm and hand respectively, When reporting the contact point coordinates between hand and ball towards the fix coordinated axes (xOy), represented in Fig. 3, one could write the relations: \u03d5\u22c5+\u03d5\u22c5+\u03d5\u22c5+= \u03d5\u22c5+\u03d5\u22c5+\u03d5\u22c5+\u2212= \u21d2 \u2032+= \u2032+\u2212= 3322111 332211 1 sinqsinqsinqhy cosqcosqcosqLx yhy xLx , (2) where: L \u2013 represents the throwing distance, By derivation with time the Eq. (2) equation system, one could obtain the initial speed of the ball: \u03d5\u22c5\u22c5\u03c9+\u03d5\u22c5\u22c5\u03c9+\u03d5\u22c5\u22c5\u03c9= \u03d5\u22c5\u22c5\u03c9\u2212\u03d5\u22c5\u22c5\u03c9\u2212\u03d5\u22c5\u22c5\u03c9\u2212= 333222111y0 333222111x0 cosqcosqcosqv sinqsinqsinqv (3) where: 1\u03c9 , 2\u03c9 , 3\u03c9 \u2013 are the angular speed of the arm rotation, the forearm and the hand, respectively. Using these components of the initial angular speed of the ball, the initial throwing angle could be determined, as follows: =\u03b1 \u2212 x0 y01 0 v v tg " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure16.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure16.2-1.png", + "caption": "Fig. 16.2. Experimental teleoperation system of one DOF", + "texts": [ + " As the time delay was bigger and the operator force changed more abruptly, the force error will be bigger. This force error will produce a position error between the master and the slave that will increase as the force error increases. However, the proposed control system is oriented to cases where there is an \u201csmall\u201d time delay in the communication channel. Therefore the force error will not be important and the position error will be small. The control method by state convergence has been tested on a experimental teleoperation system of one DoF, Fig. 16.2, considering a constant time delay [10]. The next state equations of the master and the slave have been experimentally identified: x\u0307m(t) = [ 0 1 0 \u22127.1429 ] xm(t) + [ 0 0.2656 ] um(t) (16.63) 282 J.M. Azor\u0301\u0131n et al. po si tio n ym(t) = [ 1 0 ] xm(t) (16.64) x\u0307s(t) = [ 0 1 0 \u22126.25 ] xs(t) + [ 0 0.2729 ] us(t) (16.65) ys(t) = [ 1 0 ] xs(t) (16.66) where ym and ys are the master and slave positions. These state equations have been obtained from second order differential equations that do not contain zeros, so it is possible to achieve that the slave follows the master, and to establish the dynamics of the slave-master error and the slave" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000947_978-3-540-73958-6_5-Figure5.10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000947_978-3-540-73958-6_5-Figure5.10-1.png", + "caption": "Fig. 5.10. Angles and air speed at one blade element", + "texts": [ + " For determination of mean values of forces, all components of each rotor element have to be integrated according to Fig. 5.9. The blades are assumed to be non-twisted and of constant chord from radius R1 to R2. For calculation of the force dF of one blade element the relative air speed u at this element is needed. It can be divided into two parts: vertical component uz and horizontal perpendicular to the blade ur. The third component, parallel to the blade, is not relevant for calculated forces. ur = \u03c9Mr \u2212 cos(\u03b1)vwy + sin(\u03b1)vwx (5.15) uz = vwz (5.16) Figure 5.10 shows the relevant relationships of angles \u03c6 and forces. The pitch angle \u03c6, which consists of collective pitch Pc and two cyclic pitches Px and Py 1, 1 The signs of Px and Py depend on the rotating direction of the rotor. has to be computed as well as the direction of air flow \u03b3, in order to obtain the angle of attack \u03b4. \u03c6 = Pc cos(\u03b1)Px + sin(\u03b1)Py (5.17) \u03b3 = atan2(uz, ur) (5.18) \u03b4 = \u03c6 + \u03b3 (5.19) According to [33, 22] lift dL (drag dD) is defined to be perpendicular (parallel) to the incoming air flow" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001210_iciafs.2014.7069541-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001210_iciafs.2014.7069541-Figure3-1.png", + "caption": "Fig. 3. Overall structure of the CSM.", + "texts": [ + " This paper presents the utility of force transmission in the 2-DOF system by the proposed disturbance model. Bilateral control test is carried out to validate the utility. This paper is organized as follows. In section II, the structure of CSM is presented. In section III, bilateral control method is described. In addition, disturbance modeling procedure and that results are also described. In section IV, experimental set up and results are given and section V is conclusion. 978-1-4799-4598-6/14/$31.00 c\u20dd2014 IEEE Fig. 2 shows structure inside of the CSM. Fig. 3 shows overall structure of the CSM. Table I shows the specifications of stator and mover [4]. The mover part consists of a mover case and a mover bobbin with coils, and the stator part consists of bent pipe and magnets. Both edges of stator are fixed by two retainers. The stage moves along the guide rail that is bent as same as the pipe. The mover part is fixed on the stage. The optical linear encoder attached on the side of the stage reads position information. The developed 2-DOF system does not use force sensor" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003729_detc2011-48019-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003729_detc2011-48019-Figure2-1.png", + "caption": "FIGURE 2: 3D GEOMETRICAL MODEL.", + "texts": [ + "org/about-asme/terms-of-use the racing kart was completed, and the model has been used to simulate the kart dynamic behavior. The running test of the actual racing kart was also performed, and the comparison between simulation and experiment of the dynamic vehicle behavior results are discussed in this paper. The kart frame is the main component of the racing kart to which the all other components are connected. Figure 1 shows the actual kart of our laboratory. The kart frame is built from steel tube with an outer diameter of 30 mm and a wall thickness of 2 mm. A 3D geometry model of the kart frame (Fig. 2) was created using the CAD software Pro/Engineer. The geometric data for the model was obtained by measurements using a 3D measurement device in our laboratory. The 3D geometry model of the frame was imported in the commercial FEM software ANSYS to analyze the structural behavior of the kart frame. The CAD geometry data frame was meshed with shell elements that is a four node element with six degree of freedom at each node. Several interface nodes [4] were prepared to define the joints in the process of flexible multibody dynamics simulation described in the later section of this paper" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003720_detc2014-34145-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003720_detc2014-34145-Figure3-1.png", + "caption": "FIGURE 3. SCHEMATIC FOR THE FREE FALL OF A SINGLE PARTICLE.", + "texts": [ + " [26], we will use a simplified model of the electrical motor, considering cm = 0, Tf = 0, and that the electrical time constant L/R is smaller than the mechanical time constant RJ/KEKT leads to the simplified electrical and mechanical equations of the motor, for a constant voltage: V = Ri\u2212KE \u03c6\u0307 Mm = KT i\u2212 Jm\u03c6\u0308 (16) Considering that the rotor moment of inertia, Jm, is included in the moment of inertia of the rotating parts is, J, we can expressed Mm as Mm(\u03c6\u0307) = M0(1\u2212 \u03c6\u0307 \u21260 ) (17) where M0 = KTV/R is the torque constant of the motor and \u21260 = V/KE is the angular velocity constant of the motor (see Dimentberg et al. [27]). In this section we simulate a single smooth frictionless spherical particle that is dropped from a specified height. The particle freely falls under gravity and bounces upon collision with a fixed wall. Figure 3 shows a model of a ball falling on a stationary surface. This classical bouncing ball problem has been applied by several authors (e.g. [28], [29], [30], [31], [32], [33]). Following the work of [30] and [28], we have studied a single 0.2m particle diameter falling and bouncing in a wall. The particle parameters are taken from [28] and reproduced in Table 2. The impact velocity of the particle with the wall is v0 = \u2212 \u221a 2g(h0\u2212 rp) = \u22122.801m/s, where rp is the particle radius. The MFIX\u2013DEM code has been used in the simulations for determining the position and velocity time-history" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002868_s12206-015-0143-9-Figure20-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002868_s12206-015-0143-9-Figure20-1.png", + "caption": "Fig. 20. Illustration of the moving-CG point.", + "texts": [ + " 18 shows the open loop bu response from the manual tilt angle input, where it shows good agreement, but some portion of the data exhibits disagreement because of external gusts. Fig. 19, in contrast, shows controlled bu response and prediction, and the predicted \u03b3 as internal state. The tilt angle in Fig. 19 is the output from the forward velocity controller, which is the predicted \u03b3 showing a difference from the actual data to yield a slightly different bu response. In case of the tilt-prop, tilting action of the nacelle may move the CG point as illustrated in Fig. 20. The static state moment arm length sl is shown to move back when there\u2019s a positive vertical CG offset for nacelles. It is assumed that frontal and rear arm-length at tilted state, T Fl and T Bl , respectively, can be changed from \u03b3 and the unbalanced weight of the nacelles around the pivoting point as: cos sinT P F s Zl l lg g= + , (1) cos sinT P B s Zl l lg g= - , (2) where P Zl is the vertical offset of the nacelles\u2019 overall center of mass from the pivoting point. Although this modeling is not exact, the concept of the moving-CG point was required to assess the pitching moment exhibited during tilted forward flight, and the value of P Zl was identified as 0", + "1, where the Euler angles \u03d5, \u03b8 and \u03c8 are called the roll, pitch, and yaw angles, respectively. Four motors create thrusts ( 1T ~ 1T ) and torques ( 1Q ~ 4Q ) with arm length sl . When the nacelles are tilted, and if their weights are not evenly balanced around their pivoting point, the forward and backward arm lengths may not be the same because of the change in the CG position. Therefore, in a quad tilt configuration, the forward arm length T Fl and backward arm length T Bl are separately illustrated. It was illustrated in Fig. 20 for more intuitive description for the moving-CG point. The dynamic equations are given in Eqs. (A.1) and (A.2): ( )T b b b tot m m= + \u00b4F V \u03c9 V& , (A.1) ( )T b b b b b tot = + \u00b4 \u00d7M I \u03c9 \u03c9 I \u03c9& , (A.2) where T totF and T totM are the total forces and moments in the tilted state, and m is the total mass. Superscript T is used to distinguish the tilted state from vertically fixed state. The body frame velocities are given as: b b b b u v w \u00e9 \u00f9 \u00ea \u00fa = \u00ea \u00fa \u00ea \u00fa \u00ea \u00fa\u00eb \u00fb V . (A.3) Body frame angular rate b\u03c9 and Euler angle rate \u03c9 can be converted with a conversion matrix C as: b p q r f q y \u00e9 \u00f9\u00e9 \u00f9 \u00ea \u00fa\u00ea \u00fa= = \u00ea \u00fa\u00ea \u00fa \u00ea \u00fa\u00ea \u00fa\u00eb \u00fb \u00ea \u00fa\u00eb \u00fb \u03c9 C & & & , 1 p q r f q y - \u00e9 \u00f9 \u00e9 \u00f9\u00ea \u00fa \u00ea \u00fa= =\u00ea \u00fa \u00ea \u00fa\u00ea \u00fa \u00ea \u00fa\u00eb \u00fb\u00ea \u00fa\u00eb \u00fb \u03c9 C & & & , (A" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002494_amr.1028.105-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002494_amr.1028.105-Figure7-1.png", + "caption": "Fig. 7 Vibration mode of order 2", + "texts": [], + "surrounding_texts": [ + "The static structural analysis of the drive shaft is used for pre-stressed modal calculation in Model module, therefore the natural frequencies and corresponding vibration modes of the shaft can be got. Normally we do not have to find all the natural frequencies and mode shapes. Low order natural frequencies and vibration modes have bigger impact on the vibration of the stepper motor [4],so the first six natural frequencies and vibration modes obtained in ANSYS Workbench are concerned, as shown in Figure 6-11 and Table 1.Because the amplitude is relative value after treatment, it doesnot reflect the actual amplitude [3]. 1) Natural frequency analysis. Through access to relevant information, the stepper motor has a fixed resonance region. The resonance region of the two-four phase stepper motor is generally between 180 and 250PPS (step angle of 1.8 degrees). As the drive voltage of the stepper motor is higher and the load is lighter, the resonance region is upward. However, each order natural frequency of the drive shaft is more than 1745.2HZ, so to ensure the tube reaches the specified location accurately and avoid resonance between the stepper motor and drive shaft, the operating frequency of the stepper motor must be between 300PPS and 1700PPS. 2) Vibration mode analysis. Though the analysis of the first two modal shapes (shown in Figure 6, 7), the central part of the shaft has the greatest amplitude of the resonance, therefore probably becomes the weakest portion. So without affecting the transmission accuracy, belt pulleys are arranged on both sides of the shaft as far as possible." + ] + }, + { + "image_filename": "designv11_84_0002880_ilt-06-2011-0045-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002880_ilt-06-2011-0045-Figure1-1.png", + "caption": "Figure 1 Squeezing film geometry between a long cylinder and an infinite plate, considering the PVB of lubricants", + "texts": [ + " An analytical film-pressure solution is derived from the nonlinear Reynolds-type differential equation. Compared with the case of CVL, the effects of PVB on the load capacity and the approaching time for the squeeze films are presented and discussed through the variation of the piezo-viscosity parameter. The current issue and full text archive of this journal is available at www.emeraldinsight.com/0036-8792.htm Industrial Lubrication and Tribology 66/3 (2014) 505\u2013508 \u00a9 Emerald Group Publishing Limited [ISSN 0036-8792] [DOI 10.1108/ilt-06-2011-0045] Figure 1 presents the squeezing film geometry between a long cylinder and an infinite plate, considering the PVB of lubricants. From Hamrock (1994), the film thickness h in the lubrication region can be approximated by the following relationship: h hc x2 2a (2) where hc is the central film thickness, a is the radius of the cylinder and x is the horizontal coordinate. The cylinder is approaching the plate with a squeezing velocity: h\u0307c, where h\u0307c dhc/dt and t denote the time. Assume that the fluid film is thin in the z-direction and the lubrication theory is applicable" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002479_amm.668-669.729-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002479_amm.668-669.729-Figure6-1.png", + "caption": "Fig. 6. The velocity contour of two fans under 40% load condition The left for the original fan and right for the optimized fan", + "texts": [ + " To sum up, the curvature radius of the volute exit type line is smaller, acceleration process is slower, and the volute exit pressure fluctuation is smaller than the original fan. Result analysis of variable conditions The Fig. 4 shows that compared with the original fan, the static pressure change of optimized fan is relatively stable under variable conditions, and the outlet static pressure of optimized fan is higher than the original fan under 40% load condition. Analyze flow field of two kinds of fan volute under 40% load condition, as shown in Fig. 6. Compared with the internal flow field in the volute, can find the right one more bigger than the vortex area, there are bigger energy loss. The original fan cause vortex under 40% load condition, the original fan flow area near the volute tongue volute is too large, the ring wall of volute weak restriction in fluid flow area. But the optimized fan volute tongue is lesser; volute ring wall of the fluid restriction effect is stronger, as shown in Fig.6. As a result, the optimized fan near the volute tongue does not appear larger vortex. In summary, the optimized impeller center improves export static pressure and makes export static pressure fluctuation small under variable condition runtime. At the same time, leak loss of volute tongue is small, the volute exit pressure fluctuation is small and less large eddy is few near the volute tongue in volute. So, the fan efficiency and load ability were increased as a whole. (1) The impeller center in (2, 2) is the best impeller center position of the fan" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002019_peac.2014.7038056-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002019_peac.2014.7038056-Figure6-1.png", + "caption": "Fig. 6. The ESM steady vector graph in DTC scheme", + "texts": [ + " The rules of the torque and flux magnitude hysteresis comparator are as follows: 1 0 1 * e e T * T e e * e e T T T c T T T T \u03b5 \u03b5 \u23a7 \u2212 \u2265 + \u23aa = =\u23a8 \u23aa\u2212 \u2212 \u2264 \u2212\u23a9 (10) 1 0 * s s * s s c \u03b5 \u03b5 \u03a8 \u03a8 \u03a8 \u23a7 \u03a8 \u2212 \u03a8 > +\u23aa= \u23a8 \u03a8 \u2212 \u03a8 < \u2212\u23aa\u23a9 (11) where \u03b5T and \u03b5\u03a8 are the torque and flux magnitude hysteresis comparator bands respectively. Te is calculated based on (3), and \u03a8s is estimated as the same with the FOC scheme. Assuming a stator flux linkage reference frame, i.e. the m-t coordinates, we have the vector graph as in Fig. 6. The stator flux linkage \u03a8s in m-t coordinates can be expressed as: cos sin sin cos 0 sd sd sm md fmsm sm sm s sq sq st mq ftst sm sm L i L i L i L i \u03b4 \u03b4 \u03b4 \u03b4 \u03a8 +\u03a8 \u03a8\u23a1 \u23a4 \u23a1 \u23a4\u23a1 \u23a4 \u23a1 \u23a4 \u23a1 \u23a4 = = =\u23a2 \u23a5 \u23a2 \u23a5\u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5\u03a8 +\u03a8 \u2212 \u23a3 \u23a6\u23a3 \u23a6 \u23a3 \u23a6 \u23a3 \u23a6 \u23a3 \u23a6 (12) The d-axis and q-axis current can be expressed as: cos sin sin cos sd sm sm sm sq sm sm st i i i i \u03b4 \u03b4 \u03b4 \u03b4 \u2212\u23a1 \u23a4 \u23a1 \u23a4 \u23a1 \u23a4 =\u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a3 \u23a6 \u23a3 \u23a6\u23a3 \u23a6 (13) From (3), (12) and (13), we can obtain: mq st ft sq e p st st L i i L T n i \u23a7 = \u2212\u23aa \u23a8 \u23aa = \u03a8\u23a9 (14) The rotor field current if can be expressed as: 2 2 2 2 sq f fm ft fm st mq L i i i i i L \u239b \u239e = + = + \u239c \u239f\u239c \u239f \u239d \u23a0 (15) From (12) and (15), assuming ism=0, and note that Lsq\u2248 Lmq, we can get the reference rotor field current if * as follow: 22 2 2 * * * * * s e f fm ft * md p s T i i i L n \u239b \u239e\u239b \u239e\u03a8 = + = + \u239c \u239f\u239c \u239f \u239c \u239f\u03a8\u239d \u23a0 \u239d \u23a0 (16) IV" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001136_978-3-319-19788-3_1-Figure1.10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001136_978-3-319-19788-3_1-Figure1.10-1.png", + "caption": "Fig. 1.10 Symbolic representation of a revolute joint", + "texts": [ + " The resulting number of degrees of freedom, m, is also called joint mobility, such that 0 \u2264 m < 6. When m = 1, which is frequently the case in robotics, the joint is either revolute or prismatic. A complex joint with several degrees of freedom can be constructed by an equivalent combination of revolute and prismatic joints. For example, a spherical joint can be obtained by using three revolute joints whose axes intersect at a point. \u2022 active joint: a joint which is actuated. \u2022 passive joint: a joint which is not actuated. \u2022 R joint: a revolute joint (Fig. 1.10), allowing a rotation around a given axis. If the letter is underlined (R joint), the joint is actuated. If not, it is passive. \u2022 P joint: a prismatic joint (Fig. 1.11), allowing a rotation around a given axis. If the letter is underlined (P joint), the joint is actuated. If not, it is passive. \u2022 U joint: a universal joint, allowing two independent rotations around two given axes. These joints are usually passive and can be represented by two R joints with orthogonal and intersecting axes. \u2022 S joint: a spherical joint (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure1.47-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure1.47-1.png", + "caption": "Fig. 1.47 Rounding of the contours of the plate", + "texts": [ + "45, this is in order to not generate injury and discomfort to the patient and generate more similar forms to anatomical ones. Only the rounding tool was used here. 1 Comparative Study of Interferometry and Finite Element Analysis \u2026 29 Almost to finish, an extension was added to the same piece, as shown in Fig. 1.46 with a thickness of 4 mm, equal to that of the part. This was done so that the piece is close to that of the prosthesis. Sketching and extrusion tools were used. Finally, a rounding was generated to the outline of the piece to soften the plate, Fig. 1.47, and leave it ready for assembly with the prosthesis, see Fig. 1.48. The assembly was imported in STL format for 3D printing as shown in the results. 30 J. A. Beltr\u00e1n-Fern\u00e1ndez et al. Once the jaw was scanned, and in STL format available, PTC Creo Parametric was used to convert the part, by exporting the wrap-type level 10, to a solid piece. The solid part is shown in Fig. 1.49. Thus, with a solid piece, the healthy part (branch and right condyle) was cut to start generating our prosthesis, and a plaque was added for the binding to the healthy 1 Comparative Study of Interferometry and Finite Element Analysis \u2026 31 branch that will remain to the patient after removing the part with original cancer" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002444_20131120-3-fr-4045.00035-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002444_20131120-3-fr-4045.00035-Figure1-1.png", + "caption": "Fig. 1. Quad-rotor configurations: a) conventional; b) with tilting rotors.", + "texts": [ + " The problem of navigation and autonomous flight for this class of vehicles, as well other VTOL UAVs, are well discussed in Kendoul (2012) and Michael (2012). Control of a conventional quad-rotor is based on the variation of thrust developed by the four rotors, which is achieved by variations of rotor rotation rates. In hover all rotors provide equal thrust, such that the total equals vehicle weight. Two rotors turn in one direction and two in the opposite one, so that the overall yaw aerodynamic moment is cancelled (Fig. 1.a). Yawing control moments are obtained by unbalancing these aerodynamic moments by accelerating the rotors spinning in one direction and slowing down the other two. Pitch and roll moments are obtained by unbalancing the thrust of forward and rear or lateral rotors, respectively, for a conventional cruciform configuration. A variation of the rotation rate of all the rotors allows for vertical acceleration and maneuvers at a load factor greater than one. The possibility of tilting the rotors as depicted in Fig. 1.b allows for a significantly greater flexibility in obtaining control moments. With 4 more control variables, the system becomes redundant, and it is possible to control 978-3-902823-57-1/2013 \u00a9 IFAC 232 10.3182/20131120-3-FR-4045.00035 flight condition simply unachievable for a conventional quad-rotor configuration with fixed propellers. A patent request has been issued for the configuration discussed in this paper, Avanzini & Giulietti (2012). As an example, it is possible to accelerate and fly in forward flight without the need for pitching the vehicle down", + " In the first two case, a U turn and a 360 deg yaw rotation in forward flight, the behavior of the novel configuration is compared with that of an equivalent conventional quad-rotor. The third maneuver, a 90 deg roll rotation in forward flight, represents a test case that can be flown only if the rotors are allowed to tilt. This maneuver represents a severe challenge and it fully demonstrates that the tilting rotors provide the vehicle with a significantly expanded maneuver envelope. The quad-rotor under consideration consists of a rigid cross frame equipped with four rotors as shown in Fig. 1.a. The model is derived under the assumptions that motor and rotor response is fast and their dynamics can therefore be neglected. Also, rotor blades are assumed to be rigid (i.e. no blade flapping occurs). As outlined above, control moments can be obtained either differentially changing the value of thrust and torque of each rotor changing its angular speed or tilting the rotor so that thrust can be projected into a horizontal and a vertical component in the body frame. The configuration of the quad-rotor with all the rotors tilted is shown in Fig. 1.b. The equations of motion are developed in terms of the translational and rotational velocities represented in bodyframe components, vb and \u03c9b, where attitude is represented using quaternions, (q0, q T )T , Wertz (1978): v\u0307b =\u2212\u03c9b \u00d7 vb + F m + gb (1) \u03c9\u0307b = J\u22121 [M \u2212 \u03c9b \u00d7 (J\u03c9b)] (2) q\u03070 =\u22121 2 \u03c9b \u00b7 q (3) q\u0307 = 1 2 (q0\u03c9b \u2212 \u03c9b \u00d7 q) (4) r\u0307i = T bivb (5) where F andM indicate aerodynamic force and moments, respectively, gb is gravity acceleration, J is the inertia tensor, and T bi(q0, q T ) is the coordinate transformation matrix from inertial to body frame" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003600_12.2072609-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003600_12.2072609-Figure3-1.png", + "caption": "Figure 3 Creative design and SLM manufacturing for non-assembly abacus", + "texts": [ + " It is thought that as early as 1000 BC years or so, people started to use bronze and copper imitation shells and these slowly developed into round copper coins with a hole in the middle so that they could be strung together. Inscribed bamboo-slips were used as books longest in the Chinese history before paper spread. They are consisted of bamboo and rope bindings, which are available rewinding. Combining the copper cash or inscribed bamboo-slips with abacus, the new concept of non-assembly abacus realized the perfect combination of the ancient Chinese traditional culture and modern manufacturing methods. The design and manufacturing of two cases as shown in the Figure 3. During the design process, the size of motion clearance between breads and rods, the clearance of collapsible joint, should be taken serious consideration. If the designed motion clearance is too large, the stability of movement will be affected, if the designed motion clearance is too small, the processability of SLM manufacturing will be hard. Therefore, when design the non-assembly abacus, the manufacturing factors should be considered in the design process. One of the important factors of SLM manufacturing for the motion gap can be described as geometric features resolution" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003477_jjap.53.07kf26-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003477_jjap.53.07kf26-Figure2-1.png", + "caption": "Fig. 2. (Color online) 128-element linear array probe: (a) transducer layout and (b) photograph.", + "texts": [ + " 1(b) shows a photograph of the probe. Six linear array elements are placed on both sides of a central ring element. The central through hole is 1.6mm in diameter, the inner and outer diameters of the ring transducer are 3 and 6mm respectively, and the element pitch and width of the six-element linear array transducers are 1 and 6mm, respectively. The transducer is made of a lead zirconate titanate (PZT)-epoxy 1\u20133 composite material (0.22mm thick) and is fixed on a damper using an epoxy adhesive. The driving frequency is 7MHz. Figure 2(a) shows the structure of the 128-element ultrasonic probe, and Fig. 2(b) shows its photograph. Linear array transducers composed of 60 elements for forward viewing are placed on both the left and right sides. The element pitch and width of the array transducers are 0.15 and 6mm. The linear Japanese Journal of Applied Physics 53, 07KF26 (2014) http://dx.doi.org/10.7567/JJAP.53.07KF26 REGULAR PAPER 07KF26-1 \u00a9 2014 The Japan Society of Applied Physics array transducers composed of 8 elements for detecting the needle tip are placed at the center of the probe. The pitches of the 8 elements are 0.5, 0.4, 0.3, and 0.2mm, as shown in Fig. 2(a). The transducer is made of a PZT-epoxy 1\u20133 composite material (0.2mm thick) and is fixed on a damper using an epoxy adhesive. The driving frequency is 7.5MHz. This probe has a 2mm-wide U-shaped slit at the center for inserting a needle vertically into a lesion and for removing the probe while placing the needle. The damper and overcoat are made of epoxy resin and the probe case is made of stainless steel. The elements of the linear array parts of the 13-element probe are 1mm wide, and those of the linear array parts of the 128-element probe are 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003449_amm.670-671.1350-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003449_amm.670-671.1350-Figure4-1.png", + "caption": "Figure 4. Four-rotor aircraft stress analysis", + "texts": [ + " There are a variety of multi-rotor aircraft flight attitudes which can be categorized into four parts: hover, vertical takeoff and landing (VTOL), rotation (yaw angle changes), translation (including before and after the translation (pitch changes), left and right translation (roll angle changes)). Due to the low height and speed of multi-rotor flight, the effect of the earth's rotation, the magnetic field, air resistance on its flight attitude is ignored. Thus, there are only two coordinate systems (a reference coordinate system and a body coordinate system) need to be considered inside the experiments. While in the initial state, assuming these two coordinates have the same origin and direction of each axis. Figure 4 is a stress analysis chart of a four-rotor aircraft. In Figure 4, Fi says each rotor lift, i\u2208{1,2,3,4}. Assuming the quality of the four-rotor aircraft is m, and the rotor speed is \u2126, due to its speed is proportional to the rotor to provide lift, so there are Fi=b\u2126j2, j\u2208{1,2,3,4}. S1 represents the sum of the lift provided by each of the rotor. Assuming the positive direction of the y-axis direction is the head of the body, or the positive direction of the body, in Figure 4. Have the formula 3-1. 4 4 2 2 2 2 2 1 1 2 3 4 1 2 3 4 1 1 ( )i j i j S F F F F F b b = = = = + + + = \u2126 = \u2126 +\u2126 +\u2126 +\u2126\u2211 \u2211 (3-1) Four-rotor aircraft flight attitude analysis. The four-rotor aircraft motion data collected by 3D Suit is restored in the \".Bvh\" file format, so it needs to track the corresponding to each of the basic state of four-rotor aircraft attitude data blocks from the \".Bvh\" file format. Besides using the evaluation work is built upon matlab tool-software to draw the basic state corresponding to each change of attitude angle and attitude angle changes with rotor speed curve" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure7-1.png", + "caption": "Figure 7 6-AXIS HOBBING MAHCINE", + "texts": [ + " (13) N p f sin N p cos( ) f sin 6 \u0153 c c nc c c nc c < # < < pc pc pc pc Now, the instantaneous gear ratio between the desired input gear element and its corresponding output cutter reduces to g . (14) N p d N p d mi c nc c i ni i \u0153 6 6 For f f 0, d d , p p , and the above reduces toi c i c ni nc\u0153 \u0153 6 \u0153 6 \u0153 the ratio of teeth. It may be that constant angular speeds for the gear blank and the cutter are needed. In this case, the feeds are determined for a constant speed ratio g .mi The manufacture of toothed bodies proposed here is based on a 6-axis CNC gear hobbing machine where the desired gear blank is the envelope of the cutter. Illustrated in Figure 7 is a generic 6-axis gear cutting machine. A description of the hyperboloidal cutter and its displacement relative to the gear blank is achieved by emulating the hobbing process for cylindrical gears. The cutter and gear blank in contact are two crossed hyperboloidal bodies in mesh as discussed earlier. The hobbing process includes the specification of the following parameters: A-axis: hob swivel\u00f1 B-axis: hob rotation\u00f1 C-axis: workpiece/gear rotation\u00f1 X-axis: radial in-feed\u00f1 Y-axis: hob shift (axial)\u00f1 Z-axis: gear shift (axial).\u00f1 Determination of the command signals for the 6-axis gear hobbing machine are obtained by establishing a fixed frame of reference (X , Y , Z ). The command signals are determinedf f f by assuming that each element of the machine is a rigid body. Six coordinates are used to specify the joint commands necessary for the fabrication of gears using the gear hobbing machine depicted in Figure 7. These six coordinates are indicated in Figure 8. The six parameters necessary to specify these command signals are listed in Table 1. The position and orientation of the cutter relative to the gear blank is accomplished by a combination of absolute displacements of both the gear blank and the cutter. This combination of absolute displacements is analyzed as two mechanisms operating together. 4 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002378_iet-cta.2014.0736-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002378_iet-cta.2014.0736-Figure2-1.png", + "caption": "Fig. 2 Thruster azimuth and elevation", + "texts": [ + " Although thruster can generate larger control torque than reaction wheel and magnetic actuators, it becomes one type of actuators commonly used in large-angle attitude manoeuvre. In this study, the satellite considered is controlled by using thrusters. A thruster consists of a flow control valve and a combustion chamber. When propellant passes through the combustion chamber, chemical reaction takes place generating thrust through the nozzle. Assume that N thrusters are equipped in the satellite. For the ith thruster, i = 1, 2, . . . , N , its configuration is shown in Fig. 2, the force component can be derived as Fi = Fn [ cos \u03b1i cos \u03b2i cos \u03b1i sin \u03b2i sin \u03b1i ] (7) where Fn \u2208 + is the constant thrust magnitude level, \u03b1i is the elevation angle and \u03b2i is the azimuth angle. Let ri = rxiX B + ryiY B + rziZB be the vector representing the placement of the thruster from the satellite centre of mass. Torque component provided by the ith thruster can be calculated as ui = ri \u00d7 Fi (8) where \u2018\u00d7\u2019 denotes the cross-product of two matrices. Then, the applied control torque \u03c4 generated by N thrusters is \u03c4 = N\u2211 i=1 ui = N\u2211 i=1 ri \u00d7 Fi (9) Whether because of manufacturing tolerances or warping of the satellite structure, some alignment error will exist" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001429_9781118886397.ch14-Figure14.8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001429_9781118886397.ch14-Figure14.8-1.png", + "caption": "Figure 14.8 illustrates the basis on which the angle of the conductor at the support is calculated.", + "texts": [], + "surrounding_texts": [ + "If an initial conductor length L1 and an initial conductor tension T1 are assumed, then for any subsequent motion resulting in L2 and T2, L2 \u2212 L1 = H1 \u2212 H2 aE L (14.27) or \ud835\udeffH = \ud835\udeffL L aE (14.28) where the conductor length, L, can be approximated as L = S + 8D2 3S (14.29) \ud835\udeffL can be expressed as \ud835\udeffL = 8D2 2 3S \u2212 8D2 1 3S = 8 3S (\ud835\udeffD \u2212 2D1) (14.30) where \ud835\udeffD = D1 \u2212 D2 (14.31) The change in tension can be expressed in terms of the change of vertical dis- placement as follows: \ud835\udeffH = 8aE 3SL1 \ud835\udeffD(\ud835\udeffD \u2212 2D1) (14.32)" + ] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.16-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.16-1.png", + "caption": "Fig. 7.16 The two assembly modes of the Orthoglide", + "texts": [ + "49), it can be found that x = c9z + c10 y = c11z + c12 z = \u2212c14 \u00b1 \u221a c214 \u2212 4c13c15 2c13 (7.50) where c9 = \u2212(c2 \u2212 c5)c7/(c1c4 + c1c7 \u2212 c3c7 \u2212 c4c6) c10 = \u2212(c21 + c22 \u2212 c23 \u2212 c24 \u2212 c25)c4/(c1c4 + c1c7 \u2212 c3c7 \u2212 c4c6) c11 = (c2 \u2212 c8)(c1 \u2212 c6)/(c1c4 + c1c7 \u2212 c3c7 \u2212 c4c6) c12 = \u2212(c21 + c22 \u2212 c26 \u2212 c27 \u2212 c28)(c1 \u2212 c3)/(c1c4 + c1c7 \u2212 c3c7 \u2212 c4c6) c13 = c29 + c211 + 1 c14 = 2(c9c10 + c11c12 + c1c9 \u2212 c2) c15 = c210 + c212 + 2c1c10 + c21 + c22 \u2212 d2 4 . The sign \u201c\u00b1\u201d in (7.50) corresponds to the two robot assembly modes (Fig. 7.16). Once thevalues of x , y and z are found, the expressions of thepassive joint coordinates qi2, qi3, qi4 and qi5 can be found using the expressions given in Sect. 7.1.2.4. From a geometric point of view, Eq. (7.49) are three equations of spheres Si centered respectively in O1(0, 0, d6+q11), O2(d6\u2212a+q21, 0, a) and O3(0, d6\u2212 a + q31, a). Thus, the solutions of the FGM are the intersection points of those spheres (Fig. 7.16). The FGM of the Gough-Stewart platform is probably one of the most complicated topics of the field. The ways to solve it will not be detailed here, but the aim of this section is tomake brief recalls on themost relevant works concerning this problem so that the reader can have an idea of what could be interesting w.r.t. his own objectives. In the most general case, 6\u2013UPS PKM can have up to 40 assembly modes. This result was first shown inRonga andVust (1992), and then confirmed through different approaches proposed in Husty (1996), Lazard (1993), Mourrain (1993), Raghavan (1993) and Wampler (1996)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003721_detc2014-34297-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003721_detc2014-34297-Figure5-1.png", + "caption": "FIGURE 5. Kinematic chain generated by the direct sum of two subspaces S1,u\u03021 \u2295S2,u\u03022 .", + "texts": [ + " This section introduces a class of kinematic chain that posseses screw systems of locally constant rank. By subjecting their joint coordinates and their derivatives to additional constraints, the rank of the screw systems associated with this class of chains is reduced and it remains locally reduced. As a subproduct of these conditions, the original partial kinematic chains, that in general, generates only subspaces of a Lie algebra, se(3), generate, in most cases, a different subalgebra of the Lie algebra se(3). An illustrative example of a serially connected kinematic chain is shown in Figure 5. This kinematic chain is generated by the direct sum of two subspaces S1,u\u03021 \u2295S2,u\u03022 . The subspace S1,u\u03021 , is obtained from a serial kinematic chain of the form h1,1 P,u\u03021,1,p1 \u2212h1,2 Q,u\u03021,2,p2 \u2212h1,3 R,u\u03021,3,p3 5 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use where the orientation and relative location of the helical pairs3 is completely general. Under a proper selection of the orientation, location and pitch of the helical pairs, this serial chain can generate any subalgebra of dimension 3, namely, sO, gu\u03021 , tu\u03021,u\u03022,u\u03023 or yu\u03021,p", + " The resulting subalgebra is indicated in the second summands of the fifth column of Table 1. Proposition 5. Any permutation between kinematic pairs that generate the subspace S2,u\u03022 , i.e., between h2,1 M,u\u03022,1,p4 h2,2 N,u\u03022,2,p5 , is permissible. Similarly, this result was proved in [30]. 6 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use S1,u\u03021 \u2295S2,u\u03022 The kinematic chain that is shown in Figure 5, is generated from the direct sum of two subspaces S1,u\u03021\u2295S2,u\u03022 , \u2014which is obtained from the serial chain of the form h1,1 P,u\u03021,1,p1 \u2212h1,2 Q,u\u03021,2,p2 \u2212h1,3 R,u\u03021,3,p3 \u2212 h2,1 M,u\u03022,1,p4 \u2212h2,2 N,u\u03022,2,p5 \u2014, and represents a screw systems of locally constant rank with additional constraints. The results of this section are shown in Table 1. This table shows, if there are, additional conditions to each of the subspaces S1,u\u03021 and S2,u\u03022 , the subspaces resulting after applying the relationships as well as the result of the direct sum of subspaces" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003911_j.fss.2011.01.012-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003911_j.fss.2011.01.012-Figure2-1.png", + "caption": "Fig. 2. Graphical explanation of i .", + "texts": [ + " It will be explained in Section 4 that under some circumstances, this bound is rather conservative. However, the upper bound can be easily acquired, and can help us analyze the stability of the T\u2013S fuzzy system, as described in the Main Theorem 2. 2. We emphasize that the Main Theorem 1 has an interesting graphical explanation. Generally, i means the maximum magnitude of Gi ( j ). In the Nyquist plot of Gi ( j ), it refers to the maximum magnitude Gi ( j ), when varies from zero to infinity. For example, it is the magnitude of the point O1, as depicted in Fig. 2. Thus, an upper bound of the L2-gain of the T\u2013S fuzzy model can be graphically obtained in the frequency domain. On the basis of the main theorem 1, the L2-stability of the feedback system in Fig. 1 can be investigated by the small gain theorem, which is summarized by the following theorem. Main Theorem 2. The feedback system in Fig. 1 is L2-stable with finite gain and zero bias (wb for short), if \u239b \u239d(2r+1)p\u2211 i=1 i \u239e \u23a0 \u239b \u239d2r+1\u2211 j=1 c j \u239e \u23a0 < 1, (16) where c j = sup \u2208R |Gc j ( j )| and i = sup \u2208R |Gi ( j )|" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003259_gt2013-94105-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003259_gt2013-94105-Figure3-1.png", + "caption": "Figure 3: Gear misalignment axes", + "texts": [ + " Failure of gearbox components can result from a variety of factors. These include misalignment, bearing failure, vibration, unexpected loads or overload, poor lubrication, contamination with water or debris, shaft imbalance, adhesion/abrasion, temperature effects, and material or manufacturing defects. Misalignment is a frequent source of gearbox failure. There are two types of gear-to-gear mesh misalignment: parallel misalignment (radial or axial) and angular misalignment (yaw and pitch). The different misalignment types are illustrated in Figure 3. Both types of misalignment lead to increased loading on parts of the gear teeth due to a reduction of contact area of the teeth, leading to brinelling, spalling and bending, and eventually leading to fatigue fracture. Severe misalignment can cause plastic deformation of gears, cracking and shearing of gear teeth, and damage the gear shaft. Misalignment of shafts and gears can occur when the gearbox is assembled in manufacture, which can be caused by thermal gradients across the gearbox and differential thermal expansion of the gears, shafts, housing [8], or can be caused by shaft imbalance leading to large vibrations and deflections of the shafts" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002700_chicc.2014.6896331-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002700_chicc.2014.6896331-Figure4-1.png", + "caption": "Fig. 4: The voltage output mode of over-modulation region", + "texts": [ + " It is able to switch smoothly between the seven segments PWM mode and five segments PWM mode in new linear modulation region. When 0 2 limt t , the seven segments SVPWM mode is implemented; when 0 2 ,lim limt t t five segments SVPWM mode is performed. As some regions are unable to reach directly in the actual over-modulation region 0 limt t , so we present the four optimized over-modulation strategies to convert the voltage vector into five segments boundary and three segments or one segment PWM voltage space as shown in Fig. 4. Let mu be the modified voltage before performing an over-modulation process as the green vector in Fig. 4, and pu be the practical output voltage after performing an over-modulation process as the red vector in Fig. 4. According to the adjacent basic vector time 1 2,t t and zero vector time 0 ,t we calculate vector time 1 02, ,t t t by those aforementioned algorithms. The details of the integrated SVPWM algorithm are shown in Fig. 5. In order to verify the feasibility of integrated SVPWM algorithm, we analyze the performance of the algorithm and build a test system as shown in Fig. 6(a).In the test system, the carrier frequency, dead time, minimum pulse width limitation and the modulation frequency are set as 10KHZ, 3us, 2us and 50Hz, respectively", + "s s st T T t T T t T T The integrated SVPWM implements an processing on the modified voltage vector in each region, and it always with difference between 1 2,t t and 1 2,t t . This changed difference is corresponding to a voltage difference between unregulated and regulated by the integrated SVPMW modulation. Thus, defined 1 2 1 2( ) ( ) 2 3 dc s t t t t U T u Wherein, 1 2, , , dcsT Ut t are known, 1 2, ,t t u are unknown variables, u is the function of F concerning to 1 2 0, , .t t t According to the different voltage output modes in different region as Fig. 4, the voltage difference u is analyzed in details. Marked ,lim db pwlt t t wherein, / ,db db dbsT T Tt is the dead time; / ,pwl pwl pwlsT T Tt is the minimum limitation of narrow pulse width. The concrete analysis is as follows: 1) Provided that 0 2 ,limt t directly output ,mu namely, ,p mu u at this moment, 1 1 2 2, ,t t t t thus 0u . 2) Provided that 0 2lim limt t t , directly output mu , at this moment, 1 1 2 2, ,t t t t thus 0u . 3) Provided that 02lim limt t t , the modified voltage mu is transformed into the practical voltage pu by an equal proportion compression method according to Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001136_978-3-319-19788-3_1-Figure1.20-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001136_978-3-319-19788-3_1-Figure1.20-1.png", + "caption": "Fig. 1.20 Examples of redundant robots. a A kinematically redundant PPM(3-(P) RRR robot) from the institute of mechatronics systems, Germany (Kotlarski et al. 2010). b The DualV from the LIRMM, France (van der Wijk et al. 2011): an actuation redundant PPM(4-RRR robot)", + "texts": [ + " 2008) is obtainedwhen the total number Ndof of DOF for the robot exceeds the number ndof of independent variables necessary to define the robot\u2019s platform configuration. In such a case, we have na = Ndof > ndof , na being the number of actuators. It results in an infinitude of possible solutions to the inverse kinematic problem giving the joint coordinates of the robot in terms of the platform coordinates (see Sect. 7.3.2.4). This type of redundancy occurs when extra active joints and links are added to a manipulator (Fig. 1.20a). Advantages can include larger reachable workspace, avoidance of kinematic singularities, and dexterity improvement (Ebrahimi et al. 2008). Actuation redundancy occurs when the number na of actuators is greater than the number of robot DOF Ndof . Mathematically speaking, we have na > Ndof . As a consequence, \u2022 we cannot independently choose the active joint variables as they are constrained by nc equations. \u2022 there are an infinite number of possible solutions to the inverse dynamic problem (see Sect. 8.5). Internal constraint efforts may appear. \u2022 the wrench capabilities are affected (Firmani et al. 2007) and forces of greater magnitudes can be generated. \u2022 as an advantage, the robot workspace becomes usually free of singularity. An example of an actuation redundant robot named the DualV is provided in Fig. 1.20b. 1.3.4 Other Types of PKM 1.3.4.1 Hybrid PKM The hybrid robots are composed of serially connected parallel modules like the LogabexLX4 robot (Fig. 1.21) (Charentus andRenaud 1989) and bio-mimetic snakes robots (Chablat and Wenger 2005; Khalil et al. 2007a). The serial form of these hybrid manipulators overcomes the limited workspace of parallel manipulators and improves overall stiffness and response characteristics. Cable-driven parallel robots are quite recent types of PKM in which the rigid links are replaced by cables" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003538_amm.325-326.870-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003538_amm.325-326.870-Figure1-1.png", + "caption": "Fig. 1. The overturned gripping device, [4]", + "texts": [ + " The profile verification of the bevel gears was introduced only recently. In the case of the bevel gears the measuring machine of spur gears cannot be used due to their conical shape. Tredgold\u2019s approach [2, 3] for the geometrical calculation can be used at the profile control too. Basing on this the profile measurement is carried out in a plane which is perpendicular to the cone generator. So bevel gears can be measured on any device, if they can be turned over with half angle of the pitch cone to the machine\u2019s table, (Figure 1). Such a gripping device is used for profile measuring using a universal microscope. There are other measuring instruments for profile verification which works by contact. Such an instrument is presented in [3] which is based on the idea of the uncontrollable base disc. This disc is effectively the rotating disc which rolls on the ruler of the measuring instrument without sliding, while the touching probe detects the profile. The coordinate measuring instruments are functioning based on this idea" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001133_978-1-4471-5102-9_181-1-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001133_978-1-4471-5102-9_181-1-Figure3-1.png", + "caption": "Fig. 3 VGTV X-treme (Courtesy of Recce Robotics) http://www.recce-robotics.com/vgtv.html", + "texts": [ + " Unmanned construction system consists of teleoperated robot backhoes, trucks, and bulldozers with wireless relaying cars and camera vehicles as shown in Fig. 2 and is remotely controlled from an operator vehicle. It has been used since the 1990s for remote civil engineering works from a distance of a few kilometers. Page 2 of 7 Small-sized UGVs (unmanned ground vehicles) were developed for victim search and monitoring in confined space of collapsed buildings and underground structures. VGTV X-treme shown in Fig. 3 is a tracked vehicle remotely operated via a tether. It was used for victim search at mine accidents and the 9/11 terror. Active scope camera shown in Fig. 4 is a serpentine robot like a fiberscope and was used for forensic investigation of structural collapse accidents. Large-scale fires of chemical plants and forests sometimes have a high risk, and firefighters cannot approach. Remote-controlled robots with firefighting nozzles for water and chemical extinguishing agents are deployed. Large-sized robots can discharge large volume of the fluid with water cannon, and small-sized ones have better mobility" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002525_j.phpro.2013.03.119-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002525_j.phpro.2013.03.119-Figure2-1.png", + "caption": "Fig. 2. A gas-filled target - as used today.", + "texts": [ + " This paper gives the example of a study applied to gas-filled targets. One distinctive feature of such targets is that they include a thin metallic filling tube (with an outer diameter of about 0.3 mm). This tube has to be implemented with high level requirements: the tube / hohlraum junction must be gastight (with a leakage flow rate better than 3 x 10-5 mbar.l.s-1 - measured under a helium pressure of 1 bar), this tube has to be oriented into a narrow virtual cone (under an acceptance angle of \u00b1 2 degrees). Figure 2 shows targets as they are manufactured today (i.e. with a glued filling tube). Hohlraums are made of two distinct materials: In order to simplify the overall manufacturing process, we are developing a potential version excluding this resin. In that goal we have proposed an adapted design for the hohlraum (given figure 3), with a gold thickness increased up to 100 \u03bcm. This thickness allows to drill a counterbored hole, in order to insert the A laser welding solution has been studied to bond the filling tube onto the gold hohlraum and to fulfil specifications quoted above (gas tightness and spatial orientation)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003324_iccas.2013.6704069-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003324_iccas.2013.6704069-Figure4-1.png", + "caption": "Fig. 4 Controlled system by using 4 tugboats.", + "texts": [ + " Although the actuator is constrained about the power supply to the system, the errors between actual and commanded forces are almost eliminated. A dynamic positioning of barge ship by using assistant of 4 tugboats is studied. To counteract the effect of environmental disturbances, 4 tugboats produce thrusts simultaneously. The motion of the barge is restricted in an area as small as possible. Because the tugboat cannot immediately change thrust from pushing to pulling and inversely, we will limit the tugboat thrust as limited pushing force to the vessel. The total system is shown in Figure. 4 and the system dynamic can be presented by the following model it = R( rp )v, Mv+ Dv = Tc +RT (rp)b, (22) where 1J = [x, y, rp t E 1Ft 3 represents the position and J' 3 heading angle. v = [u, v, r] E 1Ft describes surge, sway and yaw rate of the ship. M, D is the inertia and damping matrix and b is the slow frequency environmental disturbance. as In this paper, the directions of tugboats are kept fixed 3J[ J[ J[ 3J[ u] = -- , u2 =--,u, =-, u4 =- . 4 4 \u00b7 4 4 (23) The control allocation is defined in term of the forces produced by each tugboat thrust u = (up " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001126_2015-01-2515-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001126_2015-01-2515-Figure7-1.png", + "caption": "Figure 7. Another vertical inject injector cutaway view", + "texts": [], + "surrounding_texts": [ + "As discussed in the previous section, when the parallel gripper is actuated closed, the guide blocks form a space for the fastener to seat into, shown clearly in Figure 3. If there is some problem with the injection, such as the wrong fastener being called, a fastener being fed backwards, or some other malfunction, the injector can use its purge function to clear the feed tube and guide chute of jams. Figure 6 shows an injector tool in purge position. The parallel gripper has been opened, thereby opening the guide blocks and creating a clear path for purge. Some consideration must be made to prevent the purged fasteners from rocketing into the aircraft panel or otherwise finding their way into places they should not be. The injector purges with the injector positioned in front of the headstone, so purged fasteners contact the headstone before falling harmlessly to the floor. With good injector reliability, purge functions will only be used in extremely rare cases." + ] + }, + { + "image_filename": "designv11_84_0002301_s0140525x13000617-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002301_s0140525x13000617-Figure6-1.png", + "caption": "Figure 6. Behavioural studies of the use of slope cues in navigation, in rats (a and b) and in humans (c). (a) Schematic of the tilted-arm radial maze of Grob\u00e9ty & Schenk (1992b). (b) Schematic of the conical-hill arena of Moghaddam et al. (1996). (c) The virtual-reality sloping-town study of Restat et al. (2004) and Steck et al. (2003), taken from Steck et al. (2003) with permission.", + "texts": [ + " Durgin et al. (2011) have challenged this view, suggesting that experimental probes of visuomotor perception are themselves inaccurate. The question of whether there are two parallel systems for slope perception therefore remains open. How does slope perception affect navigation? One mechanism appears to be by potentiating spatial learning. For example, tilting some of the arms in radial arm mazes improves performance in working memory tasks in rats (Brown & Lesniak-Karpiak 1993; Grob\u00e9ty & Schenk 1992b; Figure 6a). Similarly, experiments in rodents involving conical hills emerging from a flat arena that rats navigated in darkness found that the presence of the slopes facilitated learning and accuracy during navigation (Moghaddam et al. 1996; Fig. 6B). In these experiments, steeper slopes enhanced navigation to a greater extent than did shallower slopes. Pigeons walking in a tilted arena were able to use the 20-degree slope to locate a goal corner; furthermore, they preferred to use slope information when slope and geometry cues were placed in conflict (Nardi & Bingman 2009b; Nardi et al. 2010). In these examples, it is likely that slope provided compass-like directional information, as well as acting as a local landmark with which to distinguish between locations in an environment. BEHAVIORAL AND BRAIN SCIENCES (2013) 36:5 529 In humans, similar results were reported in a virtual reality task in which participants using a bicycle simulator navigated within a virtual town with eight locations (Restat et al. 2004; Steck et al. 2003; Fig. 6C). Participants experiencing the sloped version of the virtual town (4-degree slant) made fewer pointing and navigation errors than did those experiencing the flat version. This indicates that humans can also use slope to orient with greater accuracy \u2013 although, interestingly, this appears to be more true for males than for females (Nardi et al. 2011). When slope and geometric cues were placed in conflict, geometric information was evaluated as more relevant (Kelly 2011). This is the opposite of the finding reported in pigeons discussed above" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002494_amr.1028.105-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002494_amr.1028.105-Figure11-1.png", + "caption": "Fig. 11 Vibration mode of order 6", + "texts": [], + "surrounding_texts": [ + "This paper is based on Solidworks to establish the 3 dimensional solid model of the shaft and uses ANSYS Workbench to analyze pre-stressed model of the shaft. Pre-stressed modal analysis is closer to reality. According to the results of the analysis, the reasonable frequency of the stepper motor can be presented, thus avoiding the generation of resonance. Meantime, this method provides reference for the selection of the stepper motor." + ] + }, + { + "image_filename": "designv11_84_0003698_oceans.2014.7003047-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003698_oceans.2014.7003047-Figure1-1.png", + "caption": "Fig. 1. The normal location relation of the leader and a follower", + "texts": [ + " Based on the basic L-F formation control method, some followers\u2019 states of the multi-AUVs system are redefined and a finite state automaton (FSA) using the state feedback of the followers is built, and then the system the corresponding control strategy will be described. A computer simulation considering the different number of system followers is carried out, and then the method on an AUV semi-physical simulation platform based on MOOS is tested. The results show that FSA based on state feedback control can effectively deal with formation of abnormal conditions. II. STATE ANALYSIS OF FOLLOWERS In this research, only the cases of abnormal followers are considered. In a moment, the normal location relation of the leader and a follower is shown in Figure 1. In the geodetic coordinate system, the real-time position and attitude of leader are defined as ( ) ( ), , , , ,L L L L L L L LP x y z A \u03d5 \u03b8 \u03c8= = , According to the L-F method, \u03d5 and \u03b9 are known and then the desired position and attitude of the follower could be obtained: ( ) ( )1 1 1 1 1 1 1 1, , , , ,T T T T T T T TP x y z A \u03d5 \u03b8 \u03c8= = , Supported by Basic Research (Grant number: B1320133015), the National High Technology Research and Development Program (Grant number: 2011AA09A105). 978-1-4799-4918-2/14/$31" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002240_sii.2011.6147563-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002240_sii.2011.6147563-Figure5-1.png", + "caption": "Fig. 5. Typical 3-D motion of microrobot", + "texts": [ + " In addition, a view around the levitated object visualized in the same way is shown in Fig. 4 (2). In order to show the motion of the levitated object in an easy-to-understand way, the Z axis of the coordinate system described in section II is depicted as a red line, and an initial stable region for the microrobot is colored in light pink. Additionally an SI Intemational 2011 appearance of initial state and the appearances of typical 3- D motions (x-axis motion, y-axis motion and z-axis motion) are shown in Fig. 5 (1) '\" (4), respectively. E. Network system This simulator can be connected to network using socket communication, it aims at having the functions to confirm the manipulation performance by connecting directly to actual robot controller, remote operation device, and so on. Indeed, the proto typing of a network between the simulator and a remote manipulation device is demonstrated in section VI. In this stage, the accuracy of temporal synchronization of the simulator is not so good, because it is built on a general non-realtime linux OS" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001813_mfi.2015.7295819-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001813_mfi.2015.7295819-Figure8-1.png", + "caption": "Fig. 8. High-speed robotic system and button spinner", + "texts": [ + " Considering a sine wave as a input trajectory of the finger tip, the parameters such as the amplitude, stroke, frequency and the center of oscillation have a crucial affect on playing the button spinner. Therefore, here we extracted the parameters obtained from successful cases and applied those parameters to the robotic manipulation of the button spinner. Our button spinner system used in the experiments mainly consists of two parts: a robotic hand with real time controller and a high-speed camera system, as shown in Fig. 8. The two parts operates at 1 kHz respectively and synchronized by 100 Mbps Ethernet and to form a high-speed visual feedback loop. In the experiment with feedforward control, the camera system is used only for recording the rotation of rotor such as position and angular position, not for the feedback control. Since we do not have the accurate model for playing button spinner, we adjusted the parameters of input sine wave based on the human finger motion (Fig. 7) and confirmed the stable condition for playing button spinner" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002136_s1068798x14070090-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002136_s1068798x14070090-Figure3-1.png", + "caption": "Fig. 3. Experimental apparatus.", + "texts": [ + " With sufficient flange thickness (hmin = 16 mm for steel and hmin = 22 mm for cast iron), \u03d5 hardly depends on h, according to [2] (Fig. 2). This indicates that the contact surface is not buckled or distorted. Thus, when h \u2265 hmin, the flanges have sufficient flexural rigidity. In the present case, with a screw diameter d = 8 mm, hmin/d = 2 for steel and hmin/d = 2.75 for cast iron. Correspondingly, lscr/hmin \u2248 3.6 and 2.6. A condition for the absence of flexural strain was proposed in [3]: 2h/(D \u2013 dro) \u2265 1. The experimental apparatus developed for study of a complex screw joint 1 (Fig. 3) consists of two coils with flanges of height h = 25 mm. The external diam eter of the flange D1 = 100 mm; the internal diameter \u03d5 f fa\u2013( )/l0 Fr 1/c 1/ca\u2013( )l0.= = Keywords: complex screw joint, flange, flexural rigidity, butt joint, contact pliability, force, screw DOI: 10.3103/S1068798X14070090 430 RUSSIAN ENGINEERING RESEARCH Vol. 34 No. 7 2014 MURKIN et al. of the flange\u2019s supporting surface D2 = 45 mm; the coils\u2019 rod diameter dro = 47 mm; the diameter of the circle on which the screws lie is Dscr = 75 mm; and the diameter of the hole for the screw is d0 = 11 mm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003854_codit.2013.6689640-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003854_codit.2013.6689640-Figure1-1.png", + "caption": "Figure 1. Mechanical unbalanced Motor", + "texts": [ + " THEORETICAL DEVELOPMENT In the following, all machines parameters are referred to stator, \" \"p denotes number of the induction machine pole pairs and \" \"s\u03c9 the supply voltage frequency. A. Vibration A rotating machine of mass \" \"m can be modeled as being mounted to a fixed support, on a spring of stiffness \" \"K with a damping viscous coefficient \" \"C . The unbalance can be 978-1-4673-5549-0/13/$31.00 \u00a92013 IEEE 770 represented by a mass \" \"\u03bc rigid at a distance \" \"r from the centre of rotation (Fig. 1). The considered fault is supposed to begin at time \" 0\"t = . When the motor rotates at speed\" \"r\u03c9 , if we assume that the machine is constrained to move vertically (single degree-offreedom), the equation of motion becomes: 2 2 2. . . . . sin( )r r d y dym C K y r t dtdt \u03bc \u03c9 \u03c9+ + = (1) At steady state, vertical displacement due to mechanical mass unbalance will be given by: 0( ) .sin( )ry t Y t\u03c9 \u03d5= \u2212 (2) Where: 1/ 22 2 2 2 0 ( ) ( )r r rY r K m C\u03bc \u03c9 \u03c9 \u03c9 \u2212 \u23a1 \u23a4= \u2212 +\u23a3 \u23a6 (3) And: 2tan( ) /( )r rC K m\u03d5 \u03c9 \u03c9= \u2212 (4) From (3), it is clear that mechanical unbalance leads to a small displacement (vibration) that can be monitored by a suitable transducer (accelerometer) delivering a signal oscillating at rotational frequency" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002499_ssp.199.338-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002499_ssp.199.338-Figure1-1.png", + "caption": "Fig. 1. Schematic illustration of the surface profile obtained by incremental method: 1 - reference lines, 2 - central line, rp - radius of the rounded fillet, rv - radius of the rounded corner, \u03b8 - angle of inclination of the model surface, A'p - surface area of the peak, A'v - surface area valleys , \u03c6 - angle of inclination layer profile, a - thickness of the layer", + "texts": [ + " The setting of model on the platform should cause the minimum surface roughness in the critical model areas. To optimize this process surface roughness models of components made by Stereolithography are created. This article introduces the various types of surface roughness models which are described in the literature and compares the consistency of mapping to the real surface structure. In the literature [1,2,3], models of roughness profile are mostly based on the model of \"sharp edges\u201d of a rectangular profile. This kind of models describes formula 1. The surface profile is obtained when (Fig. 1), the radius rp = rv = 0 and the angle \u03c6 = 0 are assumed. It takes into account only two parameters of the manufacturing process i.e. layer thickness (a) and inclination of the model surface in relation to the incremental layers (\u03b8) [1, 2] )1800(cos 4 )( 00 <<= \u03b8\u03b8\u03b8 a Ra (1) where: Ra \u2013 the roughness parameter, a - thickness of the incremental layer, \u03b8 - angle of inclination of the surface. This model is also used in computer programs that model the surface irregularities on the elements performed by additive methods [1]. Models obtained using the technologies are characterized by a specific surface structure of the so-called \u201cstairstep\u201d effect. General scheme in Fig.1 presents such a profile. A more complex model is the one that incorporates the additional layer of profile inclination angle (\u03c6). All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 128.210.126.199, Purdue University Libraries, West Lafayette, USA-24/05/15,11:46:33) The model design was created as one of the first, and described in the literature [3,4]. It is given by formulas as follows: ( ) ( ) ( ) ( ) 4 tansincos , \u03c6\u03b8\u03b8 \u03c6\u03b8 + = a Ra (2) where: Ra - roughness parameter, a - constant layer thickness, \u03b8 - angle of part inclining surface, \u03c6 - angle of the layer profile" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003854_codit.2013.6689640-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003854_codit.2013.6689640-Figure3-1.png", + "caption": "Figure 3. Mechanically Unbalanced machine", + "texts": [ + " The stator of the induction machine consists of 36 slots and the rotor is a cast aluminum type with 28 bars which is rigidly fixed to a fan having 7 blades. The load has been simulated by a magnetic brake which is stiffly coupled to the induction machine shaft and, via its control unit, allows user to apply different load torque levels and acquire speed and torque signals. Mechanical unbalance fault was simulated by fixing a blot of mass \u201c \u03bc =20g\u201d, on one fan blade at a distance \u201c r = 4 cm\u201d from the rotating axe center as it is shown in Fig. 3. Radial vibration signal was acquired by an accelerometer LUTRON TRVBT 1A4 mounted on the machine body. Stator current sensor was ensured by a clamp C100 CHAUVIN ARNOUX, and stray flux was explored thanks to two punt coils MATELCO, each of 1000 turns and diameter of 12cm. these coils was placed near the machine body in two different positions: - Axial Position: coil axis coincides with the machine rotor axis. - Transversal Position: coil axis is perpendicular to the machine rotor axis. Low-pass anti-aliasing filter is implemented in order to set the frequency bandwidth of the analyzed signals to a correct range (Butterworth of order 5 with cut-off frequency 1 KHz)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001642_coase.2015.7294190-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001642_coase.2015.7294190-Figure7-1.png", + "caption": "Figure 7. The elliptical motion and contact angle", + "texts": [ + " According to (1), the linear part of the PTZ system can be regarded as a second-order dynamic system, with inputting U and rotation angle, and the transfer function is: 2 2 2 2 2 = 2 2 1 n n n K K G s s s T s Ts . (2) While T is the output torque of the PTZ system, which is mainly influenced by the choice of pre-stress CF . The elliptic motion at the top of the stator is described as follows: cos sin t m x x t z z t \uff083\uff09 Where tx and mz represent respectively the amplitude of the elliptical vibration along different directions. As illustrated in Fig.7, the contact between stator and base begins with a, then end up with b in a period of oscillation. Where = b a represents the contact angle, mv and maxv stand for the steady rotation speed and the maximum vibration velocity respectively. The pre-stress CF has a great influence on the contact angle, which can be demonstrated by formula as follow: 2 = 2sin / 2 cos / 2C e mF k z (4) The contact pressure can be obtained: ( ) ( , ) = 0 ( ) e a a bk z z t t N otherwise (5) The friction force rf is the main driving force of PTZ rotation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001835_pamm.201310079-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001835_pamm.201310079-Figure1-1.png", + "caption": "Fig. 1: The deformed configurations of analytical, FD, CSDA scheme for Cook-type problem and comparison of CPU time for each scheme.", + "texts": [ + " To assess the performance of the perturbation techniques for the numerical computation of the spatial tangent moduli tensors, we analyze a Cook-type cantilever beam, where a quadratic distribution of in-plane shear load is applied to the right side of the beam, and clamped on the other side. We choose the set of material parameters as \u03b11 = 6.0, \u03b21 = 100.0 and \u03b22 = 2.5, \u01eb1 = 100.0, \u01eb2 = 5.0 with \u03b11, \u01eb1 and \u03b21 in unit of stress. The preferred direction is assigned as a = 1/ \u221a 3 ( 1 1 1 )T . The shear load p was successively increased in 5000 equal steps until the maximum shear load p = 5.0. Figure 1 (a-c) show the deformed configurations with the resulting stress distribution \u03c3xx. First, we have investigated the sensitivity of the rate of convergence with respect to a change in the perturbation value h as reported in the Table below. Next, we have compared CPU times (FD with h = 10\u22129 and CSDA with h = 10\u221230) as shown in Figure 1 (d). We conclude that the resulting tangent moduli with FD scheme are not accurate enough and the Newton-Raphson method loses its characteristic quadratic convergence. In contrast to that, the one with CSDA scheme shows robust and high accurate performance with sufficiently small value of h. The only drawback of the present formulatin with CSDA is higher computational load because it uses the complex numbers in computation. However, the numerical examples show that it is not expensive for finer meshes since the computing time associated with assembling becomes less significant than the time for solving the system of equations" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.4-1.png", + "caption": "Fig. 7.4 The two working modes of the leg i of the planar five-bar mechanism", + "texts": [ + "11), we get: hp(x, qa) = [ (c1 \u2212 a1)t211 + 2b1t11 + (a1 + c1) (c2 \u2212 a2)t221 + 2b2t21 + (a2 + c2) ] = 0 (7.13) from which we can find that ti1 = \u2212bi \u00b1 \u221a b2i \u2212 c2i + a2 i ci \u2212 ai (7.14) or also qi1 = 2 tan\u22121 \u239b \u239d \u2212bi \u00b1 \u221a b2i \u2212 c2i + a2 i ci \u2212 ai \u239e \u23a0 . (7.15) Note that: \u2022 the value of q11 can be found from (7.7) for i = 1 without considering the Eq. (7.7) for i = 2. \u2022 the value of q21 can be found from (7.7) for i = 2 without considering the Eq. (7.7) for i = 1. This means that Eq. (7.7) can be solved independently. In (7.15), the sign \u201c\u00b1\u201d correspond to the different working modes of the robot (Fig. 7.4). From a geometric point of view, solving these equations is equivalent to finding the intersection points of two circles (Fig. 7.4): \u2022 Circle Ci1 centred in Ai1 of radius di2, which corresponds to the vertex space of the point Ai2 when considering that it belongs to the link Ai1Ai2, \u2022 Circle Ci2 centred in Ai3 (considered as fixed if the coordinates x and y are known) of radius di3, which corresponds to the vertex space of the point Ai2 when considering that it belongs to the link Ai2Ai3. The values of q12 and q22 can be obtained using (7.9) as qi2 = atan2 (y \u2212 di2 sin qi1, x \u2212 di1 \u2212 di2 cos qi1) \u2212 qi1, for i = 1, 2 (7.16) where \u201catan2\u201d is the four-quadrant inverse tangent function" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003659_20131120-3-fr-4045.00042-Figure15-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003659_20131120-3-fr-4045.00042-Figure15-1.png", + "caption": "Fig. 15. Experimental platform.", + "texts": [ + " All the quadrotors reach the desired altitude in approximately 8 seconds (see Fig. 11) which remain constant during the whole simulation. As expected, the control input \u03c4\u03c8 = 0 Nm since the angle \u03c6 does not vary during the whole path (see Fig. 12). The total thrust is shown in figure 13 and the evolution of the quadrotors in the X-Y plane is shown in figure 14. In order to initially evaluate the position and orientation control of the scheme proposed in this work, an experimental platform was used (see Fig. 15) consisting of a quadrotor which has an embedded digital signal processor (DSP), an inertial measurement unit (IMU), four actuators (motors) with their respective power driver and a wireless communication system to accomplish autonomous flight. In order to measure the quadrotor position, a motion capture system integrated by twelve infrared cameras connected to a PC via a USB port is employed. The capture system has a set of markers that reflect the infrared light emitted by the cameras. This system provides the quadrotor\u2019s position at 100 Hz with a submillimetric accuracy" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001963_ijcnn.2014.6889627-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001963_ijcnn.2014.6889627-Figure4-1.png", + "caption": "Fig. 4. The 2-dof parallel robot. IV. SIMULATION RESULTS", + "texts": [ + " If no such a positive definite symmetric matrix P\ud835\udc56 exists, then choose new Q\ud835\udc56 and solve (49). Step 5: Obtain the controller (47) from u\ud835\udc52\ud835\udc5e\ud835\udc56 in (16), u\ud835\udc5f\ud835\udc5b\ud835\udc56 in (24), u\ud835\udc60\ud835\udc64\ud835\udc56 in (36) or in (45), u\ud835\udc52\ud835\udc60\ud835\udc61\ud835\udc56 in (38), and u\u210e\ud835\udc56 in (48) and compute the adaptive laws (40)-(43) to adjust the parameter vectors w\u0302\ud835\udc50\ud835\udc56 , h\u0302\ud835\udc56, ?\u0302?\ud835\udc56, z\u0302\ud835\udc56, and the gain vectors b\u0302 \ud835\udc56 , respectively. Repeat this Step and apply the controller as given by (47) to control the nonlinear interconnected systems. In this section, a 2-dof parallel robot as shown in Fig. 4 is used to illustrate the performance and efficiency of the proposed \ud835\udc3b\u221e adaptive integral sliding recurrent neural control scheme. From our previous work [16] and let \ud835\udefd = 3\ud835\udc50\ud835\udc5c\ud835\udc60(\ud835\udc61), we have{ ?\u0307?1\ud835\udc56 =\ud835\udc992\ud835\udc56 ?\u0307?2\ud835\udc56 = f\ud835\udc56+\u0394f\ud835\udc56+H\ud835\udc56u\ud835\udc56+ \u2211\u2113 \ud835\udc57=1,\ud835\udc57 \u2215=\ud835\udc56 F\ud835\udc56\ud835\udc57(\ud835\udc991\ud835\udc56 ,\ud835\udc991\ud835\udc57 )\ud835\udc991\ud835\udc57 (\ud835\udc61)+d\ud835\udc56 where f\ud835\udc56 = \u2212C\ud835\udc56\ud835\udc992\ud835\udc56 \u2212 g\ud835\udc56, H\ud835\udc56 = B\ud835\udc56, F\ud835\udc56\ud835\udc57(\ud835\udc991\ud835\udc56 ,\ud835\udc991\ud835\udc57 )\ud835\udc991\ud835\udc57 (\ud835\udc61) = A\ud835\udc56\ud835\udc57(\ud835\udc991\ud835\udc56 ,\ud835\udc991\ud835\udc57 )\ud835\udc991\ud835\udc57 (\ud835\udc61). From (3), we have{ ?\u0307?\ud835\udc5a1\ud835\udc56 = \ud835\udc99\ud835\udc5a2\ud835\udc56 ?\u0307?\ud835\udc5a2\ud835\udc56 = \u2212R0\ud835\udc56 \ud835\udc99\ud835\udc5a1\ud835\udc56 \u2212R1\ud835\udc56 \ud835\udc99\ud835\udc5a2\ud835\udc56 + r\ud835\udc5a\ud835\udc56 where R0\ud835\udc56 =\ud835\udc51\ud835\udc56\ud835\udc4e\ud835\udc54 { 0.0405, 0.0096, 0.0252, 0.0027 } and R1\ud835\udc56 =\ud835\udc51\ud835\udc56\ud835\udc4e\ud835\udc54 { 0.0450, 0.0728, 0.0036, 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002267_2041304110394537-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002267_2041304110394537-Figure3-1.png", + "caption": "Fig. 3 The pole placement region of S(a, r, u)", + "texts": [ + " In order to achieve this aim a suitable closed-loop pole constraint can be imposed to specify closed-loop damping and avoid fast dynamics of the controller. The region of interest for the considered control purposes is the set S(a, r, u) of complex numbers x + j y such that x\\ a\\0 , x + j yj j\\r , tan u x\\ yj j 1008 P-C Chen, S-L Wu, and K-Y Chang Proc. IMechE Vol. 225 Part I: J. Systems and Control Engineering at PENNSYLVANIA STATE UNIV on May 24, 2015pii.sagepub.comDownloaded from as shown in Fig. 3. The conditions of the poles of the closed-loop system matrix, Acl, lying in the region S(a, r, u), are characterized by the following matrix inequalities [29] 2aP + AT clP + PAcl\\0 , r P AT clP PAcl r P ! \\0 sin u(AT clP + PAcl) cos u(AT clP PAcl) cos u(PAcl AT clP) sin u(AT clP + PAcl) ! \\0 (17) By performing the congruence transformation via the matrices pT 1 , diag(pT 1 , pT 1 ), diag(pT 1 , pT 1 ) and p1, diag(p1, p1), diag(p1, p1), respectively, for the three matrix inequalities in equation (17), the following LMI conditions can be obtained Ja = 2aJp + Ja + JT a\\0 , Jr = rJp JT a Ja r Jp " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.18-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.18-1.png", + "caption": "Fig. 7.18 Example of a parallel robot (here, a five-bar mechanism) in a Type 1 singularity", + "texts": [ + " The first kind of singular configurations we will analyze are those that we call inputoutput singularities. They can be defined through analysis of the input-output kinematic relationship described in (7.62), from which three main types of singularity can be defined (Gosselin and Angeles 1990): \u2022 when matrix B is rank-deficient: such kind of singularity is called a Type 1 singularity.3 In such singularities, the PKM loses the ability to move along one (or more) direction of the workspace, i.e. a motion of the actuators does not lead to the displacement of the robot platform (Fig. 7.18). 3They are also called serial singularities in some works because it is similar to the singularities of serial robots. However, in the present book, a serial singularity has another meaning. Fig. 7.19 Example of parallel robot (here, a five-bar mechanism) in a Type 2 singularity \u2022 when matrix Ar (as well as matrices A and Ad ) is rank-deficient: such kind of singularity is called Type 2 singularity.4 In such singularities, the PKM gains one (or more) uncontrollable motion, i.e. it becomes shaky", + " As the columns of the matrix 0Ji mi are unit screws which can be seen as a Pl\u00fccker representation of lines, the methods proposed in Sect. 7.5.4 can be applied to find serial singularities. Let us consider again the five-bar mechanism presented in Sect. 7.1.2.1. From the analysis of its input-output kinematic relations defined in Sect. 7.3.4.1, we can see that: \u2022 Type 1 singularities (when matrix B defined at (7.104) is rank-deficient) appear when qi2 = 0 or \u03c0 , i.e. when the leg is full stretched or folded. An example of such type of singularity for the five-bar mechanism was provided in Fig. 7.18. \u2022 Type 2 singularities (when matrix Ar defined at (7.113) is rank-deficient) appear when q11 + q12 \u2212 q21 \u2212 q22 = 0 or \u03c0 , i.e. when the points A12, A13 and A22 are aligned. An example of such type of singularity for the five-bar mechanism was provided in Fig. 7.19. It can be also shown that, when q11 + q12 \u2212 q21 \u2212 q22 = 0 or \u03c0 , there exists a non zero vector ts = [\u2212 sin(q11 + q12) cos(q11 + q12)]T such that: Ats = 0. ts is orthogonal to the direction defined by the line passing through the points A12, A13 and A22 and represents the direction of the uncontrollable motion inside the singularity (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003856_00368791111154995-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003856_00368791111154995-Figure1-1.png", + "caption": "Figure 1 Hole-entry hybrid journal bearing geometry and coordinate system", + "texts": [ + " The bearing performance characteristics have been presented minimum fluid-film thickness ( hmin), fluid-film stiffness and damping coefficients ( Sxx, Szz, Cxx, Czz) and stability parameters ( Mc and vth) for a wide range of values of non-linearity factor ( K) for representative values of the bearing geometric and operating parameters as shown in Table I. The results presented in this paper are expected to be quite useful to the bearing designers. The geometry and coordinate system of hole-entry hybrid journal bearing system is shown in Figure 1. The generalized Reynolds equation as in Dowson (1962), governing the laminar flow of incompressible lubricant between the clearance space of journal and bearing considering variable viscosity and usual assumptions in the non-dimensional form is written as: \u203a \u203aa h3 F2 \u203a p \u203aa \u00fe \u203a \u203ab h3 F2 \u203a p \u203ab \u00bc V \u203a \u203aa 1 2 F1 F0 h \u00fe \u203a h \u203a t \u00f01\u00de Where F0, F1, and F2 are the cross-film viscosity integrals and given by the following relations: F0 \u00bc Z 1 0 1 m d z; F1 \u00bc Z 1 0 z m d z; F2 \u00bc Z 1 0 z m z2 F1 F0 d z Using Galerkin\u2019s orthogonality conditions and following the usual assembly procedure, global system equation is derived Sinhasan and Chandrawat (1989) as: \u00bd F { p} \u00bc { Q} \u00feV{ RH} \u00feX _ j{R s} \u00fe Z _{R s} \u00f02\u00de The flow of lubricant through constant flow valve restrictor in non-dimensional form is given by Sinhasan et al" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001567_icma.2015.7237783-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001567_icma.2015.7237783-Figure3-1.png", + "caption": "Fig. 3 Eyeball mechanism", + "texts": [ + " The elastic steel-wire crossed a nylon tube which is fixed on head skeleton. By pivot pins, one end of the wire is connected to the motor rocker and another end is connected to the eyebrow towing point. The steel-wire is 2 mm thick to make it plastic. When the motor rotates, the steel wire will make deformation to move through the nylon tube. Eyebrow mechanism is showed in Fig. 2 1944978-1-4799-7098-8/15/$31.00 \u00a92015 IEEE Proceedings of 2015 IEEE International Conference on Mechatronics and Automation August 2 - 5, Beijing, China Fig. 3 is CAD drawing of eye mechanism. Eyeball mechanism has 4 DOFs to pitch and yaw eyeballs. Each motor and eyeball are connected by RSSR mechanism, which is a space linkage mechanism. The two eyeballs have separate mechanism, so they can make independent movement. The eyeball is 30mm diameter. They are manufactured by 3D printer. Each eyeball is equipped with a CCD camera of 13.5mm*13.5mm size to get visual information. Eyelid mechanism is showed in Fig. 4. The mechanism, including upper eyelid and lower eyelid, has 4 DOFs to open and close eyelids" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003274_amm.325-326.375-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003274_amm.325-326.375-Figure1-1.png", + "caption": "Fig. 1 20kW PM motor and its models", + "texts": [ + " The Thermal Field Mathematical Model. Conduction, convection heat transfer modes are used in the 20kW water cooled motor. Conduction. x T kAq \u2202 \u2202 \u2212= (4) Where, q is the heat transfer rate, k is the thermal conductivity of the material and xT \u2202\u2202 / is the temperature gradient in the direction of the heat flow. )( wconv \u221e\u2212= TThAq (5) Where, h is the convection heat transfer coefficient, A is the surface area, Tw and T\u221e are the surface and fluid temperature respectively. A 20kW PM traction motor is modeled as shown in Fig. 1. Some assumptions have been made to make the calculating easy: a) Since water velocity is far less than the sound velocity, so water can be handled as incompressible fluid; b) Water velocity is a constant, so dealing the water as steady flow; c) Because strand insulation, layer insulation and major insulation are all very thin, do not handle them respectively, handle them together as an entity. (From left to right: 20kW PM motor, the whole motor model, model of fluid and 1/16 model) The Selection of the Best Inlet Velocity" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003948_amm.423-426.1936-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003948_amm.423-426.1936-Figure8-1.png", + "caption": "Fig. 8 Internal structure of upper beam", + "texts": [ + " Reference points for deflection value calculating are shown in Fig. 7. Deflection value is set as f , the calculating formula of f is 1\u03b4 and 2\u03b4 are vertical displacement values of reference point 1 and point 2 respectively, L is the width of upper beam. Through calculating, f =0.203mm/m, which is more than 0.2mm/m, this indicates stiffness of the upper beam is insufficient and need to be optimized. Structural optimization for upper beam Upper beam is consisted of standard steel plates which are welded into a whole. The names of plates are shown in Fig. 8, the top plate of upper beam is hidden to display internal structure clearly. Since upper and lower skew plate have a relatively small impact on the stiffness and mass, thickness of front and back sidewise plate, middle sidewise plate, lengthwise plate are determined as design variables for structural optimization which are written as 1t , 2t , 3t . In order to obtain relationship between stiffness, mass and design variables, experiment design method is used to determined samples of design variables which will be calculated later" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002108_20130708-3-cn-2036.00046-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002108_20130708-3-cn-2036.00046-Figure1-1.png", + "caption": "Fig. 1. Ship\u2019s co-ordinate systems.", + "texts": [ + " The considered multi-controller control system structure is studied by means of a 3DOF nonlinear mathematical model of ship\u2019s slow-varying motions, which has been developed on the basis of tests carried out on a physical model in an American ship model basin (Wise and English 1975). The yaw angle and the ship\u2019s position in DSP are defined in an Earth-based fixed reference system the axes of which are directed northwards (N) and eastwards (E), and the origin of which is located over the drilling point on the seabed. In contrast, force and speed components with respect to water are determined in a moving system related with the ship\u2019s body and the axes directed to the front and the starboard of the ship with the origin placed in its gravity center. These are shown in Fig. 1. The mathematical description of the plant is given in the form of nonlinear state space and linear output equations: 1 4 3 5 3 2 4 3 5 3 3 6 2 4 5 4 5 6 1 3 5 5 5 4 6 6 6 2 6 4 5 5 6 6 3 cos sin cos , sin cos sin , , 0.088 0.132 0.958 0.958 , 1.4 0.978 / 0.543 0.037 0.544 , ( 0.764 0.258 0.162 ) / , c c c c s s s s x x x x x V x x x x x V x x x x x V x x u x x V x V x x x x u x x x x V x x u a y = \u2212 + \u03a8 = + + \u03a8 = = \u2212 + + = \u2212 \u2212 \u2212 + + = \u2212 + \u2212 + \u027a \u027a \u027a \u027a \u027a \u027a 1 1 2 2 3 3, , ,x y x y x= = = (1) where: 2 2 4 5( ) ( )sV x t x t= + is the translational velocity of the ship masured with respect to water, 2 0.0431zza k= + describes the ship\u2019s inertia moment together with water associated with the angle motion of the ship around its vertical axis. The 2 zzk is the square of the relative inertia radius referenced to the ship\u2019s length ppL , and c V and c \u03a8 are, respectively, the velocity and direction of the sea current as indicated in Fig. 1. All the signals appearing in the equations (1) are dimensionless, i.e. related to the ship\u2019s dimensions and displacement together with the dimensionless time Wind disturbances are among the most important environmental disturbances occurring in the dynamic positioning of drillvessels. In general wind forces and moment for each degree of freedom are defined as follows: ( ) ( ) ( ) 2 2 2 0.5 0.5 0.5 . x x p a p x y y p a p y z z p a p y pp F C V S F C V S M C V S L \u03b3 \u03c1 \u03b3 \u03c1 \u03b3 \u03c1 = = = (2) where: x C and yC are the force coefficients and z C is the moment coefficient, and where a \u03c1 is the density of air, x S and yS are the transverse and lateral projected areas, while pV and p\u03b3 are the relative wind speed and directions" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002357_amm.571-572.326-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002357_amm.571-572.326-Figure2-1.png", + "caption": "Fig. 2 Feedback control", + "texts": [ + " Left and right front wheel angle is not the same, we assume that: 1 2( ) / 2\u03b4 \u03b4 \u03b4= + (1) When the vehicle is steering, steering radius and vehicle front wheel angle have the following relationship: tan L R \u03b4 = (2) Feedforward and feedback tracking algorithm There exists certain deviations, including heading deviation \u03c8 (the deviation between the actual and the desired heading) and distance deviation d (the distance between the center of rear axle and the section of path), when the vehicle is tracking one segment of the path. As shown in figure 2, in order to eliminate the deviation of two aspects and realize the vehicle\u2019s path tracking, the feedback control [2] is used firstly. Its inputs are heading deviation \u03c8 and distance deviation d . The feedback output is front wheel angle 1\u03b4 . 1 arctan kd v \u03b4 \u03c8= + (3) Where k is proportional coefficient, v is vehicle speed. The curvature of each path point can be calculated. In order to realize the vehicle tracking precisely on a small space and reduce the possibility of deviation occurring in advance, the feed forward control is added in this paper" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003143_amc.2014.6823298-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003143_amc.2014.6823298-Figure7-1.png", + "caption": "Fig. 7. Experimental system.", + "texts": [ + " First, CGC generates more accurate weight com pensation torque than the wire-pulley mechanism, without wire and pulley since it consists of rod-driven mechanism. Second, this mechanism can be installed into heavier manipulators, while ensuring high rigidity and safety, because there is no risk of amputation of wire. Therefore, CGC is adopted for bilateral control in this paper. IV. EXPERIMENTS In this section, the performance of CGC is experimentally confirmed by conducting bilateral control on it. A. Experimental Setup The master and slave system used in this study is shown in Fig. 7. The length of link is 300 mm and a scale weight is fitted up at the tip of the link. The weight of the link including the scale weight is 0.75 kg. Using theses parameters of TABLE I and Eq. (20), the spring constant can be easily calculated. This CGC contains the compression springs at the bottom of the link, which has 0.57 N/mm of spring constant. This experimental system consists of two rotary motors and CGc. When the operator manipulates the master robot, the slave robot contacts the object" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003599_1.c031306-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003599_1.c031306-Figure7-1.png", + "caption": "Fig. 7 Illustrations of a) vertical cross section of interaction between tire, stone, and ground and b) stone velocities after impact from tire and rebound from ground.", + "texts": [ + " For an aircraft tire, the vertical deflection is also a function of the aircraft speed, due to greater lift at higher speeds. An initial approximation would be to assume that the deflection varies with the square of the tire speed, taking a maximum value when the tire is stationary and having zero deflection when the aircraft reaches the takeoff or landing speed: 0 1 V=VT 2 (9) A starting point for developing a lofting model governed by a hammer mechanism involved simplifying the problem by treating the tire\u2013stone and stone\u2013ground interactions as independent events. Figure 7a details the contact between the tire, stone, and ground. This model relies on the assumption that although the stone receives forces from the two surfaces simultaneously, because the surfaces are not identical, the stone will leave one surface before the other. The simplified two-dimensional loft model developed here treats the stone, ground, and tire as rigid and later introduces friction into the interactions. A. Frictionless Tire\u2013Stone and Stone\u2013Ground Interactions Consider the motion of a tire rolling over a stone as a rigid-body collision.Amassmt is assigned to the section of tire striking the stone to enable calculation of the momentum that may be transferred. Later, the tire mass is assumed to be much greater than the stone mass. Immediately before impact (Fig. 7a), the absolute component of the tire velocity vn along the line connecting the centers of mass of the stone and the tire is vn vz cos z Rc r vz (10) where the vertical velocity component of the section of tire descending upon the stone can be approximated using Eq. (1). In the special case of frictionless interactions, vt !0 0. By combining the definition of the coefficient of restitution with conservation of momentum, the final speed of the stone after the collision with the tire is [6] vf 1 et mtvn ms mt (11) A diagram detailing the kinematics of the impact between the stone and the ground is shown in Fig. 7b. If the ground is smooth, the velocity component of the stone parallel to the ground is unaffected. The vertical component of the stone loft velocity is given by vlz egvf cos (12) Combination of Eqs. (10) and (11) and use of the geometrical relations in Fig. 7b then yields eg 1 et ms=mt 1 Rc r 2 b2 Rc r 2 vz eg 1 et ms=mt 1 1 b Rc r 2 vz (13) If the mass of the stone is much smaller than the mass of the tire, i.e., ms mt, then vlz eg 1 et 1 b Rc r 2 vz (14) The maximum value of vlz would occur when eg et 1, b 0 >vlz 2vz. The maximum value of vz within the range of expected stone and tire diameters is vz 0:5V. Hence, themaximum vertical loft velocity is approximately the speed of the aircraft. However, this scenario would not be possible, since the stone would strike the tire after it rebounded from the ground", + " If the offset distance is such that b > 0, the stonemay still clip the lower surface of the tire after bouncing off the ground. The vertical speed of the stone would be reduced by this rebound, but its maximum speed may still be approximated by the expression above. The actual loft speed would depend on the number of rebounds between the stone, the ground, and the tire. B. Tire\u2013Stone and Stone\u2013Ground Friction When friction is introduced between the tire and stone, the tangential component of the impact velocity can cause the stone to spin, as shown in Fig. 7a. Assuming that there is no slip at contact point C, the stone is given an angular velocity: !0 vt r vz sin r yvz r Rc r (15) For a spherical stone, the ground friction will not affect the vertical loft velocity, but will influence the spin and horizontal velocity. Applying general equations of motion for impacts with friction [6] to the current loft model gives vly vfy g 1 eg vfz (16) !l !0 g r l2s 1 eg vfz (17) This leads to the following translational and angular velocities for the lofted stone: vly 1 et ms=mt 1 bz Rc r 2 g 1 eg 1 b Rc r 2 vz (18) D ow nl oa de d by W E ST E R N M IC H IG A N U N IV E R SI T Y o n Ja nu ar y 27 , 2 01 5 | h ttp :// ar c" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002517_amm.658.299-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002517_amm.658.299-Figure2-1.png", + "caption": "Fig. 2. The displacement of the inner ring", + "texts": [], + "surrounding_texts": [ + "The rating life of ball bearings directly depends on balls and races contact loads. To compute these forces, a quasi-static model is requested. The technical literature presents some quasi-static and dynamic models, developed for the ball bearings analysis [1,2]. Paleu [3] developed a five degrees of freedom quasi-static model for ball bearings, but he did not study the effect of inner and outer races misalignment on balls and races load distribution. Recently, a quasi-static model was presented for roller bearings [4], emphasizing the effect of the races misalignment on load distribution within the bearing, and also on the length of the roller profile. In this paper the authors present a quasi-static model about 3 degrees of freedom for the inner ring (tilting plus axial and radial displacements). In order to solve the quasi-static equilibrium of angular contact ball bearings elements, a vectorial and modular model is developed. The effects induced by the centrifugal forces and misalignments between bearing\u2019s rings were considered. The numerical simulations of the ball bearing quasi-static equilibrium show the major contribution of the rings misalignment upon ball and races load distribution." + ] + }, + { + "image_filename": "designv11_84_0003760_gt2011-45576-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003760_gt2011-45576-Figure1-1.png", + "caption": "Figure 1 Schematic of a HAFB showing preload, set bore clearance and pad configuration", + "texts": [ + " The design was based on extensive rotordynamic analysis and parametric study to optimize the bump stiffness and top foil contour. The design is based on hydro-dynamically preloaded three-pad configuration with one hydrostatic orifice per each pad. They also extended the concept of the HAFB to thrust foil bearing [8] with novel radially-arranged bump foils for easy stiffness control and prediction. They present dynamic performance of the bearing from their computational model. The bearing discussed in this article has a three pad configuration as shown in Figure 1 with each pad having an arc angle of 120\u00b0. Each pad has its center offset from the global bearing center by a small distance, rP. This type of configuration gives varying nominal clearance around the circumference of the bearing, with maximum clearance at the leading and trailing edges of the top foil and minimum set bore clearance, CSB, at the center of the arc length of the top foil. Non-dimensional hydrodynamic preload is defined as 1 SB p p SB C R r C (1) The hydrodynamic preload is different from mechanical preload of the top foil generated by loose contact between the top foil and bump foils when the bearing is assembled onto the shaft" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002120_j.euromechsol.2013.11.016-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002120_j.euromechsol.2013.11.016-Figure3-1.png", + "caption": "Fig. 3. A rigid body undergoing a rotational motion.", + "texts": [ + " It is important to remark that unit vectors u, v and w have been completely defined in terms of the position vectors, p1, p2 and p3, of three noncollinear points pertaining to a moving rigid body. Hence the pose of the body under analysis has been fully characterized in a simple and convenient way. 2.2. Rotation matrix The attitude of the body under analysis may be represented by a suitable rotationmatrix that serve to describe the orientation of the moving frameM : UVW with respect to a fixed frame F : XYZ,3 see Fig. 3. This rotation matrix is computed next. 4 For simplicity purposes in the notation, a dot over a symbol, e.g., _q, will denote the first derivative of q with respect to time, t, that is _qhdq=dt. For more details about the importance of reference frames in connection with time-differentiation of vectors and scalars, the reader is referred to Kane and Levinson (1985). Consider the rigid body undergoing a rotational motion about a fixed point O shown in Fig. 3. Attached to the rotating body is a moving frame M : UVW . A position vector r is used to represent the position of a point P pertaining to the moving body. Under such circumstances, position vector r maintains a constant magnitude and orientation in the UVW frame, that is, its coordinates u, v and w remain fixed during the rotational motion. However, its Cartesian coordinates (measured in the fixed system F : XYZ), namely, x, y, and z are continuously changing whereas the rotational motion is occuring" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001952_robio.2011.6181704-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001952_robio.2011.6181704-Figure4-1.png", + "caption": "Fig. 4. This sketch illustrates the planes of LAURON corresponding to which the 6D vector of forces and torques shall be measured (the forces along the cutting edges of the planes, the torques in parallel to the planes. As the actual measurements are performed in the 18 joints, these individual 1D readings have to be merged and transformed into 6D reading according to the planes representing the accumulated force applied to the body.", + "texts": [ + " The task is performed in three steps: step1: For each leg: Joint-torques are estimated by mea- suring the motor currents step2: For each leg: The three individual torques are transformed to a 6D force vector relative to the robot\u2019s hip based on the system\u2019s Jacobian. step3: Finally the 6 individual 6D vectors from the legs are transformed into the body coordinate system and merged into one resulting 6D force vector corresponding to body CoG. The resulting vector shall represent the forces and torques regarding the 3 body-planes of the robot: the sagittal plane, the transversal plane and the coronal plane (see Fig. 4). In the first step our interest is to determine the motor current of each joint. The measurement has to be performed online and on-board with a maximum of precision. The UCoM creates PWM-based voltages to control the motors of each leg. As a result we find a reasonable amount of noise in the currents. We have integrated a reference resistor to measure the voltage drop of this resistor and are thus able to calculate the current I(t) which is finally filtered with a combination of a median and mean filter to reduce the noise", + " The reason can be identified when having a look on the right side of the figure. The two sketches show the center of gravity (CoG - red \u2019X\u2019) and the center of pressure (CoP - green \u2019+\u2019) of LAURON as well as the force applied by the user (illustrated by the red arrow). As can be seen the force vector does not point to the CoP. Hence the applied force is actually applied tangetially, resulting more in a torque regarding the CoP than in a force. Pushing from the side results in a torque around the Y-axis lying within LAURON\u2019s Coronal Plane (compare to Fig. 4). In result the 3D torque vector provides the more significant values for the classification rather than the 3D force vector. In the experiments it further turned out that the movement commands and the rotate commands are difficult to distinguish because the user usually pushes at an easy to reach spot of the robot which usually gives both, a torque around the X or Y axis (move command) and the Z axis (rotate command). The user would have to reflect on his intentions and LAURON\u2019s current stance to identify the direction in which he has to push to assert a sideways force/torque without giving a torque around the vertical (Z) axis" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001486_romoco.2015.7219744-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001486_romoco.2015.7219744-Figure1-1.png", + "caption": "Fig. 1: Mechanical structure model of a nonholonomic manipulator, where: S1, S2 - ball gears, u1, u2 - control signals, \u03b81, \u03b82, \u03b83 - configuration variables", + "texts": [ + " We are not providing here any details about this controller, because this paper is focused only on NHM3 kinematics transformation for singularities avoidance purpose and feedback control was used only to show that proposed solution works well. The feedback controller synthesis was made in simulation computing environment. For more details about this feedback control used for NHM3 the Reader is referred to [1], where we consider the same feedback controller using TF, but the problem of singularities occurrence existed. II. DESCRIPTION OF NHM3 The NHM3 mechanical structural model is presented in Fig. 1. NHM3 movement is controlled by two separate engines placed in the basis of the manipulator. The torque is transmitted from those engines to individual links through two nonholonomic ball gears (NBG), a set of transmission shafts, belts and toothed gears. It is worth mentioning that for NHM3 there is a theoretical possibility of the full control of unlimited number of links with use of only two actuators. The nonholonomy of NBG comes from appearance of the nonholonomic constraints in the description of its movement [12]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003369_detc2011-48166-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003369_detc2011-48166-Figure3-1.png", + "caption": "Figure 3. THE BEARING CASE SUPPORT IS CONNECTED TO THE MACHINE BASE BY TWO ORTHOGONAL LOAD CELLS (H). LEAF SPRINGS (K) DECOUPLE THE X-Y DIRECTIONS. TWO ORTHOGONAL PROXIMITY PROBES (L) MEASURE THE SHAFT ORBIT.", + "texts": [ + " The actual motor could be replaced with another one with higher power coupled with a transmission to increase the rotational speed. An instrumented bearing (pos. A in Figure 2) is placed at the non-driven end of the shaft whereas a brand-new standard and equal bearing is placed at the driven end side (pos. B in Figure 2). The bearing housings are designed to hold bearings in the load-on-pad and in the load-between-pad configurations. Each bearing case support is connected to the machine base by two orthogonal 20 kN load cells (pos. H in Figure 3). The two directions are decoupled by means of leaf springs (pos. K in Figure 3). The load is applied in the middle of the shaft by means of two hydraulic actuators (pos. C in Figure 4) placed in an orthogonal configuration at 45 degrees with respect to the load Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/idetc/cie2011/70555/ on 05/23/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use 3 Copyright \u00a9 2011 by ASME cells as shown in Figure 4. The actuators are connected to the shaft by means of two deep groove precision ball bearings (pos" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure1.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure1.2-1.png", + "caption": "Fig. 1.2. Examples of visual disadjustment due to kinematic transformations (left) or observational reasons (right)", + "texts": [ + " 1 The Human Role in Telerobotics 15 The interaction with a remote robot is done through the human system interface, which transmits operator\u2019s actions and excites human senses according to the information received from the remote environment. Multi-modal human system interfaces refer to the perceptual modalities of human beings, such as visual, auditory, and haptic 2 modalities. Thus designing new devices the human sensing ability must be taken into account. Furthermore operator teleproprioception have to be considered. Operator telepropioception implies coherence between operator\u2019s commands and their execution. Fig. 1.2 describes some examples that show disadjustments between commanded and observed motions. Such disadjustments are due to kinematic transformations, observational reasons or relative movements between object and camera, which make the guidance references incoherent to the given visual references. Moreover aspects such as information redundancy and stimulus fidelity of the information provided to the human operator are essential in obtaining an accurate perception of the remote environment. The following sections review the human sensing abilities and provide a classification of state-of-the-art human system interfaces" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003959_s105261881101002x-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003959_s105261881101002x-Figure1-1.png", + "caption": "Fig. 1. Determining the rotor unbalance by the chronometric method.", + "texts": [ + " THE APPLIED METHOD There is a method to determine the value and position of the unbalance of a rotor in a centrifugal force field over three consecutive time intervals \u0394ti (i = 1, 2, 3) whose sum is the period of a full rotation of the rotor [1]. This method, with some changes, can be used to determine the mass and coordinates of the cen ter of mass of a body. It is based on the specific character of the plane parallel motion of a rotating dise quilibrated body around a vertical axis rigidly bound to a movable base mounted on resilient supports. This base makes a progressive transportation motion along a circular path affected by the centrifugal force with the radius proportionate to the value of the unbalance. Figure 1 shows circulating rotor 1 with the mass M and angular velocity \u03a9. The vertical geometric axis of the rotor passes through the point W. The center of mass S of the rotor is displaced with respect to this axis by the position vector \u03b5. Under the action of the centrifugal force, the rotor axis shifts from the equilibrium position Q by the position vector r. The rotor is linked to a moving reference frame W\u03be1\u03b71, whose axes \u03be1 and \u03b71 are thin opaque lines plot ted on the transparent disc 2, which consistently cross the optical axis D of photoelectric detector 3" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001136_978-3-319-19788-3_1-Figure1.1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001136_978-3-319-19788-3_1-Figure1.1-1.png", + "caption": "Fig. 1.1 A general parallel robot (the gray joints denote the actuated joints)", + "texts": [ + "arallel robots, also called parallel manipulators or parallel kinematic machines (PKM), are defined in (Leinonen 1991) as robots that control the motion of their end-effectors by means of at least two kinematic chains going from the end-effector towards the fixed base. From this definition, we see that the PKM are composed of different elements (Fig. 1.1): \u2022 the (fixed) base, which is the fixed element of the robot \u2022 the (moving) platform on which is usually mounted the end-effector, \u2022 the kinematic chains, linking the base to the platform, and also called the robot legs. A leg is usually a kinematic chain of serial or tree-structure type (Figs. 1.2 and 1.3). Parallel robots are very attractive for several applications because themanipulated load is shared by several legs of the system. Consequently, each kinematic chain carries only a fraction of the total load, which allows the creation of intrinsically more rigid robots" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002864_amm.611.194-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002864_amm.611.194-Figure2-1.png", + "caption": "Fig. 2 Deformation of teeth Fig. 3 Separation of loading on the line of contact", + "texts": [ + " (ID: 136.186.1.81, Swinburne University, Hawthorn, Australia-16/05/15,13:59:06) To determine the computer model for the studies of deformation of the teeth using FEM was necessary to determine material constants, define the type of the finite element, and to select appropriate boundary conditions (geometry and power). Using FEM, the deformation of teeth, which are simultaneously in the meshing, can be solved. We will focus on the value of the total deformation in the direction of action forces (Fig. 2). To determine the deformation of gearing under load, it is necessary to know the apportionment of the load on each gearing pairs with two pairs meshing. At the beginning, let us consider the simplest, ideal load apportionment, when the load is divided on two meshing pairs by half for each couple of teeth in the meshing (Fig. 3). To determine the resulting deformation of the teeth, it is necessary to determine the deformation of individual pairs of teeth by the following equation (1) \u03b4=\u03b41+\u03b42 (1) where, \u03b4 is the deformation of a pair of teeth [\u00b5m], \u03b41,2 are deformations of teeth [\u00b5m]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002845_s11249-011-9811-9-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002845_s11249-011-9811-9-Figure3-1.png", + "caption": "Fig. 3 Movement of entrapped lubricant in the contact region, HVI650, ue = 100 lm/s, W = 20 N, hd,o = 1.35 lm. a Displacement of reference point and dimple core; b Dimple profiles at different instants; c Film thickness of dimple core and minimum film thickness at outlet region; d Interferograms of entrapped dimple at different instants", + "texts": [ + " The experiments were carried out at a temperature of 20 \u00b1 1 C. Two types of lubricants, polybutene PB1300 and traction oil HVI650, were employed, and their properties are listed in Table 1. In this experiment, a reference point was marked on the steel ball or the disc surface to enable the actual entrainment velocity to be measured. Figure 2 shows the quasistatic film shape at the end of the impact when lateral movement was about to start. Two parameters of initial dimple depth hd,o and initial gap hg,o are defined in Fig. 2. Figure 3 shows a set of typical experimental results for HVI650. Figure 3a gives the temporal displacement of the reference point and the dimple core, and their speeds can be deduced by the slope of the curves. Initially, the dimple core moves with the entrainment velocity, but slows down to a speed of about one-third of the entrainment when it has moved beyond a critical displacement value, dc. Figure 3b shows the corresponding profiles of the dimple at different instants. The temporal variations of the dimple depth, hcore, and the outlet gap, hout, are shown in Fig. 3c. It can be seen that the overall entrapped lubricant moves along the entrainment direction and its depth starts significantly reducing at the first second, when the critical displacement dc is reached as illustrated in Fig. 3a. Beyond dc, the dimple shape is no longer axi-symmetrical about the vertical axis passing the dimple core, as shown in Fig. 3b. The variation of the outlet gap size hout, as illustrated in Fig. 3c, shows the drainage rate of the entrapped oil reaches the maximum at around the first second and then decreases slowly. The interferograms of the dimple taken at different intervals are depicted in Fig. 3d. The dimple maintains axisymmetry at the initial moving stage. When it is close to the EHL contact edge, the right side of the dimple becomes steeper than the left side. Furthermore, the lubricant entrainment at the inlet region has almost no effect on the left slope of the dimple. At the time (first second) when the critical displacement is, the axi-symmetry of the dimple is lost. In this studies, the effects of different parameters such as entrainment speeds, initial dimple depth hd,o, loads, initial gap hg,o and the viscosity of lubricants on the critical displacement were investigated" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002708_tee.21852-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002708_tee.21852-Figure9-1.png", + "caption": "Fig. 9. Three-dimensional FEM model of the 6/4 SR motor", + "texts": [ + " Characterization and torque computation Characterization and torque computation is realized by a program written in MATLAB as mentioned previously. Only six sine terms were found to be the optimum for the hybrid torque estimator. Since the maximum current was limited to 15 A and the current step was adjusted to 1 A, 15 \u00d7 6 = 90 current-dependent coefficients characterize the whole operating region. In order to verify the results computed by the model, 3-D FEM analysis of the motor with the mesh structure illustrated in Fig. 9 and static torque measurement were also done. The static torque was measured by a torque meter while the SR motor was blocked by an assembly at various fixed rotor positions. All results are shown together in Fig. 10. The graph is only plotted for 5, 10, and 15 A to make the curves clearly visible. As seen in the figure, the computed static torque curves are in very good agreement with measured and FEM results. The maximum error is about 0.18 Nm, which is 6% of the maximum measured torque. 3.2. Experimental results of online torque estimation For real-time implementation, the lookup table of the measured static torque and computed torque were inserted into the control and measurement file of the SR motor" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001136_978-3-319-19788-3_1-Figure1.17-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001136_978-3-319-19788-3_1-Figure1.17-1.png", + "caption": "Fig. 1.17 Examples of spatial robots with 3 translational or rotational DOF. a The Delta robot by Clavel (1990), a TPM. b The Orthoglide of IRCCyN (Chablat and Wenger 2003). c The Tripteron developed by Gosselin (2009), a TPM . d The Agile Eye developed by Gosselin et al. (1996)", + "texts": [ + " The number of DOF of the mobile platform is denoted ndof while the number of DOF of the entire robot is denoted Ndof planar motions of the platform, especially in the direction normal to the displacement plane, a recent idea was to design spatial robots able to achieve planar motions of their platform (Fig. 1.16). The large majority of PKM have been designed in order to be able to move their platform in the space. We call them the Spatial Parallel Manipulators (SPM). The robots of this category are too numerous to mention all of them. However, we can cite: \u2022 robots with three translationalDOF (also called translational parallel manipulators (TPM)): among them, we can mention the Delta robot (Clavel 1990) (Fig. 1.17a), the Orthoglide (Chablat and Wenger 2003) (Fig. 1.17b), the Tripteron (Gosselin 2009) (Fig. 1.17c), etc. \u2022 robots with three rotational DOF (also called spherical PKM): most of them allow the platform to rotate around one given fixed point (Bonev and Gosselin 2006). The most known is probably the Agile Eye (Gosselin et al. 1996) (Fig. 1.17d), \u2022 robotswith three exoticDOF: such types of robots have usually someDOF of rotation which are constrained with the DOF of translation [(see e.g. (Bonev 2008)]. Some of them have been designed with an additional wrist which compensates for the undesirable rotations and have found some industrial applications, especially for milling (Fig. 1.18) \u2022 robots with three translational DOF and one rotational DOF around one given axis (also called Sch\u00f6nflies motion generators): they are usually used for pickand-place operations, most often at high-speed" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure21-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure21-1.png", + "caption": "Figure 21 Adjustable wheelbases (see online version for colours)", + "texts": [], + "surrounding_texts": [ + "The chain force was calculated from the engine torque and the gear ratios (crankshaft vs. rear wheel). The control system that calculated the chain force also calculated the clutch contact ratio. That is, the chain force was reduced when the clutch skidded. The front and rear sprocket exert chain forces were parallel but acted in opposite directions (see Figure 16). Thus, the chain forces were in equilibrium and did not cause any translation of the entire bike (i.e., the bike was at equilibrium). However, the chain forces affected the movement of the swing-arm relative to the chassis. This movement affected the compression or extension of the rear suspension. The chain force control system is illustrated in Figure 20." + ] + }, + { + "image_filename": "designv11_84_0001693_12.2205795-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001693_12.2205795-Figure1-1.png", + "caption": "Figure 1. a) Set-up for measurements of the reflected light intensity changes in various gaseous atmospheres: 1 \u2013 light source - LED red, 2 \u2013 photodiode, 3 - a gas supply tube over the film, 4 \u2013 film; b) circuit diagram: D1-high efficiency red LED, D2-photodiode, Uz-power supply, V-voltmeter, R-the measuring resistance", + "texts": [ + " The structure of C-Pd films was investigated with transmission electron microscopy TEM, electron diffraction of selected area and Fourier transform infrared spectroscopy (FTIR). FTIR measurements were performed with the ThermoScientific Nicolet iS10 FTIR spectrometer, using ATR (Attenuated Total Reflectance) technique in the spectral range of 650-4000 cm-1 at the spectral resolution of 4 cm-1. The investigations of reflected light intensity changes of C-Pd and C-Pd-Ni films under the influence of various gases were performed in a measuring setup designed and built in Tele and Radio Research Institute. The experimental set-up (Figure 1) consists of a high efficiency red LED, a photodiode, a gas supply tube, a DC power supply, a voltmeter, a measuring resistance. Proc. of SPIE Vol. 9662 96624F-2 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 10/20/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx ' I ,\u00a1 lk}!', L L r 1, ti \"r.,1 .. ti r ' ll F 1 \\, r Is . \"\" .': I 'i ,. `'; 4 .1 ' ; ,,' ,1' ,,; Y ' t1. ' 1,`. '' .r k. . 1,rr,,1 , '% . .;Y,r , - \\'4`,.,' !A '.* . . ,y.1 _ -' .\", l1 4 .r 1r.1\u0300 1 S . '-, . s'.``,-- w s'__..,. The investigations were carried out for 1% hydrogen in nitrogen or for 2% methane in nitrogen or for 100 ppm ammonia in nitrogen or for 5% carbon dioxide in nitrogen, under atmospheric pressure in room temperature. The reflected light from C-Pd and C-Pd-Ni films was measured by photodiode (Figure1a). The source of light was a high efficiency red LED. The light reflected by the film was directed on the photodiode. Signal from photodiode was measured with a voltmeter on the probe resistor. Changes of voltage on probe resistor correspond to changes in reflected light from the sample. Measurements of reflected light changes for investigated films were performed cyclically. Firstly, the gas was passed over investigated film by the supply tube for about 5 minutes (gas flow speed 1000 ml/min). The voltage was carried out continuously with registration of the changes every 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002562_iecon.2013.6700100-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002562_iecon.2013.6700100-Figure9-1.png", + "caption": "Fig. 9. Experimental system.", + "texts": [ + " In brief, if glpf is enough large, the unexpected force as human input have little effect on the reproduced position to environment. From Fig. 8 it is similar about the effect from F l to F ext S . In the proposed method, it is turned out that the effect from the unexpected force to the reproduced motion to environment is varied by changing glpf . From these analysis, in other words, it is possible that the reproductivity is changed arbitrarily by the cut-off frequency of LPF. In order to confirm the validity of the proposed method, experiments were conducted. Fig. 9 shows the experimental system. Fig. 9(a) shows the motion-saving system. This system is composed of two linear motors and the bilateral control is implemented between these system. Besides, another motor is used as an environment, which is the impedance-control is implemented. The stiffness K and damper D of the impedancecontrol system are set 300 N/m and 20 Ns/m respectively. By utilizing the motor as an environment, the haptic information which are position and force information can be obtained. In this system, an operator manipulates the master system and the slave system actually acts on environments. The force information and position information of master system are stored into the motion data memory. In proposal, the force information of the slave system is stored at the same time While, Fig. 9(b) shows the motion-loading system. This system uses the only slave system and the saved motion is reproduced according to the motion data memory. In addition, another motor is used as an input motor which applies the unexpected force to the motion-loading system for examining the performance of the proposed method. Table I shows the control parameters in the experiments. First, the experimental results of motion-saving system is shown in Fig. 10. From this figure, it turns out that bilateral control is conducted" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001046_20070903-3-fr-2921.00038-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001046_20070903-3-fr-2921.00038-Figure1-1.png", + "caption": "Fig. 1. Basic scheme of Optical Flow", + "texts": [], + "surrounding_texts": [ + "The IMU module is a low-cost Inertial Measurement Unit based on inertial sensors. These inertial signals are acquired using Advantech PCL-818HG (16 channels A/D) card connected to PC target. The on-board inertial sensors are three surface-micromachined angular rate sensors (gyroscope ADXRS150) arranged orthogonally and a dual-axis micromachined silicon accelerometer (ADXL203). The three angular rates ( \u00b7 \u03c6, \u00b7 \u03b8, \u00b7 \u03c8) are given by the gyros and the two angular positions (\u03c6, \u03b8) are estimated using a complementary filter (Shmuel, 1996). Data fusion is used to estimate the angular position of the aircraft by appropriately combining the accelerometers and gyros measurement signals. The gyro signal \u00b7 \u03b8g is passed through a filter F1(s) while the accelerometer measurement \u03b8a is passed through a filter F2(s). The estimate \u03b8\u0302 is given as \u03b8\u0302 = F1(s) \u00b7 \u03b8g + F2(s)\u03b8a. The complementary filters are such that sF1(s) + F2(s) = 1 where F2(s) is a first order low-pass filter k k+s . For k = 0, F2(s) = 0, and F1(s) becomes a pure integration and \u03b8\u0302 diverges due to gyro drift. On the other hand for k = \u221e, F2(s) = 1, and F1(s) = 0 and the data from the gyro is completely suppressed. The complementary filter scheme is given in figure 2. The filter parameter k has been selected in practice to obtain an estimate \u03b8\u0302 that is as close as possible to the actual value of \u03b8. 5. CAMERA CALIBRATION Camera calibration is the process of determining the optical and internal camera geometric characteristics (intrinsic parameters) and the position and orientation of the camera with respect to a certain world coordinate system (extrinsic parameters) (Hartley et al., 2004). The simple webcam logitech has been used and the camera parameters are estimated by the two planes method which gives us a simple camera characterization (J. Fabrizio et al., 2002)(K. Gremban et al., 1988). In this case the objective is to obtain a relationship between the displacement in the image plane expressed in pixels and the realworld displacement expressed in meters. The helicopter\u2019s altitude is regulated around the desired value d = 0.5 m which implies that a 0.01m relative displacement is represented by a 4 pixels displacement in the image plane. 6. ROTORCRAFT DYNAMICAL MODEL A four-rotor craft (figure 3) is controlled by varying the angular speed of each one of the rotors. The force fi produced by motor i is proportional to the square of the angular speed, that is fi = k\u03c92 i . The front and rear motors rotate counterclockwise, while the other two motors rotate clockwise. Gyroscopic effects and aerodynamic torques tend to cancel in trimmed flight. The main thrust is the sum of the thrusts of each motor. The pitch torque is a function of the difference f1 \u2212 f3, while the roll torque is a function of f2 \u2212 f4, and the yaw torque is the sum \u03c4M1 +\u03c4M2 +\u03c4M3 +\u03c4M4 , where \u03c4Mi is the reaction torque of motor i due to shaft acceleration and the blade\u2019s drag. The terms \u0393e = { \u0131\u0302e, \u0302e, k\u0302e } and \u0393b = { \u0131\u0302b, \u0302b, k\u0302b } denote the inertial and the fixed-body frame respectively and qe = [xe ye ze \u03c8 \u03b8 \u03c6]T = [\u03be \u03b7]T is the generalized coordinates which describe the vehicle\u2019s position and orientation. \u03be \u2208 R3, denotes the position of the vehicle\u2019s center of mass relative to the inertial frame, and \u03b7 \u2208 R3 are the three Euler angles. The dynamical model is obtained using the EulerLagrange approach. For simplicity the equations are separated into translational and rotational displacement as follows: Fe = m\u03be\u0308 + mg (3) \u03c4 = J\u03b7\u0308 + J\u0307 \u03b7\u0307 \u2212 1 2 \u2202 \u2202\u03b7 ( \u03b7\u0307T J\u03b7\u0307 ) where m \u2208 R denotes the mass of the vehicle, \u03c4 \u2208 R3 denotes the generalized momentum, Fe \u2208 R3 is the translational force applied to the rotorcraft, g \u2208 R3 denotes acceleration due to the gravity (g = \u2212gk\u0302e), J \u2208 R3x3 represents the inertia matrix, which is symmetric and positive definite. The Coriolis, gyroscopic and centrifugal terms are expressed as C (\u03b7, \u03b7\u0307) \u03b7\u0307 , J\u0307 \u03b7\u0307 \u2212 1 2 \u2202 \u2202\u03b7 ( \u03b7\u0307T J\u03b7\u0307 ) (4) The model equation (3) can be rewritten as m\u03be\u0308 = Fe \u2212mg J\u03b7\u0308 + C (\u03b7, \u03b7\u0307) \u03b7\u0307 = \u03c4 (5) The total thrust in the body frame is Fb = 4\u2211 i=1 Fi.k\u0302b = 4\u2211 i=1 k\u03c92 i k\u0302b (6) where \u03c9i is the angular speed of motor i and k > 0. Let us define u = \u22114 i=1 k\u03c92 i as the collective input, Fb can also be expressed in the inertial frame \u0393e as Fe = Rb eFb, where Rb e is a rotation matrix described in (P. Castillo et al., 2005). The total momentum is given by \u03c4 = \u03c4\u03c6 \u0131\u0302b + \u03c4\u03b8 \u0302b + \u03c4\u03c8k\u0302b (7) with \u03c4\u03c6 = (f2 \u2212 f1) l \u03c4\u03b8 = (f3 \u2212 f4)l \u03c4\u03c8 = \u03c4M1 + \u03c4M2 + \u03c4M3 + \u03c4M4 (8) where l is the distance from each motor to the center of gravity and \u03c4Mi is the reaction torque of motor i due to shaft acceleration and the blade\u2019s drag." + ] + }, + { + "image_filename": "designv11_84_0003678_1.4024704-Figure20-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003678_1.4024704-Figure20-1.png", + "caption": "Fig. 20 Distribution of von-Mises stress when face AA0B0B is fixed and welding is done away from the fixed face", + "texts": [], + "surrounding_texts": [ + "The present investigation, divided into Parts I and II, presents the mathematical model that incorporates all the physical phenomena involved in a welding process and utilizes it to predict the significance of the various welding parameters and helps to understand the underlying physics in better prediction of weld D/W ratio and welded joint mechanical response. Detailed parametric studies performed in Part II of the present study help to delineate the effects of various welding parameters on the weld D/W ratio. Simulations are run with temperature dependent thermophysical properties. The results have been compared with experimental data. The parameters of particular interest are welding current, welding speed, arc length or electrode gap, electrode angle, and surface active agent content. For each of these parameters, simulations were run for both low ( 40 ppm) and high ( 150 ppm) surface active agent content. The present investigation also predicts the weld solidification mode by analyzing weld pool dynamics simulation results. The CFD results show that with increase in welding speed finer dendritic substructures would be obtained. Variation of weld D/W ratio with heat input per unit length of weld is also calculated. The heat input is varied in two ways \u2013 (a) varying the welding current and (b) varying the welding speed. Changing the heat input per unit length of the weld by varying either the current or speed will influence the important weld pool driving forces differently in a nonlinear manner. The complex interplay of the driving forces determines the final weld D/W ratio. Hence, when analyzing weld D/W ratio plots versus heat input per unit length of weld, it is important that attention be paid as to how the changes in heat input were achieved. Structural analysis of the welded joint highlights the deformation of the welded butt joint under various constraint configurations. The effect of welding direction (toward/away from) the fixed edge/face has also been studied. The von-Mises stress distributions under these constraint configurations have also been highlighted. The coupled field modeling of GTA welding provides a more accurate thermal energy distribution for structural analyses. Through material modeling the effects of microstructure evolution on mechanical behavior has been indirectly incorporated. Thermally induced stress distribution and workpiece deformations 021009-10 / Vol. 6, JUNE 2014 Transactions of the ASME Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 02/19/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use have been reported. The mathematical framework developed here provides us with the tools to study the multiphysics problem of welding. The present model works very well for welding currents 200 A. The predictions from the present model will not agree very well with thicker plates and higher welding currents. Hence, the present model needs to be extended to incorporate arc pressure effects and filler material addition. Also, with higher welding current, surface deformation becomes an issue. Using volume of fluid (VOF) method, the present method can be extended to include the surface depression effects during welding. A future goal is to study stress relief methods on welded joints." + ] + }, + { + "image_filename": "designv11_84_0002996_s1052618814040098-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002996_s1052618814040098-Figure1-1.png", + "caption": "Fig. 1.", + "texts": [ + "61 1. To write the equations of motion, operator notation forms [1, 2] are applied. Following [3], let us consider the family of stationary scleronomic linear elastoviscous systems with full energy dissipation denoted below as A = (A0, A1,\u2026, AN) (Fig. 1). To each of Ar systems of family A corresponds to the dis placement field ur(xr, t) \u2208 R3, where vectors xr \u2208 Xr \u2282 R3 are the vector coordinates of the points of systems Ar, t \u2208 R, r = 0, 1,\u2026, N. It should be noted that in this paper the Newtonian collisions are always assumed to be replaced in the models by the interaction in certain rigid elastic buffers; for the figures not to be con gested, we use abridged notations (Fig. 1). The dynamics of all members of family A is determined by dynamic compliance matrix operator sys tems [1, 2] L(r)) (yr, xr; p) where p is the operator of differentiation. These operators have dimensionality [3 \u00d7 3] and assign displacement fields (1) to force fields fr(xr, t) [xr \u2208 Xr]. ur xr t,( ) L r( ) yr xr; p,( )fr yr t,( )= Every operator L(r) is an integro differential operator constructed either by means of the initial set of equations of motion and necessary auxiliary, for example, boundary, conditions or on the basis of pro cessed experimental data [4, 5]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002011_2011-01-1691-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002011_2011-01-1691-Figure1-1.png", + "caption": "Figure 1. FE model for a body-in-white", + "texts": [ + " TPA offers information and suggestion about contribution of the transfer paths and shows a dominant transfer path for taking effective measures. TPA is familiar to engineers in case of analyzing mounting of power plants and bushing of suspensions [3]. In these applications, degrees of freedom (DOF) to analyze the transmitted forces are discrete and scalar. The number of DOF is small. There are also applications of multi-points conjunction types as that of tire and wheel [4], [5]. However, conventional TPA has seldom been applied to automotive body structure consisting of many structural members and panels as shown in Figure 1. A major reason is that the vibration transmission analysis (VTA) is complicated by the many transfer paths. One approach was proposed to analyze the complex transmission of automotive body vibration by a proper coordinate transformation of analytical DOF at the cross sections [1]. In this approach, the DOF of many adjacent nodes are treated as a cluster, so analysts easily understand vibration transmission (VT). However, this approach by only one cutting section is insufficient to detect dominant paths for taking effective measures for the body that VT paths are spread on panels", + " The proposed approach is applied for cutting sections of all focused paths, and the VT characteristics of all paths are visualized. Considering the body deformation diagram indicated by influence degree (VT characteristic diagram), we are able to take more appropriate measures and efficient derivation of measures by the VT characteristics and spatial trends of all paths offers. SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 4 | Issue 2 1285 The proposed techniques are applied to booming noise using the FE model of automotive body structure as shown in Figure 1. Total DOF of this model is almost 2.5 million. Structural damping is defined 2% for glass panels and 0.1% for body panels and members. Booming noise is analyzed in the frequency band 10Hz-200Hz by input force and output response as shown in Figure 1. The input points are located at engine mount points, and the output point on the roof is selected as the representative point of interior noise assessment by vibration. NX Nastran Ver5.0 and I-DEAS Ver5.0 (SDRC) are used for analysis and handling of the model. From inspection of the frequency response of the roof as shown in Figure 7, where the highest peak is at 118Hz, reduction of this peak level was set to the target. The model for a body-in-white with numerous DOFs requires large computing time for transfer functions and transmitted forces for all DOF of the cross sections" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.10-1.png", + "caption": "Fig. 7.10 The Gough-Stewart platform (6\u2013UPS SPM)", + "texts": [ + " From a geometric point of view, solving these equations is equivalent to finding the intersection between a line Li defining the displacement of the active prismatic joints and a sphere Si that represents the displacement of the point Ai2 when the platform is fixed and the leg is virtually broken at point Ai2 (Fig. 7.9). The 6\u2013UPS PKM, also called the Gough-Stewart platform, is a robot composed of six legs, each leg being made of a passive U joint fixed on the base, followed by an active P joint and then a passive S joint (Fig. 7.10). The MDH parameters associated to the frames of Fig. 7.11 for one leg are given in Table7.5. For simplifying the computation, the base connecting points Ai1 are considered to all belong to the same plane (O, x0, y0). The parameters corresponding to the S joint are deliberately omitted. The computation of the S joint coordinates is of no interest in that section as they have no effect on the dynamic model if their corresponding friction are neglected (Khalil and Ibrahim 2007). We will deliberately limit the analysis of the IGM of the GoughStewart to the computation of the active joint coordinates only (that can be obtained through the use of the translational part of (7", + "144) As a result, AT r = [ \u03b6 r 1 \u03b6 r 2 \u03b6 r 3 ] (7.145) with \u03b6 r T i = 0rT Ai3 Ai4 . (7.146) Finally, we have Ar 0tr + Bq\u0307a = 0. (7.147) In this section, we study only the input/output kinematic relations of the GoughStewart platform introduced in Sect. 7.1.2.5. Following the method of Sect. 7.3.1, and by using the results presented in the AppendixC.4.1, we have a matrix A equal to: AT = [ \u03b6 1 . . . \u03b6 6 ] (7.148) with \u03b6 T i = 1 qi3 [ 0rT Ai1 Ai6 ( 0rP Ai6 \u00d7 0rAi1 Ai6 )T ] (7.149) where the points Aij are described at Fig. 7.10 and qi3 = \u2225 \u2225rAi1 Ai6 \u2225 \u2225 is the active joint variable for leg i , whose expression is given at (7.37). \u03b6 i is a pure force along \u2212\u2212\u2212\u2212\u2192 Ai1Ai6 which is reciprocal to all passive joint twists of the leg i (and not the active joint twists). Moreover, as shown in the AppendixC.4.1, for any leg of the robot, $i3 = $ia is a twist representing a pure translation along the P joint direction. As a result, 0$T ia = 1 qi3 [ 0rT Ai1 Ai6 01\u00d73 ] . (7.150) Finally, the matrix B is equal to: B = \u2212 \u23a1 \u23a2 \u23a2 \u23a3 \u03b6 T 1 0$1a 0 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure6.5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure6.5-1.png", + "caption": "Fig. 6.5 Access to the generative design module", + "texts": [ + " An organic type geometry resulted after the algorithms were applied to the original configuration. The functionality of the generative design was only active in two modes: (1) on a commercial version of 30 day-trial, and (2) on Autodesk accounts that have not been activated as educative versions. Account licenses can be as a monthly version or annual renewal. To start a new study of generative design we need to access the work space of Autodesk\u00ae Fusion 360\u00ae Generative Design which can be found on the upper-left corner of the window, see Fig. 6.5. At this point, the work flow is very similar to the design proposed by Krish [9] where the designer modifies the model features and the restrictions to generate the iterations, see Fig. 6.6. Once the program is in the environment frame, tools are activated and buttons will be selected from left to right, see Fig. 6.7. 2. Model editing: Strongly related to point 3; geometry is defined in this mode Fig. 6.6 Generative design process proposed by Krish [9]. Reprinted with permission from Springer Nature publishers Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure8.41-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure8.41-1.png", + "caption": "Fig. 8.41 Zones were the orthesis from the clubfoot showed deformation. Interferometry in figure (a) and FEM in figure (b)", + "texts": [], + "surrounding_texts": [ + "analysis for the clubfoot and healthy footwere 1.392mmand 1.276mm, respectively. In Figs. 8.41 and 8.42, it is shown the matched zones of both analysis in the printed ortheses.\nLike in Figs. 8.41 and 8.42 zones that match in the deformation, in Figs. 8.43 and 8.44 the lateral side zones of deformation also matches.\nAnother result observed in the FEM but not in the interferometry study is in the floor stand, where there is wear in the floor stand especially in the heel zone as it can be seen in Fig. 8.45.", + "Through 3D printing, it has been possible to develop ideas that previously remained in the drawing board to something tangible without the need of a machining center or through some manufacturing procedure that is not available to everyone. In addition", + "to 3D printing, there are technologies such as scanning process and CAD programs that help the design of 3D models through reverse engineering.\nAvailable ortheses for musculoskeletal conditions like clubfoot are generally costly. CILAB laboratory creates custom 3D printed orthoses that are accessible to their patients without a big impact, help on their economy, and correct the pathology they present." + ] + }, + { + "image_filename": "designv11_84_0002731_icma.2013.6618005-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002731_icma.2013.6618005-Figure3-1.png", + "caption": "Fig. 3 Robot used in the simulation.", + "texts": [ + " (8) These four powers i T i vf , 11 i T i vf , i T i \u03c9\u03c4 and i T i \u03c9\u03c4 1 depend on robot motion respectively (see appendix). The goal of our research is to develop a method to coordinate multiple actuators in order to appropriately distribute energy to each body of a robot through the kinetic chain according to a desired motion. In this paper, we simulated robotic throwing motions caused by the coordination of multiple actuators. And furthermore, we compared the simulated robot motions in terms of energy flow in the kinetic chain. Fig. 3(a) shows a two-link manipulator used in the simulation. For simplicity, it was modeled as a two-dimensional robot in the x-z plane. Table II shows the physical parameters of the manipulator. Fig. 3(b) shows the arrangement of six artificial muscle actuators. This arrangement is based on muscle arrangement in a human upper arm. For simplicity, the moment arms of all artificial muscle actuators about all joints were assumed to be the same and constant with changes in joint angle (i.e., 0.04 m). Fig. 4 shows the simulated robot motions. The robot motions were simulated by inputting the joint torques shown in Fig. 5 respectively. The robot motion A was simulated by applying the torques to the first joint and the second joint at the same time" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001567_icma.2015.7237783-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001567_icma.2015.7237783-Figure7-1.png", + "caption": "Fig. 7 Neck mechanism", + "texts": [ + " Each adjacent layer is connected with linear guide mechanism. The top tier and middle tier could make relative slide forward and backward. The middle tier and bottom tier could make relative slide leftward and rightward. Robot\u2019s chin and upper jaw are connected by shaft. So the upper jaw and lower jaw could open and close. Combining all the movement above, jaw mechanism could make masticatory movement. Neck mechanism has 3 DOFs to pitch, yaw and roll robot\u2019s head. It is a RSSR parallel mechanism, showed in Fig. 7. It is consist of three parts: the upper supporting plate, the lower supporting plate and the pedestal. Robot\u2019s head is fixed on the upper supporting plate. The upper supporting plate and lower supporting plate are connected by a fixed supporting shaft and two RSSR space linkage mechanism, which comprise a parallel mechanism. The lower supporting plate are connected with the pedestal by linkage mechanism. The two motors fixed on the lower supporting plate decide the direction of head rotation. When the two motors rotate same angels in the same direction, robot\u2019s head pitches" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000926_gt2008-50641-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000926_gt2008-50641-Figure2-1.png", + "caption": "Figure 2 \u2013 Two bumps strip", + "texts": [ + "s] f\u03bc Friction coefficient d\u03b8 Transmission force angle [rad] dbis\u03b8 Last bump transmission force angle [rad] soud\u03b8 Angular position of the top foil spot weld [rad] \u03b8 Angular position [rad] \u03a9 Rotating speed [rad.s-1] Superscript \u2022 Derivative with respect to time Coordinate systems ( )zyx rrr ,, Cylindrical coordinate system ( )ZYX rrr ,, Cartesian coordinate system Copyright \u00a9 2008 by ASME rms of Use: http://www.asme.org/about-asme/terms-of-use Dow THEORETICAL ANALYSIS The Foil Bearing Structural Model The mathematical model of the elastic structure is first presented for a strip made of two bumps (Fig. 2) and then it is extended to an arbitrary number of bumps. The model considers the continuous foil structure as an equivalent discrete one with a restricted number of nodes linked by linear springs. For instance, the model depicted in Fig. 3 has six degrees of freedom, namely two vertical displacements, 1v and 3v , and four horizontal displacements 1u \u2026 4u ; this is the minimum required for describing the coupling between two successive bumps. The elementary characteristics of the bump strip model are the stiffnesses ik and the transmission force angle d\u03b8 that can be analytically expressed for each type of bump" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003141_cp.2013.2541-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003141_cp.2013.2541-Figure6-1.png", + "caption": "Figure 6. CAD Model of Bent sheet", + "texts": [ + " List of Programs for a robot to perform a bending operation Master Reset Inputs/outputs Sub program for different sheet size The figure5 shows in detail the logic of the robot which handles the sheet metal part and also feeds the part into the bending machine where the bending takes place. The figure4 explains the sequence of movements which the robot follows inorder to perform the required task. The program is checked and interfaced with HMI to run in the remote mode and then the production is carried out which has been recorded and there by observed improvements in the throughput which are included in the results. Figure 6 shows the CAD model of the bent sheet which is obtained by using robot to handle and feed during bending. Figure 6. Productivity parts/hr) The graph shows a linear behavior of the productivity in terms of (parts/hr).There is an increase in productivity by 3 parts/hr when compared with the line data before programming the material handling robots By automating material handling using robots following benefits has been incurred which are as follows: Total Variable cost per annum can been reduced by 21.36%. Throughput rate can be increased by 59.72%. Productivity has been increased to 1.7 times the original. From the above studies it can be concluded that the path planning of material handling robots at Bending Workstation yields the following points" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003259_gt2013-94105-Figure12-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003259_gt2013-94105-Figure12-1.png", + "caption": "Figure 12: Idler gear instrumented tooth types", + "texts": [ + " Figure 11 shows the gear test rig configured for gear tooth inspection by a dynamic inspection idler wheel (white gear wheel), with eddy current sensors embedded in the teeth. The idler inspection wheel is easily replaced with an engineered steel gear wheel for gear-to-gear tests using a static eddy current sensor mounted adjacent to the gear wheels. For the gear-to-gear tests, a standard 10mm-diameter sensor was fitted onto an adjustable mount, and placed in various positions adjacent to the gears. The idler gear wheel used a plastic gear wheel fitted with three ferrite-cored probe coils. The coils were designed to be 2mm diameter of different lengths. Figure 12 illustrates the types. To cover larger amounts of the gear flank surface shaped eddy current sensors could be fitted, rectangular for example. 7 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/asmep/76989/ on 07/08/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use The idler gear wheel was machined to have the same module as the steel test gears except it has 24 teeth whereas the steel gears have 25 teeth, permitting a rolling scan of all teeth over 25 revs", + " The resultant signals from the scan are therefore always prominently positioned at the minima position (maximum sensing condition) of the system output. Inspection of the test data revealed that fault types 4 and 5 were not readily detected by the eddy current tooth inspection sensors, however, fault types 1, 2 and 3 produce distinct shape changes in the signal (see Figure 17), which on inspection are sufficiently different to be detected by automatic algorithms. Certainly a change from baseline operational conditions would be very easy to detect. Galling was best detected with the type 3 face sensor (see Figure 12). The type 2 sensor tended to integrate the signal from both faces of the teeth engaging either side of the scanning tooth. Figure 18 shows the signal from the type 3 face sensor, for an undamaged gear (blue trace) and compares it to the signal from the galled gear (red trace). The upper plot shows the sensor signal for a group of teeth exhibiting signs of galling damage. The reduction in amplitude for the galled gear, relative to the healthy gear, is clearly apparent for all of the teeth shown" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002673_holm.2014.7031067-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002673_holm.2014.7031067-Figure8-1.png", + "caption": "Fig 8. The permanent magnet returning structure schematic diagram", + "texts": [ + " From armature returning mode, the AEMR can be classified into four types: spring returning, permanent magnet returning, permanent magnet magnetic circuit returning and polarized magnetic circuit returning. Clapper type relay generally uses spring to make armature return, it belongs to the structure of spring returning. Balance armature type AEMR can use both spring and permanent magnet to accomplish the return process. Spring returning type relay (showed in Fig. 7) is used widely. To meet the requirement of vibration resistance, the spring can be replaced by permanent magnet (showed in Fig. 8). In Fig. 8, the permanent magnetic circuit is an independent open magnetic circuit, which has a low magnetic efficiency. Compared with the spring returning mode, the advantage of permanent magnet returning is that it has a strong retentivity force when the armature is on the steady release state. And the disadvantage is that in a small initial stage the force is very low when armature is releasing. It is because of that the gap between the armature and permanent magnet is very large. Thus the speed of armature is low" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003184_2542355.2542368-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003184_2542355.2542368-Figure4-1.png", + "caption": "Figure 4: Capture-Point \u201cDistance\u201d - Estimating the capture point \u201cdistance\u201d based on a rigid mass-less support leg.", + "texts": [ + " For a pin-point rigid-leg pendulum model, we calculate the destination leg-length for the step transitions necessary based upon the foot placement distance that would result in the mass reaching a zero velocity when vertical. We illustrate the capture-point distance and capture-point height in Figure 3. Capture-Point \u201cDistance\u201d: The capture point \u201cdistance\u201d is the specific foot position from the current projected location on the ground that will bring the pendulum to a stop (i.e., velocity will reach zero when the pendulum is standing vertically upright and straight), as shown in Figure 4. This method was proposed by Pratt and Tedrake [2006] who applied it to both a pendulum model (i.e., arc like trajectory) and linear-pendulum model (i.e., flat fixed height trajectory). Capture-Point \u201cHeight\u201d: In contrast to the capture-point \u201cdistance\u201d, which focused on finding the unknown stepping distance necessary to bring the pendulum mass to a vertical upright stop, the capture-point \u2018height\u2019 focusing on finding the final leg-height given a specific stepping distance to achieve the same task" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001446_icuas.2015.7152381-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001446_icuas.2015.7152381-Figure3-1.png", + "caption": "Fig. 3. Banked turn", + "texts": [ + " 2), it is not possible to assume that the instantaneous direction of flight coincides with the longitudinal axis of the airplane, thus an sideslip angle \u03b2 exists [11]. This way of turning is not commonly used since the sideslip angle is not desirable. In addition, the turning radius R is big, and it takes a lot of time to make the turn. B. Banked Turn One of the many discoveries made by the Wright brothers is that a fixed wing aircraft is best turned by rolling. Rolling means to perform the turn with ailerons (banked turn) instead of rudder [1]. This maneuver is employed in all the regular cases [11]. Figure 3 shows \u00b5 that is known as bank angle, where \u00b5 = \u03c6. C. Coordinated Turn An additional way to turn is by coordinating the flat turn with the banked turn. The banked turn is more efficient than the flat turn, however with the addition of the rudder action it becomes easier to control the sideslip angle. In a coordinated turn, the net side force in the aircraft\u2019s body frame is zero. A coordinated turn is produced when aileron and the rudder work together to turn and change the aircraft direction [1]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003599_1.c031306-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003599_1.c031306-Figure8-1.png", + "caption": "Fig. 8 Kinematics of asperity lofting of a stone radius r.", + "texts": [ + "0 g r l2s 1 eg vfz (17) This leads to the following translational and angular velocities for the lofted stone: vly 1 et ms=mt 1 bz Rc r 2 g 1 eg 1 b Rc r 2 vz (18) D ow nl oa de d by W E ST E R N M IC H IG A N U N IV E R SI T Y o n Ja nu ar y 27 , 2 01 5 | h ttp :// ar c. ai aa .o rg | D O I: 1 0. 25 14 /1 .C 03 13 06 !l b r Rc r r l2s g 1 eg 1 et ms=mt 1 1 b Rc r 2 vz (19) vlz eg 1 et ms=mt 1 1 b Rc r 2 vz (20) C. Asperity Lofting This section briefly considers the potential lofting of a rigid stone of irregular geometry, in which the irregularity is idealized as a single protrusion on the surface of the stone with a height p above the average radius of the stone (Fig. 8). An interaction between the protrusion and the ground can lead to the stone attaining a vertical velocity in two ways [1]. The protrusion can dig into the ground causing a horizontallymoving stone to be levered upward by rotating about the asperity tip. Alternatively, a spinning stone, whether initially translating or not, can be propelled with a vertical velocity as the protrusion strikes the ground. The following equations can be used to quantify the ratio of the vertical loft velocity to the initial horizontal or tangential surface velocity of the stone: d r p 2 r2 p (21) vz d" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000833_978-3-540-77974-2_13-Figure13.7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000833_978-3-540-77974-2_13-Figure13.7-1.png", + "caption": "Figure 13.7 One convex polygon is more extreme than another for a connected range of directions", + "texts": [ + " To this end we model the set of all directions by the unit circle centered at the origin: a point p on the unit circle represents the direction given by the vector from the origin to p. The range from a direction d1 to a direction d2 is defined as the directions corresponding to points in the counterclockwise circle segment from the point representing d1 to the point representing d2. Note that the range from d1 to d2 is d2 d1 directions between d1 and d2 not the same as the range from d2 to d1. The following observation is illustrated in Figure 13.7. Observation 13.7 Let P1 and P2 be convex polygons with disjoint interiors. Let d1 and d2 be directions in which P1 is more extreme than P2. Then either P1 is more extreme than P2 in all directions in the range from d1 to d2, or it is more extreme in all directions in the range from d2 to d1. We are now ready to prove that Minkowski sums are pseudodiscs. 293 Chapter 13 ROBOT MOTION PLANNING Theorem 13.8 Let P1 and P2 be two convex polygons with disjoint interiors, and let R be another convex polygon" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure3.35-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure3.35-1.png", + "caption": "Fig. 3.35 Displacement in screw-plate simulation", + "texts": [ + "02395 an angular fracture, since at this point, it is desired to have the least possible tension in the plate to achieve a more efficient treatment. On the other hand, it can be observed that the displacement is smaller in the conventional plate, but, likewise it could be observed that when comparing the area where the displacement is exerted, it is much larger on the conventional plate than on the lambda plate, which tells us that the conventional plate tends to have a greater displacement area compared to the lambda plate (Fig. 3.35). For the jaw-plate-screw system simulations, the forces exerted by the jaw muscles reported in the mandibular kinematics section were placed similarly to those presented in Fig. 3.36. Simulation 1: static jaw simulation with lambda-type plate under normal mandibular forces This simulation suggests that the tension distributed in the arms of the plate as the displacement is less. It can also be observed that the compression stresses are much larger than those presented on the plate (Figs. 3.37 and 3" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002780_amm.121-126.3087-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002780_amm.121-126.3087-Figure2-1.png", + "caption": "Fig. 2 Schematic of air journal bearing Fig. 3 Cross section of air bearingGoverning Equation", + "texts": [ + " A reservoir of compressed air is therefore required during start-up. So, it\u2019s not suitable for miniature gas turbine engine. Aerodynamic bearings generate the required bearing forces through the relative shearing movement between the rotor and bearing surfaces. Thus no extra energy supply is required (although energy is lost through viscous friction as for all air bearings). So, aerodynamic air journal bearings are the most promising choice for meeting the stringent bearing requirements of micro gas turbine engine. Fig. 2 shows the aerodynamic air bearing. The cross section of the air journal bearing is shown in Fig. 3. The rotational shaft is in the center, with the diameter of 1mm and 2mm in length. The air bearing is in the outside, which is wedge-shaped. The air is compressed due to the smaller clearance, then air pressure is increased. The bearing force is generated through the relative shearing movement between the rotor and the wedge-shaped bearing surfaces in the role of the air viscosity. Some governing equations should be used to solve the air status in the air journal bearing simulation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003999_s12283-014-0158-y-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003999_s12283-014-0158-y-Figure6-1.png", + "caption": "Fig. 6 Curved path of object ball O", + "texts": [ + " In mathematical terms, at the termination of impact, when n is equal to Nf, the final step of the numerical algorithm, this condition for curving is created when U 0 Nf \u00bc bNf \u00f016\u00de Here bNf denotes the direction of its velocity of the center of gravity of ball O (i.e.,), given by tan bNf \u00bc _yO G Nf _xO G Nf \u00f017\u00de and U0 being the direction of slip on the table. The subsequent curved path of the ball can be shown to describe a parabola, conveniently expressed in the X0Y0 coordinate system, which is rotated from XY axes by w (see Fig. 6). Where tan w \u00bc 1 tan U0Nf \u00f018\u00de U0Nf is obtained from the numerical algorithm explained in Sect. 2.4.3. Expressions for XV 0 and YV 0 have been derived in the detailed analysis of Hopkins and Patterson [27] on the curved path of a bowling ball. The sliding can be shown to stop at time Ts, given by, TS \u00bc 2s 0 Nf 7ls \u00f019\u00de when the ball is at S (see Fig. 6) or the coordinate (Xs 0Ys 0) in the X0Y0 system (where s 0 Nf is the slip velocity of O, on the table, at the termination of impact, obtained at the final iteration of the numerical scheme). Using the derivations of Hopkins and Patterson [27], and also using Eq. (18), it is possible to develop the expressions for Xs 0 and Ys 0, together with the velocity components _X0S and _Y 0S at the end of the slipping process. The final velocity at S, VS \u00bc _X0S 2\u00fe _Y 0S 2 \u00f020a\u00de At an angle of hS with respect to the XY coordinates, given by, hS \u00bc w\u00fe k \u00bc w\u00fe tan 1 _Y 0S _X0S " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001077_978-1-4419-7979-7_6-Figure6.23-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001077_978-1-4419-7979-7_6-Figure6.23-1.png", + "caption": "Fig. 6.23 (continued)", + "texts": [ + " If one takes a microscope one can discern a large number of domains within the structure of ferromagnetic materials. These are regions in which the dipole magnetic moments (see Fig. 6.22a with magnetic moment \u00bc qoor 2=2) of all the atoms within one domain are parallel, giving rise to a net magnetic moment for that domain. In a sample of 6.3 Magnetic Properties of Materials 243 ferromagnetic material which has not been exposed yet to any magnetic field (iron fresh from foundry), the domain magnetic moments are randomly oriented (see Figs. 6.22b [6], and 6.22c for H\u00bc 0, point 1 of Fig. 6.23a) and the net magnetic flux in the material is zero. As a consequence the operating point on the flux/flux density-current/magnetic field strength function of Fig. 6.23a is in the origin. In the followingwe perform a set of \u201cexperiments\u201d, where we increase the external magnetizing current i (or magnetic field strength H) of a coil wrapped around a ferromagnetic material in a stepwise (see points 1\u20135 in Fig. 6.23a) manner. Let us assume that at current isat the ferromagneticmaterial starts to saturate, that is, the linear B\u2013H characteristic becomes nonlinear with currents above values larger than isat. In general, when an external magnetizing force (current) is applied to ferromagnetic material the domain magnetic moments tend to align with the externally applied field H. As a result, the dipole magnetic moments of the domains add to the applied field H, resulting in amuch larger value of the flux density Bwithin the ferromagneticmaterial 244 6 Magnetic Circuits: Inductors and Permanent Magnets than would exist due to the magnetizing current (force) alone", + " Thus the effective permeability m, equal to the ratio of the total magnetic flux density B to the applied magnetizing force (magnetic-field intensity) H, that is (B/H)\u00bc mr mo, is large compared with the permeability of free space mo. This behavior continues until all the magnetic moments are aligned with the externally applied magnetic field H; at this point the all-aligned magnetic moments cannot further contribute to increasing the magnetic-flux density Bmax, and the material is said to be fully saturated. In particular if we use newly cast iron (fresh from the foundry) and measure its magnetic field H or B then we find that all magnetic dipole moments are randomly distributed and we have H\u00bc 0 or B\u00bc 0 at point 1 of Fig. 6.23a. If we increase the magnetizing current to i\u00bc isat of the coil wrapped around the ferromagnetic material then a magnetic flux density of Bsat (see Fig. 6.23a) results and we are at point 2 where the magnetic dipole moments are partially aligned with the external magnetic field Hsat. A quadrupling of the magnetizing current to 4isat leads to a relatively small increase of the flux density Bmax due to the almost total alignment of the magnetic moments of all domains with the externally applied field 4Hsat (see point 3 of Fig. 6.23a). In the absence of an externally applied magnetic intensity field H, the domain magnetic moments naturally align along certain directions associated with the crystal structure of the domains, known as axes of easy magnetization. These axes are not identical with those of the randomly oriented (Fig. 6.22c, point 1) domain magnetic moments. Thus if the applied magnetizing field is now reduced from 4Hsat to Hsat, the domain magnetic moments start to relax to the directions of easy magnetization nearest to that of the applied field. As a result, when the applied field is reduced to Hsat the magnetic dipole moments will no longer be aligned with the externally applied field, and a B larger than Bsat results. At H\u00bc 0, after having been exposed to H 6\u00bc 0, the dipole magnetic moments will no longer be 246 6 Magnetic Circuits: Inductors and Permanent Magnets totally random in their orientation; they will retain a net magnetization component along the applied-field direction (see Fig. 6.23a, point 5), and the residual flux density Bresidual is measured. It is this effect which is responsible for the phenomenon known as magnetic hysteresis. The relationship between B and H for a ferromagnetic material is both nonlinear and multi-valued due to the major loops of Figs. 6.23b, c [7]. The measured functions of Fig. 6.23b show that the enclosed area by the (B\u2013H) characteristic is frequency-dependent: the higher the frequency the larger is the area due to the generation of eddy currents within the iron core. Figure 6.23d illustrates the occurrence of major and minor loops. The intercept of the B\u2013H loop with the vertical axis is called residual flux density Bresidual and the intercept with the horizontal axis is called the coercivity Hc. Demagnetization of Ferromagnetic Material: To demagnetize ferromagnetic material either the flux density B or the magnetic field intensity H must be measured. B can be measured via a Hall sensor as depicted in Fig. 6.23e where the Hall voltage is VH \u00bc RH I B s . The coefficient RH is depending on the semiconductor material used. For indium/antimony (InSb) this constant is RH\u00bc 240 cm3/ (As). The current I is a calibration current, the thickness s is a design constant of the device and the flux density B is the quantity to be measured. Hall sensors have an error of about 1% and are not as accurate as compared with low inductive shunts for measuring currents i which are proportional to H and B\u00bc moH in air or free space. The demagnetization of current probes (see Fig. 6.23f) employing Hall sensors can be accomplished in several ways. Two methods highlight the procedure of demagnetization of current probes. This demagnetization of the iron core of a current probe is necessary in order to minimize the measuring error or such probes. Figure 6.23f depicts the iron core with demagnetization winding, Hall sensor, and conductor whose current i(t) must be measured via the flux density B developed in the iron core. Method #1 of demagnetization relies on the decrease of the amplitude of the demagnetization current (Fig. 6.23g) resulting in the decrease of the major B\u2013H loops as illustrated in Fig. 6.23h. Method #2 of demagnetization relies on the controlled decrease of the amplitude of the demagnetization current (Fig. 6.23i) resulting in the employment of major and minor (B\u2013H) characteristics as illustrated in Fig. 6.23j. A current probe is used to measure large currents of any waveshape and frequency (e.g., 100 A, triangular, 1,000 Hz). It consists of an iron core and an InSb Hall sensor (see Fig. 6.24). What is the Hall voltage VHall for B\u00bc 0.1 T, s\u00bc 1 mm, IHall\u00bc 1.0 A and RHall \u00bc 240 cm3/(As)? The Hall voltage is VHall \u00bc RHall IHall B s \u00bc 240 10 6 1 0:1 10 3 \u00bc 24mV. 6.3 Magnetic Properties of Materials 247 Exciting Current at AC Excitation: In AC power systems the waveforms of voltage and flux (created within inductors and transformers by the applied terminal voltage) closely approximate sinusoidal functions of time" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002508_ram.2013.6758569-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002508_ram.2013.6758569-Figure1-1.png", + "caption": "Fig. 1. Mobile robot with nonholonomic constraints", + "texts": [ + " Section II introduces the robot model, trajectory representation and reformulated trajectory generation problems. In section III, kinodynamic constraints and collision avoidance criterion are analytically considered. In section IV, two optimal indexes are developed and suboptimal solutions are given to deal with kinodynamic constraints, collision avoidance and optimization problem on the proposed parameter space. Simulation results are shown in section V. In section VI, conclusions are drawn. II. PROBLEM FORMULATION The mobile robot model with nonholonomic constraints is shown in Fig.1. The guidepoint(GP) is set to be the midpoint of robot\u2019s rear wheel-axle. In this paper, the mobile robot can be represented by a simple particle covered by its circumcircle. The robot kinematic model can be described as follows.\u23a1 \u23a2\u23a2\u23a2\u23a2\u23a3 x\u0307 y\u0307 \u03b8\u0307 \u03d5\u0307 \u23a4 \u23a5\u23a5\u23a5\u23a5\u23a6= \u23a1 \u23a2\u23a2\u23a2\u23a2\u23a3 cos\u03b8 sin\u03b8 tan\u03d5/l 0 \u23a4 \u23a5\u23a5\u23a5\u23a5\u23a6u1 + \u23a1 \u23a2\u23a2\u23a2\u23a2\u23a3 0 0 0 1 \u23a4 \u23a5\u23a5\u23a5\u23a5\u23a6u2 (1) where q = [x,y,\u03b8 ,\u03d5]T is the system state and (x,y) represents the Cartesian coordinates of the middle point of the rear wheel axle. \u03b8 is the orientation of the robot body with respect to the X-axis and \u03d5 is the steering angle" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002036_978-94-007-2069-5_11-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002036_978-94-007-2069-5_11-Figure1-1.png", + "caption": "Fig. 1 (a) Anti-backlash gear assembly, (b) prototype design of the backlash-free gearbox", + "texts": [ + " 2011 87 lower torques than the positioning drives require. Therefore, between the electric motor and positioning device is necessary to insert the appropriate transmission or gearbox. Gearbox, as well as any kinematic mechanism, is necessarily made with clearances and dimensional tolerances. Still current problem is the suppressing of these backlashes in reversing the shaft rotate direction. This article describes one way of suppressing the backlash in gears meshes and bearings of gearbox with spur gears and the load simulation of the gearbox. Figure 1a schematically illustrates the principle of backlash-free gearing with countershafts. Torsion bar spring is mounted at a determined preload torque, and thus act on meshing pinion wheels in opposite directions. When transmissing a rotary motion, then under the direction of rotation of input shaft performance by either one or the other path of gearing [1]. Similar methods of backslash elimination are described by [2] and [3]. The gear by the Fig. 1a is closed loop of gears. Therefore, it is necessary in design to comply with geometric and kinematic constraints, which relate to axis distances and gear circumferential speeds. In general case (Fig. 2) must be satisfied condition for pitch diameters d32 d31 D d22 d12 d11 d21 : (1) Assuming the same module of all the gears, pitch diameters can be replaced by the number of teeth z32 z31 D z22 z12 z11 z21 : (2) The advantage of this type of gearing is high variability in dimensions and therefore large area for optimization of specific applications. For the prototype design and subsequent calculations was chosen case, when d11 D d12 D d31 D d32 D d1 and d21 D d22 D d4 D d2 (Fig. 2). Table 1 lists technical parameters of the gearbox, which are used for subsequent calculations. The gearing on Fig. 1a is a combined six-member mechanism (including frame). On Fig. 3 is release of each member. It should be determine 38 unknown forces and torques, torques Mtk (torsion-bar spring preload) a M2 (load on output shaft). are parameters of equation system. There are 30 equilibrium equations, the remaining eight equations resulting from the geometry of toothing (four meshes, for each one two equations). For the dynamic analysis was created model by Fig. 4. The gear unit is connected by a ball screw with a weight, which is constrained by a ball linear guide" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003286_imece2013-64128-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003286_imece2013-64128-Figure3-1.png", + "caption": "Figure 3. Pad balance schematic diagram", + "texts": [ + "org/about-asme/terms-of-use The state variable could be defined as following: [ ]4 8 9 12 T Z Z Z Z Z= \uff0c 4 8 9 12 T Z Z Z Z Z = \u027a \u027a \u027a \u027a \u027a , where , , , , , , , T i i i xi xi i i yi yiZ x x y y\u03b8 \u03b8 \u03b8 \u03b8 = \u027a \u027a\u027a \u027a ( )4,8,9,12i = , , , , , , , , T i i i xi xi i i yi yiZ x x y y\u03b8 \u03b8 \u03b8 \u03b8 = \u027a \u027a\u027a \u027a \u027a\u027a\u027a \u027a \u027a\u027a \u027a \u027a\u027a ( )4,8,9,12i = . A traditional way of obtaining the pad velocity is finding a suitable velocity to make the 0 ti f = . Taking the pad's moment of inertia into consideration, the pad of the bearing could be regarded as one of the moving parts, and the pad motion equation needs to be coupled with the rotor dynamics equation. Figure 3 is pad balance schematic diagram. The pad angular acceleration could be calculated by Equation (4). i tiI f s\u03b1 =\u027a\u027a (4) The swing pad's angular velocity and displacement are obtained by integrating the angular acceleration once and twice respectively. Since the number of system dynamics equations is increased, the time spent on numerical integration is longer, but the time devoted to obtain the pad swing speed is saved. As mentioned before, the coordinate system used in numerical integration is attached to the center of the journal" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002910_icems.2011.6073459-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002910_icems.2011.6073459-Figure7-1.png", + "caption": "Fig. 7 shows the 3D thermal distributions by FE calculation. Figure 8 shows the thermal pictures of the motor by measurement method.", + "texts": [], + "surrounding_texts": [ + "The circuit in figure 3 applies to both the positive and negative sequence networks. The only difference between the two is the value of the \u201cload resistance\u201d RL as defined as:\n1- *\ni\ni i r\ni\nS RL R\nS (4)\nWhere\nPositive Sequence Slip: 1\n-\ns r\ns\nn n S\nn \nNegative Sequence Slip: 2 1 2 - S S\nWhere sn is the synchronous speed and rn is the rotor speed.\nNote that the negative sequence load resistance 2RL will be a negative value, which will lead to a negative shaft power in the negative sequence.\nIf the value of positive sequence slip ( 1S ) is known, then\nthe input sequence impedances for the positive and negative Sequence networks can be determined as\n * /\n( ) i i i i i\ni i i i i i i\nRc jXm Rc jXm Rr RL jXr ZM Rs jXs\nRr RL j Xm Xr\n \n (5)\nWhere i = 1 for positive sequence and i = 2 for negative sequence.\nOnce the input sequence impedances have been determined, the analysis of an induction machine operating with unbalance voltages requires the following steps.\nStep 1: Transform the known line-to-line voltages to sequence line-to-line voltages:\n0\n1\n2\nVan\nVan\nVan \n2\n2\n1 1 1\n1 / 3 1\n1\na a\na a\n \nVan\nVbn\nVcn \n(6)\nWhere a = 01 120 and Van is angle referent (0) .\nAlso 0Van and 0 Ia will be zero, due to the prototype motor is\nconnected as star isolated neutral pattern.\nStep 2: Compute the sequence line currents into the machine:\n0 0Ia (7)\n1 1\n1\nVan Ia\nZM (8)\n2 2\n2\nVan Ia\nZM (9)\nStep 3: Transform the sequence currents to phase currents:\nIa\nIb\nIc \n2\n2\n1 1 1\n1\n1\na a\na a\n \n0\n1\n2\nIa\nIa\nIa \n(10)\nFurthermore, the prototype motor is modeled by the finite element method, which is ANSIS. First of all, the sequence line current is calculated by the equation (10) as shown in the figure 4. Also figure 5 shows the measured sequence line current which is adjured by the voltage regulator. Although there are attempted for setting as the same values, the motor and load impedances is related for the error.\nV. THERMAL DISTRIBUTIONS\nIn fact, heat transfer can be reffered as:\n2 1. . T T Q k A\nL\n (11)\nThe equation (11) shows the ration of heat which is connected to the different temperatures between T1 and T2. Also the heat generation, (Q : 3watts/m ), of motor is sourced\nby the winding currents.", + "It can be denoted as:\n2\n3\ni R Q\nm (12)\nWhen i is phase current (A), R is winding resistance ( ), 3m is winding volume\nA. Balancing Voltage Condition\nFor better vision, the finite element method is shown as the thermal effect by varying unbalance voltage conditions. Figure 6 shows the 2D cross section of FE result of the motor, which is varied with 0% load and balancing three voltage phases. It can be seen that the stator winding coil is the main heat sources.\nFigure 9 shows the comparison of the thermal prototype motor which is depended by the percent of load. Table II\nshows the comparison of the thermal prototype motor as varying percent of load from 0 to 100.\nMaximum Temperature ( C )\nTest FEM Error\nLoad 0 % 47.9 47.375 1.096\nLoad 25 % 48.9 49.379 0.979\nLoad 50 % 53.1 54.831 3.259\nLoad 75 % 61.0 63.388 3.914\nLoad 100 % 73.6 74.716 1.516\nB. Unbalancing Voltage Conditions\nAccording to the percent of the both definitions [4-5], the percent of unbalancing voltage conditions is varied as 1 to 5 percent with both under and over voltage conditions are studied. There are the single under voltage, the double under voltage, the single over voltage and the double over voltage conditions. Figure 10 show the comparison of the thermal prototype motor which is depended by unbalancing voltage conditions.\n0 1 2 3 4 5 45\n46\n47\n48\n49\n50\n51\nPercent Unbalanc Voltage (%)\nM ax\nT em\npe ra\ntu re\n(C el\nsi us\n)\nUnder 1 Phase Under 2 Phase Over 1 Phase Over 2 Phase\n(a)\n0 1 2 3 4 5 70\n72\n74\n76\n78\n80\n82\nPercent Unbalance Voltage (%)\nM ax\nT em\npe ra\ntu re\n(C el\nsi us\n)\nUnder 1 Phase Under 2 Phase Over 1 Phase Over 2 Phase\n(b) Fig.10. Maximum temperature of motors\n(a) no load (b) full load", + "Table III shows the comparison of the thermal prototype motor as varying the unbalancing voltage conditions with 1 to 5 percent. It can be concluded that when the induction motor is supplied as unbalancing voltage with under voltage condition the temperature of the motor is lower that the induction motor is supplied as balancing voltage. Also the induction motor is supplied as unbalancing voltage with over voltage condition the temperature of the motor is higher that the induction motor is supplied as balancing voltage.\nThe temperatures of the motor when are supplied with over voltage conditions are higher than the temperatures of the motor when are supplied with under voltage conditions. The FEM is again the effective tool for calculating the motor performance.\nVII. ACKNOWLEDGMENT\nThe authors would like to thank faulty of engineering, Naresuan University for supporting the finance for research, materials and equipments, advising staffs in this project\nVIII. REFERENCE [1] W. H. Kersting and W. H. Phillips, \"Phase frame analysis of the effects\nof voltage unbalance on induction machines,\" IEEE Transactions on Industry Applications, Vol. 33, Issue 2, 1997, Page(s): 415-420.\n[2] Ching-Yin Lee, Bin-Kwie Chen, Wei-Jen Lee, and Yen-Feng Hsu, \"Effects of various unbalanced voltages on the operation performance of an induction motor under the same voltage unbalance factor condition,\" The IEEE Technical Conference of Industrial and Commercial Power Systems, 11-16 May 1997, Philadelphia, PA , USA, Page(s): 51-59. [3] A. Siddique, G.S. Yadava, and B. Singh, \"Effects of voltage unbalance on induction motors,\" IEEE International Symposium on in Electrical Insulation, 19-22 Sept. 2004, Page(s): 26-29. [4] Motors and Generators, ANSI/NEMA Standard MG1-1993. [5] IEEE Standard Test Procedure for Polyphase Induction Motorsand Generators, IEEE Standard 112, 1991. [6] R.C. Dugan, M.F. McGranaghan, and H.W. Beaty, ElectricalPower\nSystems Quality. New York: McGraw-Hill, 1996." + ] + }, + { + "image_filename": "designv11_84_0003058_cdc.2014.7039437-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003058_cdc.2014.7039437-Figure1-1.png", + "caption": "Fig. 1. Estimate of the domain of attraction of direct I&I adaptive control system (13) defined by (16).", + "texts": [ + " Hence, for any 0 < \u03f5 < sin \u03b8, and for all |\u03b8\u0303(0)| \u2264 sin \u03b8 \u2212 \u03f5, (15) we have that |\u03b8\u0303(t)| \u2264 e\u2212\u03b3\u03f5t|\u03b8\u0303(0)|. It is important to note that, although increasing the adaptation gain \u03b3 increases the rate of convergence of the parameter error, it reduces the domain of attraction of the zero equilibrium. Indeed, recalling that \u03b8\u0302 = \u03b8I + \u03b8P (x) the condition (15) translates into \u03f5\u2212sin \u03b8+\u03b3 arctanx(0) \u2264 \u03b8I(0)\u2212\u03b8 \u2264 sin \u03b8\u2212\u03f5+\u03b3 arctanx(0). (16) Consequently, increasing \u03b3 reduces the admissible range for the initial conditions of x. See Fig. 1. The existence of this trade\u2013off between transient performance and robustness if, of\u2013course, expected in any feedback system. Simulations of the direct I&I adaptive control system (13) were carried out with \u03b3 = 1, \u03b8 = 2 and different initial conditions. The results of the simulation are shown in Fig. 2. Remark 4: The trajectories predicted by Corollary 1 correspond only to the ones converging to zero in the figures, i.e., those with initial conditions verifying (15). Simulations show, however, that all trajectories converge to an equilibrium point" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002336_icsd.2013.6619912-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002336_icsd.2013.6619912-Figure7-1.png", + "caption": "Fig. 7. Electric field distribution on the surface of the 1st section jumper in phase A", + "texts": [ + " 6 shows the electric field distribution curve along the surface of each subconductor. It can be seen that the overall distribution pattern of all subconductor are basically the same and the maximum electric field strength were located in similar place. Owing to the shielding effect of conductors, the electric field distribution curve is low both at the beginning and the end. In the middle of the curve, with the decrease of the shielding effect of conductors and curvature radius of jumper sag, the electric field strength is high. Fig.7 shows the electric field distribution on the surface of jumper. The maximum electric field is appeared in the middle of the subconductor 6 where the curvature radius of jumper sag is small. That is just corresponding to the electric field distribution curve. The maximum electric field strength is 2317 V/mm. In 1000 kV UHV AC project of China, the maximum electric field strength on the surface of conductor and fittings should be kept below 2300 V/mm to 2500 V/mm. Table 2 lists the maximum electric field strength on the surface of jumper in three phases" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002570_ever.2013.6521586-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002570_ever.2013.6521586-Figure3-1.png", + "caption": "Fig. 3. Distribution of magnetic induction B at 3 broken rotor bars.", + "texts": [ + " Presentation of this effect points the most common localization of the rotor bar fault and explains why the adjacent bars are usually damaged. Deformation of the electromagnetic field resulting from the asymmetry is visible in the distribution of the magnetic vector potential (Fig. 2). The penetration of the flux to the damaged bars is clearly visible. In this case the mutual inductance is decreased while the leakage inductance is increased. Distribution of magnetic induction corresponding to the same time of the simulation is presented in Figure 3. After calculation of a series of FEM simulations three broken rotor bars were selected, since the electromagnetic field deformation in this case is visible globally. Field analysis provides studying the effects of the damaged rotor bars to the equivalent circuit parameters of the machine. Experimental studies are subject to uncertainties due to nonstationary model of the machine. III. NONINVASIVE ROTOR FAULT DIAGNOSTIC METHOD Analyzing reported diagnostic methods limitations an idea of the pulsed-field generation was considered" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003549_00207721.2013.775389-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003549_00207721.2013.775389-Figure1-1.png", + "caption": "Figure 1. Lockheed Martin\u2019s CEV concept.", + "texts": [ + " Keywords: aerospace control; biology; artificial intelligence; networked control The crew exploration vehicle (CEV) is NASA\u2019s proposed series of human spaceflight vehicle, one of the most important elements of Project Constellation. It is intended to supersede the space shuttle system. It will be capable of carrying astronauts beyond the low Earth orbit, landing astronauts on the Moon, and ultimately allowing human missions to the surface of the Mars and certain near-Earth asteroids. The Apollo-like capsule vehicle, as conceptually shown in Figure 1, represents one of the typical designs for the future CEV. The new CEV design will use a complicated crew and service module design principle, instead of a space plane-style lifting body used in the space shuttle system. Reliable and effective operation of such a complex vehicle system calls for technique breakthrough in various areas, one of which, apparently, is the vehicle attitude control. The past few years have witnessed some progress more or less related to this topic. For instance, linear attitude control using gyro torques has been considered in Greensite (1970), Iyer and Singh (1989) and De and Iyer (1989)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003456_ijtc2011-61259-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003456_ijtc2011-61259-Figure2-1.png", + "caption": "Fig. 2 Velocity vectors at the contact.", + "texts": [ + " Thin film colorimetric interferometry (TFCI) is used to obtain film thickness distribution from chromatic interferograms. Experiments were carried out using an experimental apparatus consisting of a ball-on-disk tribometer equipped with a microscope imaging system and control unit [9]. In the tribometer (Fig. 1) a circular EHD contact is formed between a steel ball and chromium coated glass disk. The ball axis can be set ranging from 0 up to 45 degrees and both the ball and disk can be independently driven by servomotors, thus different contact conditions can be simulated, as shown in Fig. 2. At the contact, surface velocity vectors u1 and u2 act at different directions with \u03b4 angle between them. These vectors form an entrainment ue and a sliding velocity us, which are given by equations 1 and 2 respectively, with \u03b5 angle between them. TFCI technique provides film thickness distribution from the digitized chromatic interferograms. Two series of experiments were carried out in this study. The first one dealt with the influence of size of sliding velocity vector perpendicular to the entrainment velocity direction" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003352_2014-01-1700-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003352_2014-01-1700-Figure10-1.png", + "caption": "Figure 10. 3D image of the dual park phaser", + "texts": [ + " The second lock pin locks the rotor assembly to the housing assembly when the phaser is at a full retard position. What makes this second lock pin unique, is that it utilizes pressure to engage the lock pin into the housing assembly rather than pressure to disengage as with the first lock pin, the latter being the more conventional approach. This method of lock pin engagement is ideal for stop/start systems as the duration of time in autostop mode is defined for a finite period of time. With the CTA dual park phaser as shown in Figure 10, the default intermediate lock pin is achieved simply by removing the current or duty cycle to the variable force solenoid. This allows the spring loaded spool, which is integrated within the phaser assembly to open the detent port and move the phaser back to its intermediate position by cam torque energy. By default, therefore, the phaser will also be parked in a safe position for a standard engine start. During an autostop event, the cam phaser will be commanded to the full retard position and through the spool geometry oil will be directed towards the second lock pin to engage it with the housing at the full retard position" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure4.11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure4.11-1.png", + "caption": "Fig. 4.11. The reachable workspace of the arm exoskeleton", + "texts": [ + " 4 Exoskeletons as Man-Machine Interface Systems 69 L-EXOS is characterized by a serial kinematics consisting of five rotational joints (Fig. 4.10) of which the first four are actuated and sensorized, while the last one is only sensorized. The first three rotational axes are incident and mutually orthogonal (two by two) in order to emulate the kinematics of a spherical joint with the same center of rotation of the human shoulder. The target workspace for the shoulder joint was assumed to be the quadrant of a sphere, as shown in Fig. 4.11. The orientation of the first axis has been optimized to maximize the workspace of the shoulder joint, by avoiding singularities and interferences between the mechanical links and the operator. The optimization process provided also indications for the definition of the shapes of the links. As a result from the kinematic analysis, the final orientation of the first axis (the fixed one) was chosen to be skewed with respect to the horizontal and vertical plane, while the third axis was assumed to be coincident with the ideal axis of the upper arm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003523_s10948-014-2678-x-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003523_s10948-014-2678-x-Figure2-1.png", + "caption": "Fig. 2 The geometry: its inner radius is rin; its outer radius is rout; its height is h; its magnetic polarization is axial, and the north pole is on its upper face with a surface magnetic pole density \u03c3 \u2217", + "texts": [ + "1 Geometry and Notation The magnetic field created by an axially magnetized permanent-magnet ring can be calculated analytically using the Coulombian model [5]. The z axis is the axis of symmetry of the ring magnet. Its inner radius is rin; its outer radius is rout; its height is h; its magnetic polarization is axial, and the north pole is on its upper face with a surface magnetic pole density \u03c3 \u2217. The magnetic field is expressed in three magnetic field components along the three defined axes: H\u03b8 (r, z), Hr (r, z), and Hz (r, z). The geometry, coordinate system, and some related notation appear in Fig. 2. 2.2 Analytical Calculation The Azimuthal Component The azimuthal component H\u03b8 (r, z) equals 0 on account of the cylindrical symmetry [6]. H\u03b8 (r, z) = 0 (1) The Radial Component The expression of the radial component Hr (r, z) presented here is established by Wang et al. [7]. The magnet\u2019s analytical expression was built using scalar magnetic potential method, generalized binomial theorem, and vector superposition theorem (Fig. 3). An auxiliary function is established first. HR,h r = \u03c3 \u2217R 2\u03bc0 \u221a r2+R2+(z\u2212h)2 \u00d7 (2) \u221e\u2211 i=0 [ 1 2 2i + 1 ] ( k2 R,h )2i+1 (2i+2)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002047_kem.462-463.1044-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002047_kem.462-463.1044-Figure7-1.png", + "caption": "Fig. 7 The three-dimensional Fig. 8 The deflectional deformation", + "texts": [ + "52 mm as shown in the load-deflection curve of figure 2. When the damage occurred on the femur, the maximum deflection was 0.631mm, femur modulus of elasticity was 1380MPa, poisson ratio was 0.3, and the maximum bending normal stress was 69.9Mpa. The finite element analysis was also based on the linear elastic, small deformation and isotropy CT scan of the chosen sample was taken. The obtained CT images were processed by using Mimics software. The bone boundary was extracted in order to establish three-dimensional model (figure 7). Ansys Finite Flement was used to divide mesh on the bone model and the Ansys Finite Flement was used for calculation. We can see from figure 8 that the maximum deflection of the femur sample occurred in the area of exerted load was given. The maximum deflection was 0.603mm, which is almost correspondent to the deflection result 0.631 mm in the experimental. Figure 9 shows that the maximum stress occurred at the back of the femur that bears the load. The maximum stress value was 53.3 Mpa. This maximum stress is also the first main stress given to the bone which is consistent with the area of damage on the bone" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003160_12.2013870-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003160_12.2013870-Figure2-1.png", + "caption": "Figure 2. Schematic of a turbine blade subjected to fatigue loading", + "texts": [ + " The crack growth rate per loading cycle and acoustic emission per loading cycle in a material can be related to the cyclic range of the stress intensity factor as KC dn da a (1) KC dn dN N (2) Here a is the crack length,n is the loading of the structure, N is the total number of acoustic emission events in the structure due to crack growth, C , CN, \u03b1 are constants and are dependent on the material, \u0394Kis the cyclic range of the stress intensity factor. The expressions for stress intensity factor that relates crack size to the applied loading can be found in literature [10]. For a single edge notched specimen as in Figure 2, the stress intensity factor equation at the crack tip (assuming zero tip radius) is 432 382.3071.2155.10231.0122.1 W a W a W a W a aK (3) Here K is the stress intensity factor, \u03c3is the applied loading on the specimen, Wis the width of the specimen. Knowing the fracture toughness of a material and an initial crack length the crack propagation and expected acoustic emission events can be calculated for a given fatigue loading. Proc. of SPIE Vol. 8690 86900B-3 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 07/18/2016 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx The crack growth due to fatigue loading and associated acoustic emission can be studied in a stator blade sample to obtain insights into the AE activity in the material. The blades in the turbine are made from stainless steel while the turbine casing is made of low carbon non magnetic steel. A sample was prepared from a blade cut from the turbine as in Figure 2 and loaded into an Instron tensile machine. A single notch of 0.5 inches was made at the center of the sample to simulate an existing crack. The fracture toughness of the material is assumed to be around 80 ksi inch . which is a typical value for stainless steel, a stress of around 33.8 ksi is applied and cycled at a strain rate of 0.5 inch/min on the blade sample to let the crack grow. The acoustic emission activity in the material is recorded using two wideband acoustic emission sensors, WD, from Mistras Group Inc" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure3.46-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure3.46-1.png", + "caption": "Fig. 3.46 Application and transmission of loads", + "texts": [ + " The printed jaw is painted with white spray paint, and later mottled using a black sharpie (Fig. 3.44). Once the jaw is mottled, it is mounted on the clamping set (Fig. 3.45). 114 J. A. Beltr\u00e1n-Fern\u00e1ndez et al. With the jaw mounted with its clamps, weight is added to the rectangular profile of the clamp. In this specific case, 3D printing filaments were used, each with a different weight, which was noted before performing the test. Six coils of filament were used, which weigh 200, 400, 550, 550, 650, and 1200 g respectively, which, at the end, will provide a total load of 3.5 kg (Fig. 3.46). The test is recorded in real time with a camera. At the end of the trial, the video stops and it is uploaded to a computer. 3 Biomechanical Evaluation of Sharped Fractures \u2026 115 Fig. 3.47 Polygon and section view in the manipulation of GOM correlate In the GOM correlate interface, the video of the piece tested is imported, and it is subsequently verified that the program properly detects the mottled pattern on the piece. Once the video has been successfully imported, a calibration procedure is performed, so that the program can accurately identify the travel distances" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000869_s00170-021-07749-1-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000869_s00170-021-07749-1-Figure1-1.png", + "caption": "Fig. 1 Cross section of the pipe bend including its suspension and cross-sectional view A-A", + "texts": [ + " Several critical places need to be defined on the bent part. The first such place is the outer side of the pipe bend, where the wall is thinned due to stretching. On the inside part, on the other hand, the material is compacted due to compression, and at this place the wall corrugates. Another critical area is the suspension of the extended ends after bending, this process being caused by elastic deformation. During bending, the ovality (flattening) of the pipe is further formed, which is shown in Fig. 1 [23, 24]. Before defining critical points, it is important to find out the technological criteria, such as the relative bending radius, the relative strength of the walls (thick-walled, thin-walled) and the degree of difficulty of the bending process. The determination of the bending radius is influenced by plasticity, wall thickness or the forming method. On the inside part, the thickness increases due to compression, and on the outside part, the wall thins due to stretching. After exceeding the tensile stress, which is equal to the tensile strength, the plasticity of the material is depleted and cracks are formed" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002448_icphm.2013.6621410-Figure105-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002448_icphm.2013.6621410-Figure105-1.png", + "caption": "Figure 105 Dynamic FEA model", + "texts": [], + "surrounding_texts": [ + "Fault severity assessment consists of three parts as shown in Error! Reference source not found.: Lumped parameter dynamic model, inverse model parameter estimation and the correlation with the crack size. The fundamental assumption is that the gear crack size may be inversely proportional to the mesh stiffness, suggesting that the stiffness is reduced by the increased crack size. Therefore, the accurate estimation of the stiffness is critical in this process. Lumped parameter model is used to analyze the dynamic response of the test-bed, which consists of the gear, shaft, coupling and bearing. Many literatures are available_on this subject (See for comprehensive review [20]). In this study, the model considers only the torsional degree of freedom ( B). Then the equations of motion are described by a set of parameters given in Figure 12 , where J, c, k denote the mass moment of inertia, damping and torsional stiffness, and the subscripts g, s, c represent each component - gear, shaft and coupling, respectively. The loading condition in the model are defined as the torque (T;n' 1',,111 ) of input and output shaft which are measured from the torque sensor. The gear meshing process is modeled in terms of spring and damper [21] as shown in Figure 13, in which the periodically varying stiffness is represented by square wave function [22] as shown in Figure 14. The model building and solving is carried out by DAFUL, a dynamic analysis commercial code [23]. The time history results are obtained for the TE under the given loading conditions, from which the simulated features are extracted by going through the same signal processing as stated before. In the normal gear, the stiffness variation is the same over the number of teeth as given in solid line in Figure 14. In case of cracked teeth, however, it will reduce the corresponding stiffness as in dotted line, which may result in the malfunction of the gearbox. In order to identify this, the stiffness is parameterized by its max and min values. Recall here that the fault detection process has already provided the faulted component location. Hence, in case of a single fault, which is the case in this study, we need only to determine the two max and min parameters (C, D). This is more effective in a case of wind turbine gearbox which requires the reduction of l\ufffd g.2 Figure 93 Gear mesh model with spring and damper estimated model parameter number. Inverse estimation procedure is applied to determine these by performing optimization with the objective to minimize the difference between the simulated and measured features in case of single fault. (A, B: Normal, C, D: Fault) 0.9 0.8 1,j :2 0.7 ro \ufffd 0.6 ()) 0.5 N Ow 0.4 .><: u 0.3 \ufffd () 0.2 0.1 D \ufffd .L'-\ufffd 0.2\ufffd\ufffd 0\ufffd .3-\ufffd0.4-\ufffd0\ufffd.5-\ufffd0.6\ufffd\ufffd0\ufffd.7-\ufffd0.8\ufffd\ufffd0.9 Mesh stiffness (CFaul/CNormal) Figure 86 Correlation of mesh stiffness (X axis) and crad size (Y axis) As a final step, the estimated model parameter should be correlated with the fault severity level. For this purpose, regression model is established by evaluating the mesh stiffness for each crack size. Several approaches for mesh stiffness calculation have been made for this ([24][26]). In this study, the max value (C) of the mesh stiffness parameters under crack condition are computed by carrying out the dynamic FEA that simulates the rotation of paired gears under uniform motion condition to remove the inertial force, using DAFUL. A snapshot of the DAFUL analysis is given in Model based diagnosis is a promising technique to estimate the absolute status of the fault by the help of underlying physics, which enables the prognosis that predicts the remaining useful life during the future operations. In this context, this study has proposed a method for gear fault of the gearbox under variable loading condition with the objective for wind turbine applications. While the overall concept is described here, the implementation is currently in progress and not completed yet. This will be presented in the upcoming work." + ] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure16-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure16-1.png", + "caption": "Figure 16 Applied chain reaction forces (and crankshaft torques) (see online version for colours)", + "texts": [ + " To drive the bike forward, a gear joint was created to transmit the torque from the engine to the centre of the rear wheel. The direction of the chain force was always set to be tangential to the top of the front and rear sprocket, which was the point at which the chain got into grip with the front sprocket and left the rear sprocket. This force was applied both at the front sprocket and the rear sprocket but in opposite directions, such that the opposing chain forces pointed towards each other. This force equalled the tension in the chain, which in turn led to the compression of the rear suspension. Figure 16 shows these chain forces. The rotation of the front and rear sprocket prevented the chain forces from being directly applied to these components. This problem was solved by applying the rear chain force to a triad attached to a super-node (triad) on the swing-arm at the same location as where the chain would leave the rear sprocket. At the front, the reaction chain force was applied to the engine housing on a super-node (triad). The purpose of the clutch is to disconnect the transmission between the crankshaft and the rear wheel", + " The engine control system also included blocks to determine the engine torque that should be delivered for specific bike or engine speed ranges. These blocks may act as rpm-limiters on real bikes or as speed limiters, such as those used when MotoGP bikes ride in the pit-lane. The chain force was calculated from the engine torque and the gear ratios (crankshaft vs. rear wheel). The control system that calculated the chain force also calculated the clutch contact ratio. That is, the chain force was reduced when the clutch skidded. The front and rear sprocket exert chain forces were parallel but acted in opposite directions (see Figure 16). Thus, the chain forces were in equilibrium and did not cause any translation of the entire bike (i.e., the bike was at equilibrium). However, the chain forces affected the movement of the swing-arm relative to the chassis. This movement affected the compression or extension of the rear suspension. The chain force control system is illustrated in Figure 20. The aerodynamic drag force is the most important factor affecting the behaviour of the bike. A lifting force from air flow also acts on the bike" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003233_amr.338.611-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003233_amr.338.611-Figure1-1.png", + "caption": "Fig. 1 Coordinate system of a single pad", + "texts": [ + " A corresponding computer program is developed base upon the theoretical model. Using the program, the effect of turbulence on journal equilibrium position, pads inclinations and complete dynamic coefficients is investigated for given load cases. All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP, www.ttp.net. (ID: 137.222.24.34, University of Bristol, Bristol, United Kingdom-21/03/15,00:54:09) Governing equation. Figure 1 shows the coordinate system of a single pad. In Fig. 1, Op\u03be\u03b7 is the local coordinate of a journal - pad, Op and Oj are geometric centers of the pad and journal respectively. The dimensionless oil film pressure distribution satisfies the following Reynolds equation: 3 2 3 z 1 1 ( ) ( )cos ( )sin 2 2 p D p h G h G L \u03b8 \u03be \u03b7 \u03b8 \u03b7 \u03be \u03b8 \u03b8 \u03b8 \u03bb \u03bb \u2202 \u2202 \u2202 \u2202 + = + + \u2212 \u2202 \u2202 \u2202 \u2202 (1) Where 1 cos sinh \u03be \u03b8 \u03b7 \u03b8= \u2212 \u2212 is dimensionless oil film thickness, p is dimensionless oil film pressure, ( )2z L\u03bb = is dimensionless axial coordinate, \u03be , \u03b7 , \u03be and \u03b7 are dimensionless displacement and velocity of journal respectively, G\u03b8 and zG are turbulence correlation coefficients, which can be written as: \u03b8 \u03b8 b \u03b8 h 1 12 G a R = + , z z b z h 1 12 G a R = + (2) Here, h eR R h= is local Reynolds number, Re=\u03c1CU/\u00b5 is the Reynolds number, \u03b8a , \u03b8b , za and zb are turbulence coefficients, which are defined by Ng and Pan [9]", + " (11), the Jacobian Matrix J1 can be expressed as: T T T T 1 T T T T 2 2 1 1 2 2 1 1 2 2 \u03b7 \u03be \u03b7\u03be \u03be \u03be \u03b7 \u03b7 \u03be \u03b7 \u03b7\u03be \u2202 \u2202 \u2202 \u2202 \u2202 \u2202 \u2212 \u2212 \u2212 \u2202 \u2202 \u2202\u2202 \u2202 \u2202 = \u2202 \u2202 \u2202 \u2202 \u2202 \u2202 \u2212 \u2212 \u2212 \u2202 \u2202 \u2202 \u2202 \u2202\u2202 p p K p p K f p f p J p p K p p K f p f p 1 1 (19) Because of the similar calculative property, Eqs. (7), (15), (16) can be solved simultaneously. Once the vector p and its partial derivatives with respect to\u03be ,\u03b7 , are obtained, the oil film forces and their Jacobian matrices can be directly calculated from Eqs. (9), (12) and (19), respectively. Coordinate transformations and superposition of subsystem. Considering moving relationship between pads and the shaft, as shown in Fig. 1, an exact coordinate transformations from the global coordinate to the moving coordinate is presented by Liping Wang, Tiesheng Zheng et al.[6,7], which can be written as: [ ]T 5 2 0m= +\u03b6 T q T (20) Where [ ]T\u03be \u03b7=\u03b6 , [ ]T x y \u03b5=q , [ ]5 3 4=T T T , 3 2 0=T T T , 4 2=T T e , 2 cos sin sin cos \u03b1 \u03b1 \u03b1 \u03b1 \u2212 = T , ( ) ( ) ( ) ( )0 cos sin sin cos \u03b3 \u03d5 \u03b3 \u03d5 \u03b3 \u03d5 \u03b3 \u03d5 + + = \u2212 + + T , [ ]T 0 1=e . Here, m is preload coefficient, \u03b3 is the pivot position angle, \u03d5 is the pad angle displacement. Similarly, the relationship of transformations about the oil film forces and their Jacobian matrices between [ ]T\u03be \u03b7 and [ ]T x y \u03b5 based upon the transformation matrix T5 are defined as: ( ) ( )T 5 J, ,G =F q q T F \u03b6 \u03b6 (21) T T T 3 J 3 3 J 4 G 5 J 5 T T 4 J 3 4 J 4 \u2212 \u2212 \u2212 = \u2212 = \u2212 \u2212 T K T T K T K T K T T K T T K T (22) T T T 3 J 3 3 J 4 G 5 J 5 T T 4 J 3 4 J 4 \u2212 \u2212 \u2212 = \u2212 = \u2212 \u2212 T D T T D T D T D T T D T T D T (23) Where ( ),GF q q , GK , and GD are the fluid force vector, stiffness matrix and damping matrix in global coordinate" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002176_amm.712.81-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002176_amm.712.81-Figure5-1.png", + "caption": "Fig. 5. Hardness measurement method of gear wheel - five measurement points", + "texts": [ + " The chemical composition of the phases in the sintered material, % wt. Element Cu [%] Ni [%] Mo [%] Fe [%] Light phase 2.85 4.27 - rest Dark phase - - 1.65 rest Fig. 3. The microstructure of the steel 42CrMo4 samples (without coating), nital etched Based on the analysis, it was found that the products do not comply the technical specifications for static strength tests. Breaking point of the gear wheel made of sinter material which was identified during static strength test is shown in Figure 4. Hardness measurement points shown in Fig. 5. In addition, was carried out statistical analysis of the strength of gear wheels made of two materials analyzed to illustrate the strength of the test samples for a period of two years for sinter Sint - D32 (Fig. 6). For sinter material were also statistically evaluated static strength results for the batch of products among which there were identified non-conformed products (cracking during static strength test) for period of three months (Fig. 7). For comparison, performed statistical analysis of the static strength of gear wheels made of steel 42CrMo4 (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003762_s1068798x1410013x-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003762_s1068798x1410013x-Figure4-1.png", + "caption": "Fig. 4. Determining the pressure function.", + "texts": [ + " To assess the possible gap, we use Targ\u2019s approxi mate solution for the rolling of viscous liquid between rf1,2 z1 2, 2\u03c1f* 1 \u03b1lsin\u2013( )\u2013 z1 \u03b1kcos ;= \u03c7f1,2 \u03c0 z1 2, \u03d5P1,2;\u2013= \u03d5P1,2 2y0* y2 1,*\u2013( ) z1 2, ;= y0* 0.5\u03c0 \u03c1f* \u03b1lcos\u2013 \u03c1a* \u03b1dcos \u03b1dsin \u03b1ltan+( ),\u2013= y0* ra1,2 z2 z1 1+\u239d \u23a0 \u239b \u239e 1 \u03b1wcos rf2 1,\u2013 2\u03b4* z1 \u03b1kcos .\u2013= \u03c7a1 = 1 2ra1 \u03b1kcos ra1 2 \u03b12 kcos 4\u03c1a* 2 z1 2\u2013 z1 2 \u2013 \u239d \u23a0 \u239c \u239f \u239b \u239e \u239d \u23a0 \u239c \u239f \u239b \u239e arccos \u03d5P1;+ \u03c7a2 = z1 2z2ra2 \u03b1kcos ra2 2 \u03b12 kcos 4\u03c1a* 2 z2 2\u2013 z1 2 \u2013 \u239d \u23a0 \u239c \u239f \u239b \u239e \u239d \u23a0 \u239c \u239f \u239b \u239e arccos \u03d5P2.+ RUSSIAN ENGINEERING RESEARCH Vol. 34 No. 10 2014 OPTIMIZATION OF NOVIKOV GEARS: NUMERICAL SOLUTION 611 two cylinders [4]. In Fig. 4, we illustrate the calcula tion of the pressure function in the liquid. The pressure is applied to the opposite sides of the midplane of the hydrodynamic layer. Hence, the pres sure function is symmetric relative to the horizontal axis. Therefore, we need only consider half the layer thickness. Taking account of the relevant scale factor (rn = 1), we may write the pressure function in the hydrodynamic layer in the form where f(\u03c8) = 1 \u2013 cos\u03c8 + \u0394 is the dimensionless auxil iary function; \u0394 is half the relative gap; \u03c8 is the devia tion of the cross section from the plane passing through the gear axis; \u03b7 is the dynamic viscosity, Pa s; \u03c9 is the speed of the driving gear; k is the ratio of the gap to the layer thickness on entering engagement" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003533_detc2014-35036-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003533_detc2014-35036-Figure5-1.png", + "caption": "FIGURE 5: FREE BODY DIAGRAM OF THE MRL-SSL RoboCup ROBOT.", + "texts": [ + " (10) tp = positivemin \u2212\u2016vt\u2016\u00b1 \u221a \u2016vt\u20162 +2utdt ut . (11) th = min{ts, tp}. (12) where un < 0 is the constant longitudinal deceleration and ut is the constant lateral acceleration. Besides, dn and dt are depicted in Fig. 4. It is noteworthy that tp is the smallest positive solution of Eq. (11). In this section, the kinematics and dynamics of the fourwheeled MRL-SSL RoboCup robot will be discussed and its velocity and no-slippage acceleration space will be derived. Table 1 represents the parameters of this robot. Robot Velocity Space Figure 5 is the free body diagram of MRL-SSL RoboCup robot. As it can be observed, {w} is the world coordinate system and {b} is the body coordinate system attached to the center of the robot. The kinematic equations of this robot are expressed in {w} as follows: wX\u0307 = w b R(\u03c6)bX\u0307 = w b R(\u03c6)r G\u2212\u2020 q\u0307. (13) 4 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/31/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use G = \u2212sin\u03b11 \u2212sin\u03b12 \u2212sin\u03b13 \u2212sin\u03b14 cos\u03b11 cos\u03b12 cos\u03b13 cos\u03b14 l l l l 3\u00d74 ", + " This method focused on dynamic equations of robot which govern the vehicle dynamics [20]. The force and moment balance equations have been described in the world coordinate system {w} as follows: mwX\u0308 = w b R(\u03c6) 4 \u2211 i=1 fiDi + 4 \u2211 j=1 n j 0 0 1 \u2212mg 0 0 1 . (17) wH\u0307G = w b R(\u03c6) 4 \u2211 i=1 (Pi\u2212PM)\u00d7 fiDi + 4 \u2211 j=1 (P j\u2212PM)\u00d7n j 0 0 1 (18) where m is the robot mass; wX\u0308 = ( x\u0308w y\u0308w z\u0308w )T is the Cartesian acceleration and wH\u0307G = ( H\u0307Gx H\u0307Gy H\u0307Gz )T is the time derivative of angular momentum of the robot in {w}. As it can be observed from Fig. 5, fi is the friction force, and ni is the normal force between the wheel i and contact surface. PM is the position of 5 Copyright \u00a9 2014 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/31/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use the center of mass and Pi is the position vector of wheel i with respect to the center of mass. Di is the traction direction vector of wheel i which is orthogonal to the Pi. Finally, Eq. (19) could be determined which describes the robot acceleration space with control inputs ui \u2208 [\u22121,1]", + " (24) as desired velocity vd to be explored to the next time step and this is repeated until reaching the goal. From line 17 of Algorithm 1, it can be inferred that by using velocity filtering equation, Eq. (16), this is also capable to plan orientation of the robot along the desired trajectory. This means in each time step, the suggested planner, can compute not only the best admissible linear velocity, but also the desired orientation of the robot. During the experiment, the yb-axis of the robot (see Fig. 5) should be always tangent to the trajectory (for dribbling and ball handling maneuvers). Therefore at each time step, in order to make the angle between vd and the robot yb-axis equal to zero, the command \u03c6\u0307d is produced by the planner. The suggested motion planner was implemented successfully by some simulated scenarios in crowded static and dynamic environments. The performance of the suggested motion planner is presented in Table 2. All time results are taken on an Intel core i5 CPU at 2.5 GHz with 4 GB RAM" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003907_iros.2013.6696859-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003907_iros.2013.6696859-Figure4-1.png", + "caption": "Fig. 4. Condition for a FC grasp by [5]: Four vectors \u03b8-positively span R3. The intersection of the trihedra formed by all triples of vectors belonging to the cones C1, C2 and C3 is shown in grey. The cone C4 is depicted in orange and its opposite lies in the intersection.", + "texts": [ + " Complexity: Let N the total number of elements and k be the number of elements in one dimension of the grasp space. A brute-force method runs in O(N2) in 2D. The presented method runs in linear time O(N) in 2D and O(k2(p\u22121)) otherwise. Note that the complexity increases rapidly with the dimension of the space. Starting with a precomputed set of ICRs, the method presented in this section extends as much as possible the initial regions. The approach is inspired by the necessary and sufficient condition proposed in [5] for finding an n-finger frictional FC grasp, as illustrated in Fig. 4: Definition 3.1: n vectors (representing the generatrices of n cones) \u03b8-positively span Rn\u22121 when, for any (n\u22121)-tuple of these vectors, the nth cone centered on the direction opposite of the nth vector lies in the interior of the intersection of the convex polyhedron formed by all (n\u22121)-tuple of vectors belonging to their cones. An application to the more complex case of n-finger ICRs is possible if the iterative extension of each ICR is considered, and the cones in the previous definition are replaced by convex cones for each ICR" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003659_20131120-3-fr-4045.00042-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003659_20131120-3-fr-4045.00042-Figure1-1.png", + "caption": "Fig. 1. Coordinates frames for the quadrotor.", + "texts": [ + " Then, the control strategy should be established in terms of the coupling error as this is where the position and synchronization errors get combined. The quadrotor can be represented as a rigid body evolving in space with mass m and inertia matrix J . The movement of the object is affected by the gravitational force, one main force and three momentums. The four electric motors dynamics are relatively fast and, therefore, they are usually neglected as well as the flexibility of the blades. Let us consider an inertial coordinates frame I = {Ex Ey Ez}, fixed to the ground, and a body fixed coordinates frame B = {ex ey ez} (see Fig. 1). Therefore, the generalized coordinates for the quadrotor are written by \u03be = [x y z]T (4) \u03a6 = [\u03c6 \u03b8 \u03c8]T (5) where (4) represents the position of the center of gravity and (5) represents the three Euler angles: roll, pitch and yaw, respectively. The motion equations are given by the Newton-Euler equations (Kendoul et al. (2006)), this is m\u03be\u0308 = TRez \u2212mgEz (6) J\u2126\u0307 =\u2212\u2126xJ\u2126 + \u0393 (7) where T is the total thrust of the four motors in the body fixed coordinates frame in the ez direction; the quadrotor weight is in the Ez direction" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002984_cca.2014.6981499-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002984_cca.2014.6981499-Figure1-1.png", + "caption": "Fig. 1. Schematic of the cart-beam system", + "texts": [ + " The performance of the control law is demonstrated through simulations. Index Terms\u2014 Cart on a beam, underactuated, energy- based control, passivity In this paper, we address the asymptotic stabilization of a non-symmetric rigid beam balanced by a actuated cart. The system consists of a rigid beam supported by a fixed point located in any place of the beam. The rigid beam freely rotates around that point and a cart traverses on it. The only input to the system is the force acting on the cart (see Figure 1). The problem of stabilization of the CaB system was first introduced in [1], where the dynamics of the system were greatly simplified by neglecting the contribution of the cart to the inertial forces. This assumption paved the way for designing a nonlinear change of state and input coordinates, that transformed the constraints on the original states to the new states. Further, the authors in [1] have considered the problem of state-constrained stabilization of beam-balance systems, having two different actuator configurations: a cart driven on a beam and in the sec- Department of Electrical Engineering, Indian Institute of Technology Madras, Chennai, India (e-mail:arun dm@iitm", + " The domain of the configuration space that results in the constraints being enforced is obtained using the sub-level sets of the closed-loop potential function. The paper is organized as follows. In section 2 the equations of motion of the CaB system and the control objective are presented. The control law synthesis in the framework of IDA-PBC is carried out in section 3. Simulations results are shown in section 4 and the conclusions 978-1-4799-7409-2/14/$31.00 \u00a92014 IEEE 1244 are made in section 5. The system consists of a beam of length l and mass m, the center of the beam is free to rotate on a pivot, under the influence of gravity as shown in Figure 1. The beam geometry is such that the center-of-mass is at a distance lc from the center of the beam. A cart of mass M traverses on the beam and is actuated by a force F and it is assumed that the cart does not slip on the beam. The total energy of the CaB system is given by E(q, p) = 1 2 p\u22a4D(q2) \u22121p+ ( q2 + mlc M ) M g sin q1 (1) where the inertia matrix D(q) is given by diag{J + Mq22 ,M}, J denotes the moment-of-inertia of the beam about its pivot, q1 is the angle made by the beam with the horizontal, q\u03071 is the angular velocity, q2 is the cart position, q\u03072 is the cart velocity and g is the acceleration due to gravity" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002322_1.4029294-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002322_1.4029294-Figure1-1.png", + "caption": "Fig. 1 A branched-chain system composed of four moving articulated rigid bodies", + "texts": [ + " Section 5 includes the final development, an investigation of the accuracy of the experimental SESC in order to provide insight into the number of poses needed to obtain an accurate SESC given measuring error in CoM and joint angles. Section 6 concludes with noting that the developments presented in this paper move SESC modeling toward more practical applications. This section reviews the classical determination of the CoM location of a multilink chain in order to define the SESC. Manipulations of the equations to compute the CoM of the original chain as the end-effector of the SESC are also presented. 2.1 Classical CoM Determination. The determination of the CoM for the articulated spatial four-body system shown in Fig. 1 is given as an example to illustrate the method. Figure 2 shows the kinematic parameters including the joint displacement vector d and joint rotation matrix A. Mass property information is also shown, including the mass m and the relative link CoM location c. Attaching a reference frame at each joint, the CoM of the system is defined by the parameters mi, ci;di, and Ai, where the subscript i indicates the numbering scheme used for identifying each body. The CoM location of each link in its local reference frame is indicated by ci 2 R3", + " In a similar manner, any general branched chain composed of rigid bodies articulated by revolute, spherical, or universal joints defines an SESC. The links, or si, of the SESC are determined by the masses (mi), CoM locations of the links (ci), and the distances between the joints \u00f0di\u00de of the original system. The joint angles of the SESC correspond to the joint angles of the original system. In addition, the SESC maintains the same degrees of freedom (DOF) as the original system [3]. 2.3 CoM Estimation. Continuing with the four-jointed example of Fig. 1, Eq. (3) may be manipulated in yet another way C \u00bc B1s1 \u00fe B2s2 \u00fe B3s3 \u00fe B4s4 \u00bc \u00bdB1 B2 B3 B4 s1 s2 s3 s4 8>>><>>: 9>>>=>>; (7) where B1 \u00bc A1; B2 \u00bc A1A2; B3 \u00bc A1A3; B4 \u00bc A1A3A4 (8) Equation (7) may be generalized to any spatial articulated rigidbody system as C \u00bc \u00bdB1 B2 Bn0 s1 s2 .. . sn0 8>>><>>>: 9>>>=>>>; \u00bc BS (9) where n0 is the number of bodies in the system, B 2 R3 3n0 (B 2 R2 2n0 for the planar case) is the aggregation of rotation matrices Bi, and S 2 R3n0 (S 2 R2n0 for the planar case) is the vector representation of the SESC" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002737_ecticon.2013.6559506-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002737_ecticon.2013.6559506-Figure1-1.png", + "caption": "Fig. 1 Equivalent circuit model of induction motor", + "texts": [ + " In this paper, a simple and efficient technique based on energy conservation law is presented to find a single equivalent circuit of a group of induction motors. The obtained steady-state parameter of the aggregate motor is shown and compared with those appearing in the open literatures. Moreover, the dynamic simulation results obtained from the sum of individual model and from aggregate model are compared to verify the computing accuracy. II. AGGREGATION TECHNIQUE FOR MULTIPLE INDUCTION Generally, the steady-state model of induction motor is represented by the equivalent circuit as shown in Fig. 1. Rs and Rr are stator and rotor resistance. Xls and Xlr are stator and rotor leakage reactance. Xm is mutual 978-1-4799-0545-4/13/$31.00 \u00a92013 IEEE reactance. Let considering the group of induction motors which are connected at the same buses as shown in Fig. 2. In the grouping procedure, it is initially assumed that all parameters of each motor are known. These parameters are required to be adjusted to the same common MVA base. If the operating slip of each individual motor is not available, it can be alternatively computed using (1) with terminal voltage of 1.0pu (Vs) as, 0m eT T\u2212 = (1) Where, ( )2 0 (1 ) (1 )mT T A s B s C= \u2212 + \u2212 + (2) ( ) ( ) ( ) 2 2 2 2 2( / ) ( ) ( ) m s r e th r th lr s m ls X V RT sR R s X X R X X = + + + + + (3) 2 2 2( ) s m th s m ls R X R R X X = + + (4) 2 2 2 ( ) ( ) s m m ls m ls th s m ls R X X X X X X R X X + + = + + (5) The input current, active, and reactive powers of all motors are then computed using the equivalent circuit in Fig. 1. In order to find the aggregate model or single unit model of them, the law of energy conservation is applied in this paper. The apparent power absorbed by the aggregate motor is equal to total power absorbed by all motors. Hence, the total stator and rotor currents of the aggregate motor in complex form can be expressed by, 1 n agg s si i I I = =\u2211 (6) 1 n agg r ri i I I = =\u2211 (7) Based on the energy conservation law, the circuit parameter of the aggregate motor can be derived as follows, 22 1 n agg agg s si si s i R I R I = =\u2211 (8) 22 1 n agg agg r ri ri r i R I R I = =\u2211 (9) 22 1 n agg agg si lsi sls i X I X I = =\u2211 (10) 22 1 n agg agg ri lri rlr i X I X I = =\u2211 (11) 22 1 n agg agg agg m si ri mi s r i X I I X I I = = \u2212 \u2212\u2211 (12) With the same technique, the air-gap power of the aggregate motor can be expressed by, ( ){ }2 1 Re n agg ag si si si si i P V I I R\u2217 = = \u2212\u2211 (13) The slip of the aggregate motor can be computed by, 2agg agg agg agg r r ags I R P= (14) The moment of inertia of a group of the motor can be found by the kinetic energy conservation law as follows, 1 n agg agg i i i H H S S = =\u2211 (15) Where, 21 2i i i siH S J \u03c9= (16) 1 n agg i i S S = =\u2211 (17) It is noted that Si is the rated kVA of individual motor" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003440_j.jappmathmech.2011.05.007-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003440_j.jappmathmech.2011.05.007-Figure1-1.png", + "caption": "Fig. 1.", + "texts": [ + " The contacts at the support points are assumed to be unilateral and subject to the law of dry (Coulomb) friction. The dynamics of possible motions of such a body under the action of gravity forces and dry friction is investigated. In the case of a plane body, it is possible to obtain particular integrals of the equations of motion. \u00a9 2011 Elsevier Ltd. All rights reserved. 1. Description of the model, the equations of motion and the formulation of the problem. Consider a rigid body, supported at its points Ak (k = 1, 2, 3) on a rough horizontal plane (see Fig. 1), where, when moving in a gravity force field, the body does not detach from the support plane, i.e., it performs plane-parallel motion. We will derive the equations of motion of such a body.1 We will introduce a system of coordinates Oxyz, rigidly connected to the body, where the point O is the centre of mass of the body, the z axis is directed vertically upwards, and the xy plane is parallel to the support plane. Everywhere henceforth, unless otherwise stated, the subscript k takes the values 1, 2, 3" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003477_jjap.53.07kf26-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003477_jjap.53.07kf26-Figure7-1.png", + "caption": "Fig. 7. (Color online) Experiment on needle tip detection: (a) experimental setup and (b) photograph.", + "texts": [ + " This phenomenon was confirmed by simulation and experiment. The simulation model is shown in Fig. 4. Transducers A and B with 1.5mm width were placed on the water surface, and the stainless needle was inserted. Insertion depth was varied from 1 to 20mm, and the intensity of the signal from the needle tip was simulated. The signal intensity was evaluated from the vertical displacements of transducers A and B. The experiments on detecting the needle tip in water using the 13 element-probe were performed to verify the simulation results. Figure 7 shows an experimental setup and its photograph. The probe was placed in water. A pulse voltage of \u00b948V at 7MHz was applied to the probe by the pulserreceiver, and the received signal intensity was measured. The stainless steel needle of 1.5mm diameter and 10\u00b0 tip angle was inserted into water. The depth of the needle was varied from 1 to 20mm from the front of the probe. Figure 8 shows the simulation and experimental results. The receiving signal intensity increased with insertion depth up to 15mm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003915_robio.2011.6181260-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003915_robio.2011.6181260-Figure2-1.png", + "caption": "Fig. 2. Physical model of needle", + "texts": [ + " The intra-operative planning task is to estimate pre-planned needle-insertion path and analyze the risk of needle-insertion path respectively, in order to confirm the accuracy and feasibility of the surgical planning. In the following parts, we present the physical simplified model of the needle, based on which the surgical planning is proposed. With the construction of the needle-insertion path, we can figure out that: lesion target BP is known; entry point AP is needed to be planned. So the needle-insertion path could be simplified as different posture with the assumption that lesion target BP is considered as the joint point of ball hinge joint sports. The coordinate frame is set up as shown in Fig.2, where z axis is vertical-up. Without the consideration of the obstacles (vessels, ribs etc.), the workspace of the needle-insertion path is supposed to be a cone. Thus the feasible solution of needle-insertion path is the posture subset of the cone, whose corresponding joint variables are rotation angle \u03b8 around y-axis and rotation angle \u03d5 around z-axis. On this basis, we define the CFRW as a posture set, which the needle can reach without the collision with the obstacles. Next, the solution of CFRW is analyzed in detail", + " According to [15], ribs, vessels are regarded as third-order continuously differentiable implicit functions. Here, we assume that obsf is a third-order continuously differentiable implicit function of the curved surface of obstacles (ribs, vessels etc). The posture of the needle at a moment is recorded as q , and then local coordinate system '\u03a3 is set up, where 'z -axis coincides with the own axis of the needle. The '\u03a3 rotates around y-axis and z-axis in global coordinate system\u03a3 , and the corresponding rotation angles are q\u03b8 and q\u03d5 , respectively. As shown in Fig.2, the homogeneous transformation matrixT of '\u03a3 with respect to\u03a3 , is: cos sin 0 0 cos 0 sin 0 sin cos 0 0 0 1 0 0 0 0 1 0 sin 0 cos 0 0 0 0 1 0 0 0 1 q q q q q q q q \u03d5 \u03d5 \u03b8 \u03b8 \u03d5 \u03d5 \u03b8 \u03b8 \u2212\u23a1 \u23a4 \u23a1 \u23a4 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5= \u23a2 \u23a5 \u23a2 \u23a5\u2212 \u23a2 \u23a5 \u23a2 \u23a5 \u23a3 \u23a6 \u23a3 \u23a6 T The cylinder equation is used to represent the needle segment inserted within the lesion organ. The description equation of the needle surface in '\u03a3 is: ' ' ' ' T ' 2 ' 2 2 a x y z x y([ , , ] ) 0f a a a a a r= + \u2212 = (1) where ' ' ' T x y z[ , , ]a a a is the coordinate of the point on the needle surface in the local coordinate frame '\u03a3 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002681_amr.328-330.186-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002681_amr.328-330.186-Figure1-1.png", + "caption": "Fig. 1 Section of high-pressure internal gear pumps Fig. 2 Illustration of Willis theorem", + "texts": [ + " In this paper, a method of generating tooth profiles for conjugated internal gear is presented. Based on gear geometry and meshing theory, the parametric equations of involute conjugated internal gear are derived. With the help of MATLAB and SolidWorks, 3D models of the gear couple are built to evaluate the feasibility of the method. High-pressure internal gear pumps have significant advantages over other types of pumps, such like high volumetric efficiency, long service life, little flow pulse rate and low noise level, which make them wildly applied in modern industry. As shown in Fig. 1, conjugated internal gear is the key component of high-pressure internal gear pumps, which decides volumetric efficiency, capacity of fluid discharge, wear life, flow pulse, vibration and noise. A proper tooth profile of the gears is the basis to gears and very important to the performances of the pumps. The study on tooth profiles of internal gears employed in the pumps has been done by many researchers [1-4]. Once the general equations of the tooth profiles are obtained, it is possible to calculate the theoretical performance and operational indexes of the pump, which is helpful to the optimization of the pump" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001388_047134608x.w4531.pub2-Figure15-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001388_047134608x.w4531.pub2-Figure15-1.png", + "caption": "Figure 15. Change of static domain structure in an ideal Gossoriented grain caused by compressive stress along the rolling direction (48).", + "texts": [ + " In this particular case, the process of magnetization involves domains reorienting but remaining along {100} directions unless the magnetizing field is very high. When amechanical stress is applied to a crystal of cubic material such as silicon\u2013iron, magnetoelastic energy, E, is introduced given by E \u00bc 3 2 l100s a2 1g 2 1 \u00fe a2 2g 2 2 \u00fe a2 3g 2 3 1 3 3l111s a1g1a2g2 \u00fe a1g1a3g3 \u00fe a2g2a3g3\u00f0 \u00de (26) where l100 and l111 are magnetostriction constants for the material at a particular temperature, s is the stress (positive when tensile), and the direction cosines of the stress and magnetization are (a1,a2,a3) and (g1,g2,g3) with respect to the crystal axes. Figure 15a shows an ideal domain structure in a perfectly oriented grain of grain-oriented silicon\u2013ironwhen no stress is applied. The antiparallel bar domain structure shown in the demagnetized state is primarily due to the high magnetocrystalline energy of the material. Suppose a tensile stress is applied along the [001] direction, then in any domain a1\u00bc 1, a2\u00bc 0, a3\u00bc 0, and g1\u00bc 1, g2\u00bc 0, g3\u00bc 0, so equation 26 simply reduces to E \u00bc l100s (27) In iron and silicon\u2013iron alloys, l100 is positive, so E is negative, showing that the tensile stress reduces the free energy of this domain structure, which initially was in a low-energy state, so the structure remains unchanged. If the material were magnetized along the [001] direction under AC conditions, the presence of the stress would have no effect on the loss. If the stress applied to the structure in Figure 15a is compressive, then s is negative, so the free energy expressed by equation 27 will be positive and a net energy increase will occur. Typically, the magnetoelastic energy given in equation 27 will be of the order of 1% of the total energy dominated by magnetocrystalline energy, but it causes a major reorganization of the domain structure to that shown in Figure 15b. Here, the magnetocrystalline energy is unaffected since all domains still lie along {100} directions, but the bulk of the material is now magnetized along the [100] and [010] directions perpendicular to the stress direction and the magnetoelastic energy of these domains is now negative (this can be shown by putting the appropriate values of a1, a2, a3 into equation 26). In practice, the domain structure is far more complicated as indicated in Figure 9, so the above treatment is an oversimplification because domain walls, grain boundary interactions, and misorientation of grains are ignored but it shows the principle of the process", + " Obviously, if the material is now to be magnetized along the [001] direction, the domains oriented along the [100] and [010] directions must rotate through 90 , so a far more lossy process occurs. Figure 16 shows the effect of tension and compression applied along the rolling direction of electrical steels and in iron-based amorphous material for comparison. In the grain-oriented steel, tension has a small effect as expected from the previous basic analysis, and there is a sharp increase in loss at a certain compressive stress when domains in individual grains take up a structure similar to that shown in Figure 15b in the demagnetized state. Stress applied along the rolling direction of nonoriented steel has a smaller influence on the loss because of the random grain orientation. If stress is applied along any direction apart from the rolling direction, its effect on the domain structure and hence losses can be predicted from equation 26. Figure 17 shows how complex stress made up of orthogonal components parallel and perpendicular to the rolling direction can have a wide ranging effect on the losses (49)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001384_iic.2015.7150723-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001384_iic.2015.7150723-Figure2-1.png", + "caption": "Fig. 2 Schematic diagram of a rolling element bearing.", + "texts": [ + " \u2022 The shape of the pulse generated by impact is modelled by cubic hermite spline The restoring force generated by ball-race contact deformation of the ball is of nonlinear nature because of the Hertzian contact. The local Hertzian contact force and deflection relationship for bearing may be written as (1) where K is the constant for Hertzian contact elastic deformation which depends on the contact geometry. In general the deflection of the ith ball located at any angle e is calculated by following expression (refer Fig. 2): l5 =(xcosei + ysinea - (y) (2) x and y are the deflections along X and Y direction respectively and y is the internal radial clearance which is the clearance between an imaginary circle, which circumscribes the balls and the outer race. At the time of impact at the defect, a pulse of short duration is produced and is accounted in the model by the term !J. i.e. additional deflection. Equation (2) is modified by adding !J. to internal radial clearance and is given by l5 =(xcosei + ysinea - (y + " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001803_1.4031066-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001803_1.4031066-Figure3-1.png", + "caption": "Fig. 3 Contact between rotor and track: (a) configuration at contact and (b) contact model using equivalent spring and damper (view from origin O to ef 1 direction)", + "texts": [ + " The angular velocity components xf 1, xf 2, and xf 3 around each axis and their derivatives are represented as xf 1 \u00bc _h sin a cos d _a sin d xf 2 \u00bc _h cos a\u00fe _d xf 3 \u00bc _a cos d\u00fe _h sin a sin d 9= ; (2) _xf 1 \u00bc \u20ach sin a cos d\u00fe _h _a cos a cos d _h _d sin a sin d \u20aca sin d _a _d cos d _xf 2 \u00bc \u20ach cos a _h _a sin a\u00fe \u20acd _xf 3 \u00bc \u20aca cos d _d _a sin d\u00fe \u20ach sin a sin d\u00fe _h _a cos a sin d \u00fe _h _d sin a cos d 9>>>>= >>>>; (3) By substituting them into the Euler\u2019s equation represented in the floating reference frame ef [7], the following set of equations of motion is derived [5]: I1 _xf 1 \u00fe \u20acc \u00bc 2 d dj jRaFb \u00fe caR2 t _a sin d ccR 2 a _c I2 _xf 2 \u00fe xf 3 I1 xf 1 \u00fe _c I2xf 1 \u00bc 2 d dj jRa sin d Rt cos d Fa I2 _xf 3 \u00fe xf 2 I2xf 1 I1 xf 1 \u00fe _c \u00bc 2Rt\u00f0Fb \u00fe Fc\u00de caR2 t _a cos d 9>>>= >>>; (4) where I1 and I2 are the polar and diametrical moment of inertia of the rotor, respectively. ca and cc are the damping coefficients for the precession and spin motions, respectively. Fa \u00bc Faep3 is the normal contact force between the rotor and the track, Fb \u00bc Fbep2 is the frictional force in the tangential direction due to slide, Ra is the radius of rotor at the contact position, and Rt is the rotor\u2019s length from the center of the rotor to the rotor\u2019s tip, which is the contact point. Figure 3 shows the contact model using equivalent spring and damper. Z \u00bc Rt tan d is the displacement of the rotor\u2019s tip to ep3 direction, and it is the value at the positive side of ef 1 direction. Z0 is the radius gap between the rotor and the track. The velocity vc of the rotor relative to the track at the contact point P in the tangential direction is represented as vc \u00bc d dj jRa _c\u00fe Rt _a (5) When the rotor is in the sliding condition on the track, velocity vc at the contact point P relative to the frame et, which is fixed to the track (refer Fig", + "\u201d While, when the contact occurs and Z > 0 (nutation angle is negative, d < 0), the rotor contacts with the upper side of the track and it is referred to as \u201cupper side contact.\u201d In this paper, the contact force Fa is represented by the combination of the restoring force 061010-2 / Vol. 137, DECEMBER 2015 Transactions of the ASME Downloaded From: http://vibrationacoustics.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/journals/jvacek/934234/ on 02/08/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use with spring constant k and the damping force with damping coefficient c, as shown in Fig. 3(b). The direction of the base vector ep3 is set to the positive direction of the contact force Fa. The contact force in the case of upper side contact, Fau, and the one in the case of lower side contact, Fal, are represented as Fau \u00bc k\u00f0Z Z0\u00de c _Z Fal \u00bc k\u00f0Z \u00fe Z0\u00de c _Z Z \u00bc Rt tan d 9>= >; (6) Then, the contact force Fa is represented as [5] Fa \u00bc Fau : \u00f0Z0 < Z and Fau < 0\u00de 0 : f Z0 < Z and Fau 0\u00f0 \u00de or Z0 Z Z0\u00f0 \u00de or Z < Z0 and Fau 0\u00f0 \u00deg Fal : \u00f0Z < Z0 and Fal > 0\u00de 8>>>< >>: (7) The frictional force Fb changes depending not only on the upper/ lower sides of the contact but also on the sign of relative velocity vc at the contact point" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001157_978-1-4471-2330-9_5-Figure5.12-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001157_978-1-4471-2330-9_5-Figure5.12-1.png", + "caption": "Fig. 5.12 M 1 and M 2 are the centers (midpoints) of the crank and connecting rod, respectively", + "texts": [ + "11 ), then ticking the square beside the y-axis value and changing the value to 3. Click OK when done. 18. Click on File > Save as and save it in the same directory where you will have your M-script fi les. Name the fi le virtual_scene . The extension of your Virtual Reality fi le will be .wrl . In what follows, the virtual scene, virtual_scene.wrl , will be animated by assuming that the crank angle is an input to the system. The animation will be recorded into a movie fi le crank_slider.avi . Figure 5.12 shows the centers (midpoints) M 1 and M 2 of the crankshaft and the connecting rod, respectively. In addition, Fig. 5.12 shows b and g , the angles of the crankshaft, and the connecting rod with the vertical, respectively. The parameters that will be changed in the virtual world are the following: 1. The translation property of Piston 2. The rotation property of the Rod (around its center M 2 ) 3. The translation property of Rod (translation of its center M 2 ) 4. The rotation property of Crank (around its center M 1 ) 5. The translation property of Crank (translation of its center M 1 ) One might ask: How to defi ne the centers M 1 and M 2 of Crank and Rod ", + " In this particular example, the centers M 1 and M 2 will be left at their default position (at the midpoint of the cylinders). The script that comes before the for-loop consists of two parts. The fi rst (refer to Part 1 of crank_slider.m ) is the one dealing with opening and viewing the virtual world. The second (refer to Part 2 of crank_slider.m ) deals with setting the recording of the animation to a *.avi fi le. In this example, the movie will be saved as crank_slider.avi . In the for-loop, the crank angle, b, (Fig. 5.12 ) is being varied from 0 to 4 p rad with increments of 0.08 rad. The rod angle, g, (Fig. 5.12 ) is being computed accordingly. Note that both angles are measured with respect to the vertical line (a zero angle means that the object is aligned with the vertical line). The translation properties of the piston, the crankshaft, and the rod are varied. The rotation properties of the crankshaft and the rod are also being varied. Refer to Sect. 5.2 to see how the coordinates of points N and P are being computed. After you run crank_slider.m , a video fi le crank_slider.avi should be created in the same directory as the MATLAB fi le" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002868_s12206-015-0143-9-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002868_s12206-015-0143-9-Figure7-1.png", + "caption": "Fig. 7. Additional thrust requirement due to tilted nacelle.", + "texts": [ + " However, it is common from a linear control design that a small nonlinearity can be neglected, and from this basis, small tilt angle is to be neglected during the attitude control. This assumption of \u2018small angle\u2019 allows the elimination of any additional aerodynamic control surfaces, and the control de- sign is simplified to an equivalent quad-rotor UAV during forward flight. Another assumption of \u2018small thrust variation\u2019 is also used. During the forward flight mode, the tilted nacelle produces forward thrust as well as vertical thrust as illustrated in Fig. 7. When the PAV is hovering with the nominal thrust 0T , it will then require additional thrust of AT because the vertical component of additional thrust VAT is required to maintain the same vertical thrust. The horizontal thrust HT will also be affected by horizontal component of additional thrust HAT [8]. However, this nonlinear effect is also neglected for attitude control design because the variation of required thrust and power are small up to 20\u00b0 regardless of the maximum takeoff weight as analyzed from the Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001600_ecc.2013.6669841-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001600_ecc.2013.6669841-Figure2-1.png", + "caption": "FIGURE 2. Schematic drawing of multi-DOF planar robot with revolute joints and with a single reaction wheel attached to each link.", + "texts": [ + " Since the controller needs to gather the measurements from the sensors, simulate the system for one-step ahead prediction, compute the control sequence and transmit it to the actuators in one loop, the following condition must hold true: Tsens + Tsim + Tsolv + Ttran \u2264 Ts (8) 4620 VOLUME 4, 2016 However, even if at a given time step this condition is violated, the system will still be fault-tolerant to a limited extent, since it has simulated and generated controls forKNMPC steps ahead in the previous time step; therefore, it can skip one step and apply the next control inputs in the queue. For a more detailed description of this framework the reader is referred to [19]. III. REACTION WHEEL INTEGRATED VSA ROBOTS A. SYSTEM MODELING In this subsection we describe the dynamics of VSA robots with reaction wheels. For simplicity, we limit our treatment to planar VSA robots which have one reaction wheel attached to every link. We also assume that robot links are moved by rotational joints only. The schematics of a such VSA robot is shown in Fig. 2. Without loss of generality, it is assumed that the center of mass of the momentum wheels are coincident with their axis of rotation. The described procedure can be extended to non-planar multi-DOF VSA robots with rotational joints. Angular position and velocity of link i are described by \u03c8i and \u03c8\u0307i, respectively. Angular position and velocity of the wheel, attached to the link i, are denoted by the generalized coordinates \u03b6i and \u03b6\u0307i. A joint i is connected to the proximal end of the link i and causes its motion" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002460_ichqp.2014.6842802-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002460_ichqp.2014.6842802-Figure6-1.png", + "caption": "Figure 6.b. Magnetic flux for one pole.", + "texts": [ + " 3D GENERATOR ANALYSIS The model of the generator for the 3D analysis was made in Solidworks [8]. The 3D analysis was performed in Ansoft Maxwell software [9]. Because of the computational efforts were too large for the analysis of the entire generator, the model was divided in 3 parts and just one phase has been analyzed. From 3D simulations, the supposition that the magnetic flux is enclosed between the closest poles and the rotor yoke is proved. We used this supposition in the 2D simulation. We can observe the magnetic flux vectors in figure 6.a and 6.b. In figure 7 we can see that the higher values for the magnetic flux is obtained in rotor yoke, near the magnets/poles. In the absence of load, the distribution of magnetic flux is stationary. To obtain the voltage values for one phase of the generator were performed independent magnetostatic simulations at different angles from the initial position. The average flux for the three coils per phase was calculated, for each simulation, using the program. Using the results we calculated the voltage in no-load conditions" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003286_imece2013-64128-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003286_imece2013-64128-Figure1-1.png", + "caption": "Figure 1. Turbocharger model and simplified model", + "texts": [ + " A nonlinear oil film force database method is applied in this simulation and the pad inertia is considered. The dynamics characteristics are presented in bifurcation diagrams within a certain rotor speed range. The result depicts the alternation of the rotor system among the periodic, quasi-periodic and chaotic motions caused by the change of rotor speed and also presents the influence of journal diameter and pad inertia on the nonlinear vibrations of the journal and rotor center. DYNAMICS MODEL Figure 1(a) shows the discussed model of the turbocharger. The unbalances are applied at M4 and M12. The tilting pad bearings are supported at M8 and M9. Figure 1(b) shows the simplified model. Four lumped masses locate at shaft ends M4, M12 and both bearing position. The disc's moment of inertial at M8 and M9 are omitted. The two bearings are tilting 4 pad bearings. As shown in Fig. 2, for each disc, a coordinate system used in numerical integration is selected. 2 Copyright \u00a9 2013 by ASME Downloaded From: https://proceedings.asmedigitalcollection.asme.org on 12/05/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use The mathematical expressions of the simplified model can be described by twelve differential equations" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003026_amm.401-403.254-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003026_amm.401-403.254-Figure9-1.png", + "caption": "Figure 9 The diagram of rolling element bearing on vibrating screen with two pitting fault", + "texts": [ + " 41, \u03c6wn\u2014The location of each pitting in load distribution area, n=1,2 1201 coscos \u03b3\u03b3 mcwmcw FFGF ++=\u2211 (42) (2)Vibration model of rolling element bearing inner ring on rotating machinery with two pitting fault When the inner ring have pitting corrosion fault, because of the relative position of pitting and load distribution is changing. Impact load Fmcwn of each pitting position is a variable. So: > \u2264\u00d7 = L Ln mcin Qtd F \u03d5\u03d5 \u03d5\u03d5\u03d5 0 )()( (43) \u03d5\u03b3\u03b3\u03b3 cos))cos(cos( 10201 +++=\u2211 mcimci FFGF (44) Put Eq. 40 and Eq. 42 into Eq. 18, we can get vibration model of rolling element bearing on rotating machinery with two pitting fault. (1)Vibration model of rolling element bearing outer ring on vibrating screen with two pitting fault Rolling element bearing of vibrating screen with two pitting fault was show in Figure 9. > \u2264\u00d7 = L Ln mcwn Qtd F \u03b2\u03b2 \u03b2\u03b2\u03b2 0 )()( (45) )cos(coscos)( 10101 \u03b3\u03b3\u03b3\u03b1\u03d5 +++=\u2211 mcwmcwr FFFF (46) (2)Vibration model of rolling element bearing inner ring on vibrating screen with two pitting fault > \u2264\u00d7 = L Ln mcin Qtd F \u03b2\u03b2 \u03b2\u03b2\u03b2 0 )()( (47) )cos()cos(cos)( 10201 \u03d5\u03b3\u03b3\u03d5\u03b3\u03b1\u03d5 +++++=\u2211 mcinmcinr FFFF (48) Put Eq. 46 and Eq. 48 into Eq. 35, we can get vibration model of rolling element bearing on vibrating screen with two pitting fault. The test of vibrating screen model is SDM00, is double-shaft motor multifunctional vibration screen" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure2.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure2.2-1.png", + "caption": "Fig. 2.2. Typical 7 DOF anthropomorphic robot arm", + "texts": [ + " Once the new haptic application has been rudimentarily developed using ViSHaRD10 and the feasibility is verified, a tailored, highly specialized haptic display with exactly matching mechanical properties can be developed. The first considerations regarding the kinematic design of a redundant interface have been focused on the class of standard kinematical designs consisting of a 3-jointed spherical shoulder, a single elbow joint, and a 3-jointed spherical wrist. These arms can be described as anthropomorphic after [30]. Exemplarily, a configuration with a wrist in roll-pitch-roll configuration is illustrated in Fig. 2.2. The strength of these mechanisms is the size of the workspace which is optimum for 7 DOF robots in terms of the ratio of the arm length to the working volume. The translational workspace is a sphere with an interior singularity at the center. The angular workspace is 360\u25e6 around each axis since singularities in the wrist can be avoided by rotating the elbow around the line from the shoulder to the wrist. A kinematic analysis of the design shown in Fig. 2.2 is presented in [31]. Among the drawbacks we identified for 7 DOF anthropomorphic arms are: \u2022 Gravitational load: Only the first joint axis is designed to be vertical for arbitrary positions and orientations of the end-effector. As a consequence high motor torque is required to compensate for gravitational load. \u2022 Interior singularity: The singularity in the center of the workspace impairs the dexterity and thus the performance of the device when moving the end effector close to the shoulder. An elimination of this singularity requires at least two additional redundant joints placed between the shoulder and the wrist" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002487_mhs.2011.6102192-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002487_mhs.2011.6102192-Figure1-1.png", + "caption": "Figure 1 Inchworm type microrobot", + "texts": [ + " Some microrobots realized the linear displacement by the principle of an inchworm. These microrobots consist of three piezos and three electromagnets(1). One of three electromagnets is not excited and the electromagnet moves on a surface by the deformation of the piezos connected. While the non-excited electromagnet moves, the electromagnet slides on the surface, and the friction between the electromagnets and the surface. In addition, a current energy source for the electromagnet and a voltage energy source for the piezos were needed. Figure 1 shows the microrobot which uses piezos. The microrobot is composed of three metal blocks with different weight and two horizontal piezos. This microrobot moves by the expansion and contraction of the horizontal piezos. When the microrobot moves, the friction between the microrobot and an operation surface disturbs the 978-1-4577-1362-0/11/$26.00 \u00a92011 IEEE - 273 - displacement of the microrobot. The friction force is usually proportional to the normal force. Quick deformation of a vertical piezo change both the normal and friction force(2)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003606_amm.339.510-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003606_amm.339.510-Figure3-1.png", + "caption": "Figure 3 A single tooth of gear calculation area", + "texts": [ + " Aim for simulating real actual working condition of gear, ensuring rigidity of teeth and body of gear, and simplifying finite element analysis scale, three teeth on the gear are built in the process of modeling, and the rest part is showed with cylindrical which critical dimension is pitch circle; Choose the condition when gears are meshing on the middle pitch line, and use finite element model of gear thermal analysis based on SOLID 70 element, as is shown in figure 2. Loading and solving thermal analysis model. For a single tooth of gear, according to the differences of heat transfer style of each part, there are 5 calculation areas. As is shown in figure 3, the calculation of boundary condition is shown in reference [5] . In the figure 3, GM area is the meshing surface;GT1, GT2, GT3 are separately top, root and no meshing surface of tooth; GD area is end surface; Gw area is bottom surface; Gj, GJ are sections surfaces of tooth. Besides, load average heat flux density and convection heat transfer coefficient of meshing surface on the pitch cylindrical surface. Solve the thermal analysis model and get the temperature field distribution nephogram of capstan and driven gear, as is shown in figure 4. From the figure, the temperature field distribution of standard involute cylindrical gear follows the rules: (1) It is shown from the result that the high speed and heavy load on the gear teeth produces the bigger temperature gradient, and the distribution style of each tooth on the same gear is the same" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001046_20070903-3-fr-2921.00038-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001046_20070903-3-fr-2921.00038-Figure3-1.png", + "caption": "Fig. 3. Four-rotors rotorcraft", + "texts": [ + " The simple webcam logitech has been used and the camera parameters are estimated by the two planes method which gives us a simple camera characterization (J. Fabrizio et al., 2002)(K. Gremban et al., 1988). In this case the objective is to obtain a relationship between the displacement in the image plane expressed in pixels and the realworld displacement expressed in meters. The helicopter\u2019s altitude is regulated around the desired value d = 0.5 m which implies that a 0.01m relative displacement is represented by a 4 pixels displacement in the image plane. 6. ROTORCRAFT DYNAMICAL MODEL A four-rotor craft (figure 3) is controlled by varying the angular speed of each one of the rotors. The force fi produced by motor i is proportional to the square of the angular speed, that is fi = k\u03c92 i . The front and rear motors rotate counterclockwise, while the other two motors rotate clockwise. Gyroscopic effects and aerodynamic torques tend to cancel in trimmed flight. The main thrust is the sum of the thrusts of each motor. The pitch torque is a function of the difference f1 \u2212 f3, while the roll torque is a function of f2 \u2212 f4, and the yaw torque is the sum \u03c4M1 +\u03c4M2 +\u03c4M3 +\u03c4M4 , where \u03c4Mi is the reaction torque of motor i due to shaft acceleration and the blade\u2019s drag" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002502_fie.2014.7044079-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002502_fie.2014.7044079-Figure2-1.png", + "caption": "Fig. 2. Upper: Artistic impression of the SEM car test bench. Lower: Preliminary work on the bench.", + "texts": [ + " Additionally and in order to be competitive, the car design should take into consideration the race track (length of straight sections of the track, sharpness of turns, slopes in the track, surface finish of the track, weather conditions, etc.). The driving strategy (acceleration, cruising speed, cutting the power to the motor at strategic locations in the track, avoiding the use of breaks) is of paramount importance in energy efficiency; this strategy should be well prepared before the race. Therefore an SEM-dedicated computer-controlled test bench is being developed for this purpose at Qatar University (See Figure 2). It is anticipated that this test bench will contribute to significant improvement in the achievements in future SEM competitions. IV. RESEARCH QUESTIONS AND INVESTIGATION HYPOTHESES A number of research questions have been formulated to investigate the impact of Shell Eco-Marathon on students, these are: - RQ1: Does the SEM design experience have an impact on attitudes and confidence toward engineering as compared with pre- participation? - RQ2: Does the SEM design experience have an impact on 21st Century engineering skills compared with University experience" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002480_2015-01-0680-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002480_2015-01-0680-Figure1-1.png", + "caption": "Figure 1. Schematic of the bench rig", + "texts": [ + " After the rig was constructed, tests were run and the scuffed pins from the rig were compared to scuffed pins run in fired engines. CITATION: Zhang, R., Zou, Q., Barber, G., Zhou, B. et al., \"Scuffing Test Rig for Piston Wrist Pin and Pin Bore,\" SAE Int. J. Fuels Lubr. 8(1):2015, doi:10.4271/2015-01-0680. 16 The bench test rig consists of loading system, drive system, synchronization system, data acquisition system and specimens. A schematic drawing of the main frame of the bench rig is shown in Figure 1. The loads on the piston pin are produced by a PZT (Piezoelectric Transducer) actuator which is held in a linear bearing. A load cell measures the dynamic loading during the test. The load is applied to the pin by the small end of the connecting rod, hence on each of the piston wrist pin and pin bore bearings there is half of the load. In the bench test rig, the piston is placed top side down. With such configuration, there will be more contact area between piston and piston holder which increases the stiffness and hence aids in producing a load on the piston wrist pin during operation", + " Hence the load applied in the piston assembly is correlated with the rotating position and cycle of the connecting rod. The Piezoelectric transducer is a solid-state ceramic actuator which converts electrical energy directly into linear motion (mechanical energy) with high resolution. The PZT (Piezoelectric Transducer) actuator itself has very high stiffness for compression (66 kN/60 \u03bcm). Due to its stiffness at both ends of the PZT (Piezoelectric Transducer) actuator, even a small displacement can generate a high load. In the schematic drawing (Figure. 1), the PZT (Piezoelectric Transducer) actuator is in contact with the load cell components at one end. On the other end a fine pitch screw is utilized to apply a preload on the piston. The compression force generated by the actuator is applied on the small end of the connecting rod; hence on each side of the piston wrist pin and pin bore there is only half of the compression load. The response time of the PZT (Piezoelectric Transducer) actuator is 100 ms (Figure. 3); thus the actuator needs 100 ms to increase to its maximum range and also needs 100 ms to return from the maximum range to the original position" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001502_intmag.2015.7156611-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001502_intmag.2015.7156611-Figure1-1.png", + "caption": "Fig. 1. Sizing optimization and the skewed model of proposed TFPMG. (a) Schematic structure. (b) Side view of the rotor and the stator. (c) Cogging torques with different ks and kr. (d) Top view of the skewed model.", + "texts": [], + "surrounding_texts": [ + "INTERMAG 2015\nCogging Torque Optimization of a Novel Transverse Flux Permanent Magnet Generator with Double C-hoop Stator for Wind Power Application. Z. JIA, H . LIN Southeast University, Nanjing\nI . INTRODUCTION Transverse flux permanent magnet generator (TFPMG) is especially suitable for wind power application for the merits of large pole numbers, decoupled magnetic circuit, and high power density . The distinguishing feature of TFPMG is the magnetic flux existed in three-dimensional space and three-dimensional finite element method (3-D FEM) is employed to analyze its characteristics . Such as flux-switching TFPM generator [1], many TFPMGs with new topologies have been proposed . However, they commonly have a drawback that only half of PMs do work at the same time and the cogging torque vibrations are unacceptable and desiderated to be optimized . The proposed 12 pole-pairs TFPMG overcomes these shortcomings, which schematic structure is shown in Fig . 1 (a) . The generator is constructed by the double C-hoop stator cores inserted into machined cavities in the stator holder, the doubled PMs screwed onto two rotor disks with opposite polarities to enable the flux-concentrated effect, and the armature winding bundling all stator hoops . II . COGGING TORQUE ANALYSIS The sizing optimization of the stator and the rotor to reduce the cogging torque is analyzed and the side view of the rotor and the stator with indicated size parameters is shown in Fig . 1(b) . The previous investigations demonstrated that the cogging torque is significantly influenced by the ratios of ks and kr, which denote the ratios of circumferential widths of stator hoop and rotor core to pole pitch, respectively [1-2] . The 3D-FEM is employed to investigate the magnetic field distributions and the size relationship . The ratios comply with the inequation of (kri > 1 - ksi) (kri >kso) = kri > 1 - 0 .788 * ksi are described in detail to optimize the cogging torque performance . Fig . 1(c) shows that the amplitude of cogging torque of a single phase TFPMG with optimized ratios of kr and ks is significantly reduced . The circled effect indicates that the optimized cogging torque varies smoother than the original one due to the increased ks with decreased kr . In view of the mentioned factors the optimum values of ks and kr are 0 .76 and 0 .71 to achieve the best performance of the cogging torque and the flux linkage . III . COGGING TORQUE OPTIMIZATION According to the above analysis, the method of skewing the rotor core is applied to improve the flux linkage and minimize the cogging torque of the generator . The top view of the skewed rotor core versus the stator hoop is shown in Fig . 1(d) . The static and transient characteristics of the skewed models at a series of rotor positions from 0 to 6 mechanical degrees are calculated . The no-load flux density distributions under unskewed and skewed rotor core of 3 .6 degree based on 3-D FEM are obtained . As shown in Fig . 2(a) the unskewed rotor core is easier to saturation and the maximum magnetic field intensity is 2 .15 T . On the contrast, as shown in Fig . 2(b) the skewed rotor core of 3 .6 degree does benefit to bring fairly well-distributed flux at the average magnetic field intensity of 1 .5T . Meanwhile, the effect of the skewing method does benefit to the utilization of uniform distribution of magnetic field intensity in both stator and rotor cores . The back-EMF waveforms of skewed rotor core are shown in Fig . 2(c) . It can be seen that when the rotor core is skewed too deep to 4 .8 degree the amptitude of back-EMF will decline sharply . Apparently, when the rotor core is skewed to 3 .6 degree, the back-EMF waveform is the closest to the sinusoid with relatively low total harmonic distortion (THD) . The influences of skewed rotor core on the cogging torque waveforms without and with skewed to 3 .6 degree rotor core are compared in Fig . 2(d) . Obviously, the amplitude of the cogging torque of the skewed rotor core is reduced significantly over 50% of the unskewed structure with minimized cogging torque ripple . The predicted cogging torque waveforms without and with skewed to 3 .6 degree rotor core verify that the optimization is feasible . Hence, the electromagnetic torque will increase enormously if the load current is applied and then the cogging torque is negligible . So the influences of skewing method on static and transient characteristics are equilibrated to consider the \u201cSkewed 3 .6\u201d as the optimal choice .\n1) Jianhu Yan, Heyun Lin, and et al . \u201cCogging torque optimization of flux-switching transverse flux permanent magnet machine,\u201d IEEE Trans. Magn., 49(5) 2169-2172, (2013) . 2) A . Masmoudi, A . Njeh, and et al . \u201cOptimizing the overlap between the stator teeth of a claw pole transverse flux permanent magnet machine,\u201d IEEE Trans. Magn., 40(3) 1573\u20131578, (2004) ." + ] + }, + { + "image_filename": "designv11_84_0002334_2013-01-1478-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002334_2013-01-1478-Figure9-1.png", + "caption": "Figure 9. Module cooling structure", + "texts": [ + " By using batteries with lithium titanate negative electrode, the level of capacity fade to take into account was much lower than batteries with carbon electrode. There are some other major factors of battery degradation, such as stored conditions at high temperature and at high SOC (state-of-charge). The cooling structure of the battery pack has enabled to suppress battery degradation. An indirect cooling structure is adopted where cold air passes through the cooling paths under battery cells and contributes to heat transfer, as shown in Figure 9. A thermal conductive material is attached on the bottom of the battery modules, so that it can transfer the heat from the cell to the cooling passage. Simulations were conducted to ensure the cooling structure to efficiently cool all battery cells, keeping the battery temperature within the designated range. Figure 10 shows the results of an investigation into the battery system cooling capacity. Without the cooling system, battery temperature will continue rising during repeated driving and recharging cycles and it will eventually exceed its limit" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003520_detc2014-34759-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003520_detc2014-34759-Figure9-1.png", + "caption": "Figure 9 Contact Pattern of the Existing and New Tooth Profiles", + "texts": [ + " Time (s) Figure 8 Output angular velocity\u2019s hysteresis Figure 8 is circular spline\u2019s output as a function of time. Series1 is the theoretical data at each time step, series 2 is the simulation data at each time step. The hysteresis phenomenon can be observed which has been proved by Curt Preissner [4] through phenomenological approach using skillfully designed dynamometer. The simulation data can be used by control algorithm design to improve output dynamic response. Contact ratio is another criteria to evaluate tooth profile. Figure 9 shows the teeth number engaged in the new tooth profile are 29, 61% more than the existing design. The new tooth profile improves the stress concentration condition of the root by even stress distribution of more teeth. This paper presents the development of tooth contact analysis. With the capacity of computer and parallel calculation growing, applying explicit dynamics to TCA is possible. First the explicit algorithm will allow the FEA simulation model include all of the three nonlinear source: nonlinear geometry, nonlinear material, and compacted contact" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure1.44-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure1.44-1.png", + "caption": "Fig. 1.44 Elliptical extrusion", + "texts": [ + "43, these are useful as the anchor areas to both the bones and muscles. For the creation of these holes, the sketching and extrude tools were used. By the top line, the circles are at 6.25 mm; while on the bottom line are separated 6mmbetween each center, andwith a diameter 28 J. A. Beltr\u00e1n-Fern\u00e1ndez et al. of 2.5 mm all circles, as shown in Fig. 1.43. Finally, these circles were extruded for 4 mm. Next, it was to create an elliptic shape to the top of the part with an elliptical extrusion, as shown in Fig. 1.44. The sketching and extrude tools were reused. For the next step, five rounds were applied as shown in Fig. 1.45, this is in order to not generate injury and discomfort to the patient and generate more similar forms to anatomical ones. Only the rounding tool was used here. 1 Comparative Study of Interferometry and Finite Element Analysis \u2026 29 Almost to finish, an extension was added to the same piece, as shown in Fig. 1.46 with a thickness of 4 mm, equal to that of the part. This was done so that the piece is close to that of the prosthesis" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002320_s1068798x11020122-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002320_s1068798x11020122-Figure2-1.png", + "caption": "Fig. 2.", + "texts": [ + " In calculating the decrease in distance between the cylinder axes, we take account of three components of the total deformation: two contact deformations in the contact zones of the cylinders with the two planes; and the compressive deformation of the cylinder. The compressive defor mation Wc is found from Eq. (4). The contact defor mation Wco at the two planes is found from Eq. (2), where we assume that z = z1 = z2. The decrease in distance between the planes is determined by the sum W1 = Wc + 2Wco where EXAMPLE 2 Two contacting cylinders are placed between two compressive planes (Fig. 2). The decrease in distance between the planes is determined by the sum of the two compressive deformations Wc1 and Wc2 of the cylin ders, the two contact deformations Wco1 and Wco2 in the cylinder\u2013plane contact zones, and the contact deformation Wco3 in the cylinder\u2013cylinder contact zone where i = 1; 2. W1 2\u03bbq 2R b1 \u239d \u23a0 \u239b \u239eln 2 2z b1 \u239d \u23a0 \u239b \u239eln 0.407+ + ,= b1 4\u03bbqR= . W2 Wc1 Wc2 Wco1 Wco2 Wco3;+ + + += Wc1 Wc2+ = 2q \u03bb1 2R1 b1 ln 0.407+\u239d \u23a0 \u239b \u239e \u03bb2 2R2 b2 ln 0.407+\u239d \u23a0 \u239b \u239e+ ; Wco1 Wco2+ q \u03bb1 4z1z2 b1 2 ln \u03bb2 4z1z2 b2 2 ln+ ;= Wco3 \u03bbq 4z1z2 b2 ,= bi 2 4q\u03bbRi,= P W 1 P Fig", + " Com parison shows that the calculation results and experi mental data are in good agreement (Fig. 4). This is explained by the adoption of the value z/b = 7.5 and the recommendations of Belyaev and Petrusevich. The agreement obtained verifies the theory and model here proposed. Analysis of Fig. 4 shows that disregarding the contact deformation of the bodies leads to significant error. Apparatus. Dinnik presented experimental data for the decrease in distance between elastic bodies in lin ear contact in [3]. The loading configuration is shown in Fig. 2. His apparatus is a 5 t press. Samples. Two identical disks are employed (radius R = 2.98 cm; length L = 1.94 cm). The disks are made of steel (E = 2.1 \u00d7 106 kg/cm2; \u03bcm = 0.3). Measurement system. The decrease in distance between the centers of the disks is measured by means of a mirror instrument. Test results. The test results are shown in Fig. 5 (circles). Calculation. In this case, the decrease in distance between the disk axes consists of the contact deforma tion of the disks and two halves of the compressive deformation of the disks" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001946_amm.698.552-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001946_amm.698.552-Figure1-1.png", + "caption": "Fig. 1. Bricard\u2019s linkage", + "texts": [ + "ome additional conditions ensure their mobility. The article identifies these special conditions providing the assemblability and mobility of Bricard\u2019s linkage modifications. R. Brikard proposed a mechanism consisting of six one\u2014degree-of-freedom kinematic rotary pairs in 1927 [1]. This mechanism is interesting due to its paradoxical mobility [2] which does not agree with theoretical mobility. This feature allows finding the application of this mechanism in practice [3]. Brikard\u2019slinkage (Fig. 1) consists of six links AB, BC, CD, DE, EF, and FA connected by rotary joints (hinges) A, B, C, D, E, and F. Link AB is fixed motionlessly and acts as a rack. Link BC is a driving link and its rotation drives the whole mechanism. The driving link rotation angle is the rotation angle of link BC relative to the fixed link AB (hinge B angle). The rotation angle of link FA relative to the fixed link AB (hinge A angle) is the driven link rotation angle. The kinematic analysis of the mechanism has allowed defining the relation between the driving link rotation angle and the driven link rotation angle", + " In Brikard's linkage each of the six links turns the direction vector through 90 \u00b0 clockwise therefore the mechanism can be designated as (-90-90-90-90-90-90). Links ABand BC were joined so that the direction vector of the B hinge as a part of the AB link coincided with the direction vector of the B hinge as a part of the AB link. When the mechanism is open, the links built in one line form the chain AB-BC-CD-DE-ED-FA\u2032 shown in Fig. 3. The direction vector turns through 90\u00b0 clockwise when passing each hinge from A to A \u2032. Closing the chain by aligning hinges A and A\u2032 forms Bricard\u2019s mono-mobile linkage (Fig. 1). The mechanism is assemblable if its chain (Fig. 3) can be closed at least for one value of the driving link rotation angle. The mechanism is mobile if its chain can be closed for a continuous range of driving link rotation angles. The mechanism has a crank if its chain can be closed for any value of the driving link rotation angle. It has been shown above that Brikard's linkage (-90-90-90-90-90-90) is assemblable and mobile. Within this work Brikard's linkage modifications (\u03b11 \u03b12 \u03b13 \u03b14 \u03b15 \u03b16) assemblability condition is defined", + " The mechanism consists of six units, which can be divided into three pairs of units: P1, P2, P3 (Fig. 8). Brikard's linkage modification (\u03b11 \u03b12 \u03b13 \u03b14 \u03b15 \u03b16) is mobile if all links turn the direction vector in one direction (clockwise or counterclockwise), and each of the three link pairs rotates the direction vector through 180\u00b0. So, Brikard's linkage modification is mobile if: , where k=\u00b11 (1) Let us verify the satisfaction of condition (1) of the mechanisms discussed above. For Bricard\u2019s linkage (Fig. 1): \u03b11= -90\u00b0 \u03b12= -90\u00b0 \u03b13= -90\u00b0 \u03b14= -90\u00b0 \u03b15= -90\u00b0 \u03b16= -90\u00b0 So: , where k=\u00b11 (2) System (2) is correct for k=-1 so Bricard\u2019s linkage is mobile. For the mechanism shown in Fig. 6: \u03b11= -90\u00b0 \u03b12= -80\u00b0 \u03b13= -70\u00b0 \u03b14= -100\u00b0 \u03b15= -120\u00b0 \u03b16= -80\u00b0 So: , where k=\u00b11 (3) System (3) is not correct for k=\u00b11 so the mechanism is stationary. Let us change link parameters to provide mobility of the mechanism: \u03b11= -100\u00b0 \u03b12= -80\u00b0 \u03b13= - 70\u00b0 \u03b14= -110\u00b0 \u03b15= -120\u00b0 \u03b16= -60\u00b0 The assembly of mechanism links in a different order can not affect the assemblability, but it may affect the mobility" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001803_1.4031066-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001803_1.4031066-Figure1-1.png", + "caption": "Fig. 1 Model of gyroscopic exercise tool power ball", + "texts": [ + " The developed hybrid dynamical model is used for a numerical simulation, and both the uniform precession motion and the periodic reversing motion are observed and investigated. The dynamical behavior in these motions is explained in detail focusing on the transition between the sliding mode and the rolling mode. Furthermore, the influence of both the magnitude and the frequency of input motion for the occurrence of these motions is investigated numerically and confirmed by the experiment. The physical model used in this paper, which is the same one used in the previous paper [5], is shown in Fig. 1. The center of the rotor is fixed at the origin. Therefore, the translation displacement is kept to zero and only the spatially rotational motion is considered. This assumption is made in order to reduce the DOF in the model and to consider only the essential motion in the modeling. The validity of this assumption is confirmed in Sec. 5 experimentally. The input periodic rotational motion with the angle h is Contributed by the Technical Committee on Vibration and Sound of ASME for publication in the JOURNAL OF VIBRATION AND ACOUSTICS" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003671_detc2011-48794-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003671_detc2011-48794-Figure1-1.png", + "caption": "Figure 1. Typical Toroidal CVT.", + "texts": [ + " The main friction type CVTs are: V-Belt, chain and Toroidal traction-drives. In this paper we will focus on Toroidal CVTs. Toroidal CVTs have higher torque capacity compared to V-Belt and chain CVTs and can thus be used in Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use 2 Copyright \u00a9 2011 by ASME mid-size and full-size passenger cars. Toroidal CVTs contain input and output shafts with toroidal disks mounted on each shaft, forming a toroidal cavity (Fig. 1). Motion transmission between the shafts is provided by multiple rollers in contact with the two disks. The rollers can change the speed ratio by tilting their angles. Very high contact pressures (2-3 GPA) are needed between the rollers and the disks to transmit the required torque. The rollers and the disks don\u2019t have any metal to metal contact. A thin film of lubrication oil separates the two surfaces. Traction between the rollers and the disk can be modeled using the elasto-hydrodynamic lubrication (EHL) theory [2, 3]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003608_isciii.2011.6069753-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003608_isciii.2011.6069753-Figure2-1.png", + "caption": "Fig. 2. The cross section of the SRM in study", + "texts": [ + " HAVING DIFFERENT FAULTY CONDITIONS All the winding faults of a SRM cause unsymmetrical field distribution inside the machine. The best way to emphasize these changes is to perform a precise numeric field analysis of the SRM. The main data of the simulated SRM are: i.) Rated power 350 W ii.) Rated voltage 300 V iii.) Rated current 6 A iv.) Rated speed 600 1/min v.) Number of stator poles 8 vi.) Number of rotor poles 6 The cross section of the motor together with its pole's notations is given in Fig. 2. The numeric field computations were carried out by using the FEM based Flux 2D program package produced by Cedrat (France) [10], [11]. The simulations were performed for the healthy machine and for the following three winding fault conditions: i.) coil A having 20% of turns shorted ii.) coil A having 50% of turns shorted iii.) coil A having all its turns shorted. The most significant results are given in Figs. 3 and 4, where the flux lines obtained via field computations are shown for the A stator pole being aligned, half-aligned and unaligned Rare\u015f Terec et al" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003219_amm.87.30-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003219_amm.87.30-Figure1-1.png", + "caption": "Figure 1. Powertrain mounting system model", + "texts": [ + " The system is simplified with concerning the actual arrangement of the mounting system of the HEV [8]. (1) The three axial stiffness of mounting is concerned separately with ignoring the role of mounting damp, and the three axial stiffness\u2019 direction is considered along the three principal elastic axes which are perpendicular to each other; (2) The powertrain system is regarded as a rigid body, and the subframe and the powertrain system are considered as an integrated part. The system model is established with powertrain system centroid being the origin of coordinate which is shown in Figure 1. The six degrees of freedom (DOFs) are x translation, y translation, z translation, x rotation, y rotation, and z rotation. In Fig. 1, x, y and z are the translation DOFs along lateral Ox, longitudinal Oy and vertical Oz respectively; \u03b1, \u03b2 and \u03b3 are the rotation DOFs around lateral Ox, longitudinal Oy and vertical Oz respectively; i=1, 2, 3, 4 are the signs of the mounting of front, back, left and right separately; pi, qi and ri are the three mutually perpendicular principal elastic axes of the ith mounting part, and stiffness of them are kpi, kqi and kri; \u03b8pi, \u03a6qi and \u03a8ri are the intersection angles between the three principal elastic axes and Ox, Oy and Oz of the ith mounting part" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001157_978-1-4471-2330-9_5-Figure5.1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001157_978-1-4471-2330-9_5-Figure5.1-1.png", + "caption": "Fig. 5.1 A sketch for the crank-slider mechanism", + "texts": [ + " The position of the ball is controlled on the plate by means of two independent actuators. The problem will include developing the equations of motion of the ball, constructing the virtual scene, developing the two PID controllers for the plate, and animating the ball and the plate. The electronic version of all the M-script fi les and VRML models in addition to the recorded movies for the animated crank-slider mechanism and the controlled motion of the ball on the plate can be downloaded from Springer\u2019s web site http:// extras.springer.com/ . Figure 5.1 shows a sketch for the crank-slider mechanism. Point O is the center of the crankshaft. ON is the crankshaft. Its length is 10 (cm).r = NP is the connecting rod. Its length is 20 (cm).l = Since P is constrained to move in the vertical direc- Chapter 5 Crank-Slider Mechanism of a Piston Matlab\u00ae and Simulink\u00ae are registered trademarks of The Mathworks, Inc. tion and the rest of the components are coplanar, this system has one degree of freedom which will be considered to be the crankshaft angle b" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002291_s1064230713020056-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002291_s1064230713020056-Figure5-1.png", + "caption": "Fig. 5. D method.", + "texts": [ + " D Method In the case of D method, there are bipolar pulses (UM = {+UPOW, \u2013UPOW} or UM = {\u2013UPOW, +UPOW}) on the winding or bipolar pulses with the back emf at the end of the PWM period (UM = {+UPOW, \u2013UPOW, } Tavg* \u03b3 \u03a9avg*t0/T.\u2013\u00b1= JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL Vol. 52 No. 2 2013 AUTOMATION OF CONSTRUCTION OF CHARACTERISTIC CURVES 261 or UM = {\u2013UPOW, +UPOW, }) depending on the load torque and the PWM duty cycle. Therefore, the expression for the speed\u2013torque characteristics of the D method (Fig. 5) is a combination of descriptions of the speed\u2013torque characteristics for impulse modes II and IV: 7.4. ND and AD Methods The possible presence of impulse modes I and V on the winding is a specific feature of the ND and AD methods. Therefore, the mathematical description of the speed\u2013torque characteristics of the ND and AD methods (Fig. 6) is a combination of the corresponding expressions for impulse modes I and V: 7.5. SN and SA Methods When the SN and SA methods are used, there may be impulse modes I\u2013IV on the winding depending on the load torque, the PWM duty cycle, and the sequence of alteration of voltage pulse signs within the PWM period" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001803_1.4031066-Figure13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001803_1.4031066-Figure13-1.png", + "caption": "Fig. 13 Direction of initial angular velocities _a0 and _c0, and setting of the initial precession angle", + "texts": [ + " Once the rotor enters the rolling mode, the condition may change to the uniform rotational motion. 5.1 Experiment Procedure. The experimental system used in this paper is almost the same as the one used in the previous paper [5] (Fig. 12). The procedure of the experiment is shown below: (1) Set the values of input angular velocity X and input amplitude H. (2) Pull the string of the power ball to give the moderate initial spin speed _c (2300\u20132400 rpm) to the rotor. Only the case with directions of _c and _a shown in Fig. 13 is chosen and observed. In this case, the side black pattern drawn is in the upper contact. (3) Observe the spin speed shown by the tachometer and start the input of sinusoidal motion to the motor with 061010-10 / Vol. 137, DECEMBER 2015 Transactions of the ASME Downloaded From: http://vibrationacoustics.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/journals/jvacek/934234/ on 02/08/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use the amplitude H and angular frequency X when the spin speed decreases and reaches _c0 (1900 6 50 rpm or 2000 6 50 rpm)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure6.10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure6.10-1.png", + "caption": "Fig. 6.10. Convergence to perceived depth", + "texts": [ + " The function of the observer\u2019s accommodative system is to manage the adaptation of the eyes\u2019 crystalline lenses to achieve sharp focus of light from the scene point onto the retina. The accommodative system is coupled to the convergence system because under natural viewing conditions the convergence distance as dictated by the intersection of look directions is equal to the optical distance to the scene point, i.e. for an angle of convergence \u03b1f , the accommodative system adapts the crystalline lenses to focus light from a distance of df = 1 2 IPD/ tan \u03b1f 2 , where IPD is the distance between the eyes\u2019 pupils, and \u03b1f is their angle of convergence, see Fig. 6.10. A problem arises in stereoscopic 3-D viewing if the perceived distance df does not match the optical distance do (c.f. Sec. 6.6.1). To avoid the retinal image becoming blurred, the natural accommodative response must be de-coupled from convergence. De-coupling introduces visual discomfort, loss of visual acuity, and after-effects. It is often cited as the main reason that stereoscopic 3-D displays become intolerable after long periods, e.g. [24]. A solution to the problem of accommodation-convergence conflict that has been suggested on several occasions, and evaluated most recently by Akeley et al" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003421_ilt-01-2011-0007-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003421_ilt-01-2011-0007-Figure1-1.png", + "caption": "Figure 1 Journal-bearing", + "texts": [ + " This is because in a bearing of finite width (z-direction) there is a large pressure gradient \u203ap/\u203az at right angle to the main motion in the u-direction, since the pressure falls to zero at the ends of the bearing. This pressure gradient gives rise to a leakage flow and a consequent loss of pressure, and hence load capacity in the bearing. Thus, we have to resort to three-dimensional lubrication theory, taking into account the shear-thinning variation of viscosity. The notations are the usual ones. The geometryof the journal bearing is shown inFigure 1.The difference between the radii of the shaft and journal is ho, and the inner surface turns at a speed v. Usually ho/R is 0(1023). The offset along the x-axis is 1 howhere 1 is the eccentricity ratio. The quantities of interest are the torque M to turn the inner cylinder and the x and y components of the resultant load generated by the motion. The problem was analyzed for the second-and third-order fluids for arbitrary 1 values when no cavitation occurs (Davies and Walters, 1973). Applications of this and similar analyses should be restricted to cases where the negative pressures are small compared with atmospheric pressure (Tanner, 1988)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002354_carpi.2014.7030038-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002354_carpi.2014.7030038-Figure9-1.png", + "caption": "Figure 9. Sensor set for monitoring column inclination", + "texts": [], + "surrounding_texts": [ + "The operation of the Elevator IV is monitored by a sort of sensors and controlled by a Programmable Logic Controller (PLC). Fig.5 shows the dashboard of the Elevator, including buttons for commanding basic movements. The Elevator is operated using this panel in an emergency. Under normal conditions, the Elevator is operated using a remote control unit (Fig.6) that the electrician carries inside the bucket. The remote control unit communicates with the PLC panel through radio frequency. The dashboard is the main control component and shows the operation status of the Elevator. It includes buttons for commanding: (a) simple movements, (b) pre-programmed tasks, (c) turn on and turn off the device and (a) an emergence stop. The simple movement buttons are: Front, Back, Right, Left, Up and Down. The pre-programmed task buttons are: Work and Rest. The buttons functions are as follow: Front: Incline the column toward rested position; Back: Incline the column back to the vertical position; Right: Rotates the column to the right; Left: Rotates the elevator to the left; Up: Moves bucket upward; Down: Moves bucket downward; Work: Automatically set the column to the vertical position and the bucket to the inferior limit position; Rest: Automatically moves the column to the horizontal transportation position (if the basket is empty and the column aligned with the vehicle main axis); Emergency: When pressed, the remote control unit is ignored, any automatic movement is stopped and commands of the control panel are followed ignoring sensor signals. The dashboard also contains a display where a sort of instructions and warnings are shown to the electrician. Fig.7 shows the logic control diagram of the Elevator. The core of the control system is a Programmable Logic Controller (PLC). The PLC receives command signals from the dashboard or from the remote control unit. It also receives signals of a sort of sensors that monitors the state of the Elevator. Based on received signals, the PLC executes a routine that verifies the safeness of the system and sends commands to the actuators through power electronic circuits: relays, in the case of electric motors and solenoid valves in the case of hydraulic actuators. So as to save electric power, the electric motor of the hydraulic pump unit is turned on only when a hydraulic actuator is required. All actuators are controlled in ON/OFF basis. Only the electrical actuator for elevation has an acceleration ramp, avoiding sudden motion that is uncomfortable to the electrician, and impacts of the bucket structure against mechanical stroke limits. At present, simultaneous motion in more than one D.O.F. is not allowed. The safeness check routine avoids various situations that bring risks of damages to the Elevator or injuries to the operator. The most important routine is one that avoids the fall of the entire elevator together with the vehicle. This verification is done based on the elevation, the inclination and the rotation of the column. At moment, the inclination of the vehicle is not monitored. In future works, also this parameter will be measured using an electronic level. Thus a more precise safeness check against falling will be possible. Fig.8 shows the approximate position of sensors. Figs. 9 and 10 give more details about the sensors for detecting, respectively, the column inclination and rotation. The Table II describes the function of each sensor. Although details are not shown, the control system of the Elevator IV also avoids vehicle moving while the Elevator is at working position. The vehicle can be moved only if the column is set and locked in horizontal position. III. TESTS Several tests were conducted to ensure the structural safety and correct operation of the Elevator IV. After that, the prototype was tested in field. Fig.11 shows the Elevator IV with a load of 120kg set in the bucket, the bucket elevated until the maximum reachable height and the column inclined until the maximum value. This is one of several critical conditions tested in the field. In all situations, no abnormal deformation of the structure was observed and vehicle with the Elevator kept the stability. No additional device was required to the vehicle suspension system to assure the stability. The Elevator IV is enough compact so that it was not necessary removing the cabinets to store tools and components for maintenance. Thus, the Elevator IV showed to be secure in mechanical aspects. Concerning the control system, exhaustive tests were conducted so as to confirm that the algorithm assures the safe operation of the Elevator avoiding accidents such as fall of the entire structure, run of the vehicle with the Elevator in the working position, unexpected movements, etc. Also, the behavior of the control system was verified to overcome emergency situations. For example, when \"Emergency\" button is pressed during the execution of the \"Rest\" automatic operation described above. In such situation, all interlocks are overridden and all motions can only be commanded manually through dashboard buttons. Therefore the Elevator can be carefully conducted to a safe situation. IV. CONCLUSIONS The CPFL and the Escola Politecnica of Sao Paulo University jointly conducted a project aiming improvements in the working conditions of the electricians in the maintenance of electric power distribution network. By applying Mechatronics and Robotics, the objective was to develop devices that reduce risks of immediate accidents or long-term injuries to electricians. Based on an ergonomic study, the project focused two activities: the tree trimming and the work at high positions. One of results is the automatic elevator that helps electricians to conduct maintenance of power distribution lines at high positions. In a past work, authors presented the most recent version of the automatic elevator, the Elevator IV, describing its mechanical aspects and load tests demonstrating the effectiveness and the safeness of the device. In addition, this work described the control system of the Elevator IV. Exhaustive situations were simulated in the Elevator IV in field tests. In any situation, the control system assured the stability of the device and the safeness of the electrician. The development of the Elevator IV concludes a cycle of studies concerning improvements in the working conditions of electricians in the maintenance of distribution power lines. Prototypes of all developed devices are now being tested in field so as to effectively introduce them to the maintenance activity. The key point in the development process of all devices was the correct definition of the automation level of each device. This depends not only on the access to the most advanced Mechatronics and Robotics technology and the cost. It depended on the long-term policy of the enterprise concerning human resource, the level of satisfaction of electricians and other peculiarities of the Brazilian power distribution system. Because of that, solutions adopted in other countries may not be effective in Brazil." + ] + }, + { + "image_filename": "designv11_84_0002176_amm.712.81-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002176_amm.712.81-Figure4-1.png", + "caption": "Fig. 4. Image of breaking point of the gear wheel", + "texts": [ + " Table 3. The chemical composition of the phases in the sintered material, % wt. Element Cu [%] Ni [%] Mo [%] Fe [%] Light phase 2.85 4.27 - rest Dark phase - - 1.65 rest Fig. 3. The microstructure of the steel 42CrMo4 samples (without coating), nital etched Based on the analysis, it was found that the products do not comply the technical specifications for static strength tests. Breaking point of the gear wheel made of sinter material which was identified during static strength test is shown in Figure 4. Hardness measurement points shown in Fig. 5. In addition, was carried out statistical analysis of the strength of gear wheels made of two materials analyzed to illustrate the strength of the test samples for a period of two years for sinter Sint - D32 (Fig. 6). For sinter material were also statistically evaluated static strength results for the batch of products among which there were identified non-conformed products (cracking during static strength test) for period of three months (Fig. 7). For comparison, performed statistical analysis of the static strength of gear wheels made of steel 42CrMo4 (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002494_amr.1028.105-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002494_amr.1028.105-Figure6-1.png", + "caption": "Fig. 6 Vibration mode of order 1", + "texts": [ + " The static structural analysis of the drive shaft is used for pre-stressed modal calculation in Model module, therefore the natural frequencies and corresponding vibration modes of the shaft can be got. Normally we do not have to find all the natural frequencies and mode shapes. Low order natural frequencies and vibration modes have bigger impact on the vibration of the stepper motor [4],so the first six natural frequencies and vibration modes obtained in ANSYS Workbench are concerned, as shown in Figure 6-11 and Table 1.Because the amplitude is relative value after treatment, it doesnot reflect the actual amplitude [3]. 1) Natural frequency analysis. Through access to relevant information, the stepper motor has a fixed resonance region. The resonance region of the two-four phase stepper motor is generally between 180 and 250PPS (step angle of 1.8 degrees). As the drive voltage of the stepper motor is higher and the load is lighter, the resonance region is upward. However, each order natural frequency of the drive shaft is more than 1745.2HZ, so to ensure the tube reaches the specified location accurately and avoid resonance between the stepper motor and drive shaft, the operating frequency of the stepper motor must be between 300PPS and 1700PPS. 2) Vibration mode analysis. Though the analysis of the first two modal shapes (shown in Figure 6, 7), the central part of the shaft has the greatest amplitude of the resonance, therefore probably becomes the weakest portion. So without affecting the transmission accuracy, belt pulleys are arranged on both sides of the shaft as far as possible. This paper is based on Solidworks to establish the 3 dimensional solid model of the shaft and uses ANSYS Workbench to analyze pre-stressed model of the shaft. Pre-stressed modal analysis is closer to reality. According to the results of the analysis, the reasonable frequency of the stepper motor can be presented, thus avoiding the generation of resonance" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001952_robio.2011.6181704-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001952_robio.2011.6181704-Figure7-1.png", + "caption": "Fig. 7. Left: A sketch of LAURON\u2019s legs from three points of view (top, front, left side). The leg marked in red is chosen for the construction of the matrix. Right: sketch of a leg with indicated joint configuration. This particular leg is marked with the red circle in the left-hand side sketches.", + "texts": [ + " The matrix is constructed in columns: each of the n joints creates one column ji depending on it\u2019s joint axis zi and the vector pi from the joint to the footpoint: ji = ( zi \u00d7 pi zi ) (12) Where zi represents the rotational axis of the joint i and the vector pi point to the corresponding footpoint i. J(CJ) = ( j1, ..., jn) (13) The resulting Jacobian matrix J(CJ) has 6 rows (for the 3- dimensional case) and n columns. Obviously the complexity of this construction rises linear with the number of joints. In the following we will create the Jacobian for an exemplary joint of an exemplary joint configuration CJ of LAURON as sketched in Fig. 7. We need one of these Jacobian matrixes for each leg of the robot. These are defined for the kinematic chain running from the hip (the alpha joint - see Fig. 3) to the footpoint of the same leg. To include the 3 degrees of freedom of the leg we need 3 columns of the Jacobian. First of all the vectors pi and zi are created corresponding to the sketch from Fig. 7 which then in turn give us the columns of H F J(CJ). z0 = \u239b \u239d 0 0 1 \u239e \u23a0 p0 = \u239b \u239d 0 y0 \u2212z0 \u239e \u23a0 z1 = \u239b \u239d 1 0 0 \u239e \u23a0 p1 = \u239b \u239d 0 y1 \u2212z1 \u239e \u23a0 z2 = \u239b \u239d 1 0 0 \u239e \u23a0 p2 = \u239b \u239d 0 y2 \u2212z2 \u239e \u23a0 z \u00d7 p0 = \u239b \u239d \u2212y0 0 0 \u239e \u23a0 z \u00d7 p1 = \u239b \u239d 0 z1 y1 \u239e \u23a0 z \u00d7 p2 = \u239b \u239d 0 z2 y2 \u239e \u23a0 (14) H F J(CJ) = \u239b \u239c\u239c\u239c\u239c\u239c\u239c\u239d \u2212y0 0 0 0 z1 z2 0 y1 y2 0 1 1 0 0 0 1 0 0 \u239e \u239f\u239f\u239f\u239f\u239f\u239f\u23a0 (15) By adding the individual 6D force and torque vectors acquired from the six individual limbs, while taking their corresponding point of application into account, we gain the resulting 6D force and torque vector for the robot\u2019s body" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002001_robionetics.2013.6743610-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002001_robionetics.2013.6743610-Figure9-1.png", + "caption": "Fig. 9. The geometry of the Freehand Robot [8]", + "texts": [ + " The robot is a 2 rotational link assembly with a drop-and-pick pen as an end-effecter. The robot is shown in Fig. 8. The freehand drawing robot is constructed using Aluminum 2\" square pipes. The first link L 1 is 8 cm and the second link L2 is 7 cm. Both links are rotational and actuated by dual bearing, metallic geared Power HD-1501MG - Standard Servo motors. The motors have a torque of 15-20 Kg-cm. They operate on analog pulse width modulation with 4. 8 - 6.0 volts. The motors have an operating range of 0-180\u00b0. The geometry of the robot assembly is shown in Fig.9 [8]. The coord inates (xeff, Yeff) represent the location of the pen or marker. The angles 8} and 82 represent the angles of the 2 links. The geometry is kept in such a way that the rotation is of \u00b1 90\u00b0 from the reference. The normalized workspace (inverted) of the robot arm is shown in Fig. 10 [8]. Figure lO. Workspace of two link robot [8] As soon as the points are extracted from the image, angles 8} and 82 are calculated using the inverse kinematics of two link robot. The feedback provides the actual position of the pen which is used to calculate the angles acquired using forward kinematics" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure6.11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure6.11-1.png", + "caption": "Fig. 6.11 Definition of structural loads", + "texts": [], + "surrounding_texts": [ + "the formal exploration given by the algorithms yielded by the explained parameters and reduce the objects mass through the same algorithms that will form the connections between the green areas to be preserved, see Fig. 6.12. 7. Once the manufacturing methods are established it is important to highlight that the generative design it\u2019s not only to be used with additive manufacturing since there are other methods such as subtractive numerical control with 2 to 3 axis in which parameters could be established for a cutter, see Fig. 6.13. 6 Numerical Simulation of Cranial Distractor Components \u2026 171" + ] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.21-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.21-1.png", + "caption": "Fig. 7.21 Case 2 of serial singularity: in that case, the actuator being fixed, the leg gains one internal mobility and the motion of the passive joints does not lead to a platform motion", + "texts": [ + " 7.20). We will call them Leg Active Joint Twist System (LAJTS) singularities. \u2022 Case 2: the sub-system 0Jdi is rank-deficient (this also corresponds to the degeneracy of the matrix Jtdi in (7.87) and, as a result, to the degeneracy of the matrix Jtd in (7.91))\u2014in that case, a displacement of the passive joints of the leg does not necessarily bring a displacement of the end-effector along one given direction. Moreover, in such configuration, the leg instantaneously gains an uncontrolled motion (Fig. 7.21). Later in the book, we will call them Leg Passive Joint Twist System (LPJTS) singularities. \u2022 Case 3: the system [ 0Jai 0Jdi ] is rank-deficient, while the systems 0Jai and 0Jdi are not\u2014in that case, a displacement of any types of the leg joint does not necessarily bring a displacement of the end-effector along one given direction (Fig. 7.22). Such singularities are usually similar to the Type 1 singularities (see Sect. 7.5.1). It will be shown later that the degeneracy of the system 0Jdi lead to the degeneracy of the PKM dynamic model" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002135_15502287.2013.833996-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002135_15502287.2013.833996-Figure6-1.png", + "caption": "FIG. 6. Hand made Microsystems wings. (Color figure available online.)", + "texts": [ + " The reason for reduced tensile properties at higher temperatures is the melting of nanofibers. Nylon-66-based nanopaper and reinforced nanopaper were used for different applications, such as fabrication of microsystems wings, miniaturized embedded antenna, and flexible solar cells applications. Results of insect wing components are presented here. An attempt was made to build wings using a pultruded carbon rod/fiber as grid and electrospun nanopaper as membrane. A schematic of the structural grid in% of fibers is shown in Figure 6a starting with 1k and 3k tow. The grid structure was constructed using DP 460 resin and pultruded carbon rods were used at the root of the tearing edge to provide additional stiffness. The grid structure was then smeared by resin using a suitable brush. The carbon tow and rods were then bonded together by curing the resin coated grid at atmosphere for three hours. After curing, the excess resin was trimmed. Figure 6b shows the picture of the bare wing. In the final step, the electrospun Nylon-66 nanopaper was fused to the wing by the process discussed in Section 2.2 without applying pressure on the vacuum bag. The excess nanopaper was trimmed. Figure 6c shows the photograph of the nanopaper-fused wing. Wings were also constructed with carbon-fabric-reinforced nanopaper as membrane. These wings were constructed initially; fusing was done without applying pressure on the vacuum bag, which resulted in improper adhesion. The nanopaper was used to fabricate several insect wing configurations. They are FlyTech dragonfly wing (Figure 7a) and Daedalus flight system wing (Figure 7b) fabricated with carbon-reinforced nanopaper. The flyability and durability of the reinforced nanopaper were demonstrated with remote control operation of the FlyTech dragonfly" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002868_s12206-015-0143-9-FigureA.1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002868_s12206-015-0143-9-FigureA.1-1.png", + "caption": "Fig. A.1. Thrust and torques of a quad rotor.", + "texts": [ + " Taguchi, Two-degree-of-freedom PID controllers, International Journal of Control Automation Systems, 1 (4) (2003) 401-411. [24] K. J. Astrom and T. Hagglund, PID Controllers: Theory, Design and Tuning, Second Ed. Instrument Society of America, Research Park, USA (1995). [25] K. J. Astrom and T. Hagglund, The future of PID control, Control Engineering Practice, 9 (11) (2001) 1163-1175. [26] A. Visioli and Q. C. Zhong, Control of integral processes with dead time, Springer-Verlag London Limited, London, UK (2011). Appendix A.1 Equation of motion Fig. A.1 shows the thrusts and torques acting on a quad rotor configuration and Fig. A.2 shows those acting on a quad tilt configuration. Two configurations are the same if the tilt angle \u03b3 is zero degree, thus all equations are equal for VTOL mode and forward flight mode by substituting \u03b3 = 0\u00b0. The body fixed reference frame B: ( bO , bx , by , bz ) and the earth fixed inertial reference frame W: (O, x, y, z) are shown in Fig. A.1, where the Euler angles \u03d5, \u03b8 and \u03c8 are called the roll, pitch, and yaw angles, respectively. Four motors create thrusts ( 1T ~ 1T ) and torques ( 1Q ~ 4Q ) with arm length sl . When the nacelles are tilted, and if their weights are not evenly balanced around their pivoting point, the forward and backward arm lengths may not be the same because of the change in the CG position. Therefore, in a quad tilt configuration, the forward arm length T Fl and backward arm length T Bl are separately illustrated" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003348_detc2014-34040-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003348_detc2014-34040-Figure1-1.png", + "caption": "Figure 1. A spur gear pair model.", + "texts": [ + " Under combined internal and external periodic excitations, the multi-valued properties and jump phenomena occur not only in primary resonance frequency but also in super harmonic frequency, the excitation force amplitude has less influence on the nonlinear dynamic characteristics and the increase of the excitation force amplitude could no longer control the nonlinear vibration of gear system. A two-degree-of freedom semi-definite model of the spur gear pair with periodic time-varying stiffness, backlash, and static transmission error as well as torque fluctuation as shown in Figure 1 is considered in this paper. The shafts and bearings are assumed to be rigid. All the symbols used are given in the Nomenclature. The differential equation of torsional motion of gear pair can be written as 1 1 1 1 1 2 2 1 1 1 2 2 1 b b b b b b I cr r r e t r k t f r r e t T t (1a) 2 2 2 1 1 2 2 2 1 1 2 2 2 b b b b b b I cr r r e t r k t f r r e t T t (1b) where T1(t)=T1m+T1a(t) and T2(t)=T2m+T2a(t) are torques on pinion and gear, f is a non-analytical function essentially describing the nonlinear elastic restoring force of gear pair" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003888_j.ijleo.2014.01.021-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003888_j.ijleo.2014.01.021-Figure2-1.png", + "caption": "Fig. 2. Schematic illustrations of three test parts.", + "texts": [ + " Six different pH alues of solution (pH 10.5, pH 11.68, pH 11.5, pH 12 and pH 12.5) ere first used to investigate the linear regression equation for preicting the pH value of solution. Then, five different pH values of olution (pH 11.13, pH 11.68, pH 12.21, pH 12.34 and pH 12.39) ere used to investigate the relative error of linear regression equaion. Finally, three test parts were used to investigate the efficiency f removing support material from RP parts using pH value comensation technology. Fig. 2 shows the schematic illustration of hree test parts, which was designed by Pro/ENGINEER software. he model was then exported to the FDM QuicksliceTM software Please cite this article in press as: C.-C. Kuo, Y.-T. Siao, On-line pH va process, Optik - Int. J. Light Electron Opt. (2014), http://dx.doi.org/10. ia the stereolithography format. Fig. 3 shows the designed three est parts with given dimensions. Table 1 shows the parameters for hree test parts. Fig. 4 shows the slicing results of three test parts sing the CatalystEX slicing software" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002766_qr2mse.2013.6625903-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002766_qr2mse.2013.6625903-Figure3-1.png", + "caption": "Figure 3. Accelerometer mounting locations on the rear pmg housing.", + "texts": [ + " Two mechanical faults were simulated on the test rig as follows: (1) a ball bearing outer race defect was deliberately introduced, and (2) a rotor imbalance fault was simulated by attaching a mass block to the surface of the load disc. The general arrangement of main equipment is illustrated in Fig. 2. 978-1-4799-1014-4/13/$31.00 \u00a92013 IEEE 1698 2.25 kHz piezoelectric single-axis accelerometers were used due to their wide frequency range and suitability for limited space inside PMG. The four (4) accelerometers were placed as shown in Fig. 3. The results verify that the best location was location 3, which is immediately below the bearing housing. The signal power is the largest at this location, which makes intuitive sense because this is inside the load zone. The vibration signals collected at the location 2 and 4 were random signals, which meant these two vibration signals was dampened out and can\u2019t be used for signal processing. Adhesive is a simple and effective means of mounting vibration transducers. Wax was used to mount the 4 accelerometers at the mounting locations. Small flat surfaces were milled at mounting location 1 and 3 in Fig. 3. Extra electric tape was wrapped around the outside rim of the bearing housing. For more security, a pipe clamp was used to fasten the accelerometers. IV. PRELIMINARY ANALYSIS The two mechanical \u201cfaults\u201d listed earlier, were simulated together. There are formulas [6] that can be used to calculate bearing characteristic frequencies. An even simpler rule of thumb is that the frequency of ball passing outer race is approximately 40% of the number of rolling elements multiplied by the shaft rotating frequency: 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003963_detc2013-12427-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003963_detc2013-12427-Figure1-1.png", + "caption": "FIGURE 1. THE REFERENCE CONFIGURATION (THIN LINE) AND THE RING CONFIGURATION AT TIME t (THICK LINE) WITH THE ASSOCIATED BASE LINES.", + "texts": [ + " It is shown that, in agreement with the results obtained for purely flexural modes, the extensibility still leads to softening extensional-flexural modes of linearly elastic rings. The derivation of the geometrically exact equations of motion in space of elastic rings undergoing extension, shear, and bending is discussed following [10, 11]. Let {e1,e2,e3} be a fixed right-handed orthonormal basis for Euclidean 3-space, with (e1,e2) being the plane in which the planar base curve of the ring lies in its natural (stress-free) configuration (see Fig. 1). The orientation of the cross section in the reference and in the current configurations is described by the intrinsic frames {bo 1(s),b o 2(s),b o 3(s)} and {b1(s, t),b2(s, t),b3(s, t)}, respectively. The unit vectors bo k can be expressed in terms of ek as bo k = Ro(s) \u00b7ek where the orthogonal tensor Ro(s) describes the rotation \u03b8 o(s) about e3. On the other hand, the rotation of bo k into bk can be described by the incremental rotation tensor R(s, t) according to bk = R \u00b7bo k. The arclength parameter s along the base curve is employed to describe the position of material cross sections of the ring", + " If the base point is taken to coincide with the center of mass of the cross section and (b2,b3) are collinear with the principal axes of inertia, then: \u03c1i = o and \u03c1J becomes diagonal. The constitutive functions for the generalized stress resultants, expressed as N (s, t) 7\u2192 \u02c6N (\u03bd ,\u03b72,\u03b73,\u00b51,\u00b52,\u00b53, \u03bd\u0307, \u03b7\u03072, \u03b7\u03073, \u00b5\u03071, \u00b5\u03072, \u00b5\u03073), etc., are assumed to have as many derivatives as appear in our analysis. When the ring undergoes planar motions, major simplifications arise, namely, bo 3 = b3 = e3, the vector b1 makes the angle \u03b8 with bo 1 (see Fig. 1) and the rotation matrices reduce to the following expressions Ro(s) = cos\u03b8 o sin\u03b8 o 0 \u2212sin\u03b8 o cos\u03b8 o 0 0 0 1 , R(s, t) = cos\u03b8 sin\u03b8 0 \u2212sin\u03b8 cos\u03b8 0 0 0 1 . (7) As a consequence of the planar motion assumption, \u03b73 = 0 = \u00b51 = \u00b52, H3 = 0 =M1 = M2 and the displacement vector reduces to the form u(s, t) = u(s, t)bo 1 + v(s, t)bo 2. The stretch vector and flexural curvature vector are thus expressed as \u03bd := \u03bd b1 +\u03b7 b2 and \u00b5 = \u00b53 = \u03b8 \u2032 where, here and henceforth, the prime indicates differentiation with respect to s" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.7-1.png", + "caption": "Fig. 2.7. Successive rotations of an object about axes of fixed frame", + "texts": [ + " In view of this, it can be concluded that two rotations in general do not commute and its composition depends on the order of the single rotations. Example 2.3 Consider an object and a frame attached to it. Figure 2.6 shows the effects of two successive rotations of the object with respect to the current frame by changing the order of rotations. It is evident that the final object orientation is different in the two cases. Also in the case of rotations made with respect to the current frame, the final orientations differ (Fig. 2.7). It is interesting to note that the effects of the sequence of rotations with respect to the fixed frame are interchanged with the effects of the sequence of rotations with respect to the current frame. This can be explained by observing that the order of rotations in the fixed frame commutes with respect to the order of rotations in the current frame. Rotation matrices give a redundant description of frame orientation; in fact, they are characterized by nine elements which are not independent but related by six constraints due to the orthogonality conditions given in (2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002749_ever.2014.6844133-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002749_ever.2014.6844133-Figure2-1.png", + "caption": "Fig. 2 Pulsating voltage signal injection.", + "texts": [ + " Moreover, (2a) shows the negative-sequence carrier component which has contained the rotor position information in its phase angle. In the synchronous reference frame with the estimated carrier frequency, the negative-sequence carrier current response can be represented as _ _ cos(2 ) sin(2 ) \u03b8 \u03b8 \u23a1 \u23a4 \u0394\u23a1 \u23a4 =\u23a2 \u23a5 \u23a2 \u23a5\u0394\u23a2 \u23a5 \u23a3 \u23a6\u23a3 \u23a6 neg d n nneg q i I Ii (3) Similarly, a high-frequency pulsating carrier voltage signal (4) is injected into the d-axis estimated synchronous reference frame as two superposition rotating carrier vectors with opposite direction as shown in Fig. 2. cos , 0 dh c c qh e e v V t v \u03b1 \u03b1 \u03c9 \u03d5 \u23a1 \u23a4 \u23a1 \u23a4 = = +\u23a2 \u23a5 \u23a2 \u23a5 \u23a3 \u23a6\u23a2 \u23a5\u23a3 \u23a6 (4) With the aid of d-axis pulsating voltage signal injection on the estimated synchronous reference frame, the high-frequency carrier current response can be given by cos(2 ) sin sin(2 ) dh p n m n mqh e e i I I Ii \u03b8 \u03b8 \u03b1 \u03b8 \u03b8 \u23a1 \u23a4 + \u0394 +\u23a1 \u23a4 = \u22c5\u23a2 \u23a5 \u23a2 \u23a5\u0394 +\u23a3 \u23a6\u23a2 \u23a5\u23a3 \u23a6 (5) It is clearly described that the carrier current response signal is basically amplitude modulated by the rotor position information (cross-saturation angle \u03b8m is constant)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001312_icuas.2015.7152309-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001312_icuas.2015.7152309-Figure10-1.png", + "caption": "Figure 10: CFD simulation of pressure contours around the device body", + "texts": [ + " Thus a righting moment is achieved to return the device back to zero angle of attack. By altering the \u03b2d, this point will change to a non-zero value, causing the device to stabilize at a rolled angle; a lift force is induced, creating a horizontal component of velocity. This method is used to maintain separation with the skydiver during free fall. The point where each curve intercepts the horizontal axis, correlates to these new angles of attack. The moment coefficient is taken from the center of mass. Figure 10 shows the pressure contours around the device. The size of the pressure lobes on either side of the device are similar in magnitude as \u03b2d is employed, showing that the device is close to its new equilibrium point. Lift Behavior Lift coefficients follow a positive gradient (figure 13), with respect to angle of attack, a similar behavior to a symmetrical airfoil with a flap, under the same flow. The coordinate system for measuring lift is set to positive lift in the upward direction, with positive \u03b2d also in this direction" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003671_detc2011-48794-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003671_detc2011-48794-Figure6-1.png", + "caption": "Figure 6. Penetration of the rollers into the toroidal disks resulting in a contact ellipse.", + "texts": [ + " Figure 5 shows a sketch of the system along with the major system dimensions. Table 2 shows the values of the dimensions, mass/inertia and discretization parameters of the toroidal drive model. In this system the input/output speed ratio can be adjusted from 0.5 to 2.0 by changing the roller angle \u03c6. In this simulation the speed ratio is kept constant and set to 1, i.e. \u03c6 = 0. There is a very small penetration of the rollers into the toroidal disks. This penetration results in an elliptical contact patch between the roller and the toroidal disk (Fig. 6). The penetration is set such that the maximum normal contract stress at the center of the contact ellipse is 2.6 GPa. This normal stress is calculated using: dk p=\u03c3 where d is the penetration between the roller and the toroidal surface at the center of the torus. This means that for this model the penetration d is: 0433.0== pkd \u03c3 mm Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use 7 Copyright \u00a9 2011 by ASME This penetration results in the contact ellipse shown in Fig. 6. The total normal force between the roller and the disk is given by: \u2211\u222b \u2192= == ni iipN AdkdAF 1AreaEllipse \u03c3 This integration is numerically performed by summing the normal contact penalty forces from n contact points on the roller which are on the contact ellipse (Fig. 6), where Ai is the contact area and di is the penetration, associated with contact point i. This normal force model closely mimics Hertz contact theory, where the two surfaces deform and form a contact ellipse at the contact interface that is approximately the same as the contact ellipse that is based on geometry of the roller and the toroidal disk shown in Fig. 6. The material stiffness is modeled using the normal penalty springs. In our model FN evaluates to 24750 N or about 2.5 tons. Note that this is a very high normal contact force. The normal contact force between the roller and the disk in toroidal CVTs is normally around 2.5 tons and can reach a maximum of 10 tons. The very high normal contact force is needed to enable the transfer of large torques. The simulation starts by ramping up the input shaft angular velocity from 0 to 2000 RPM in 1 sec, then it is kept constant for the rest of the simulation (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001216_icit.2015.7125200-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001216_icit.2015.7125200-Figure5-1.png", + "caption": "Fig. 5. Result of magnetic field analysis", + "texts": [ + " Existence of the q-axis bridges Even though the permanent magnet buried shallowly, if the q-axis bridges between magnets are made as shown in Fig.3, the decrease of q-axis inductance caused by the magnetic saturation is relieved because q-axis flux flows enough as shown in Fig.3. The effect of making the q-axis bridges is shown by magnetic field analysis using FEM. TableI shows the specification of the motor model for magnetic field analysis. Fig.4 shows the motor model. When the U-phase and V-phase currents flow as shown in Fig.4, in case of the rotor position shown in Fig.4, the q-axis flux is maximum. Fig.5(a),(b),(c) show the results of the magnetic field analysis at 3A when the width of the q-axis brigdes tq is 3mm,5mm,7mm, respectively. Fig.5(a),(b),(c) show that the magnetic field saturation occur at the part [A] of Fig.4 at any tq because the permanent magnet buried shallowly. However, Fig.5(a),(b),(c) show that q-axis flux flows enough through the part [B] of Fig.4 with increasing tq . Therefore, the existence of the q-axis bridges cause that the q-axis flux flows enough as shown in Fig.3 even though the permanent magnet buried shallowly. The sound noise caused by torque ripple increases when the permanent magnet is buried shallowly because of the shortage of the flux due to permanent magnet \u03d5s which flows the teeth of the stator as shown in Fig.6 Therefore, the grooves of the rotor are made as shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003912_detc2013-12420-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003912_detc2013-12420-Figure10-1.png", + "caption": "FIGURE 10. INITIAL CONFIGURATION FOR A SIX-DOF NEWTON\u2019S CRADLE.", + "texts": [ + " Since most notable works neglect friction in the model, it is difficult to present a comparison of the result from this case. However, the solution demonstrates that the effect of differing friction properties between pairs of balls in the chain affects the motion obtained. It is possible to adjust the friction coefficients to match results obtained in experiment. This is not explored further in this work. Case III: Six-DOF with e\u2217 = 1 and \u00b51/\u00b52 = 1 A six-DOF Newton\u2019s cradle is now considered as shown in Fig. 10 using the proposed approach. This system is composed of three balls as presented for the three-DOF case in Fig. 1 but has six-DOFs. The three additional DOFs are included in the model to allow each ball to rotate freely with respect to the massless string it hangs from. Similar to the three-DOF cases, Ball A is released from rest at an arbitrary height and impacts Balls B and C, which are initially at rest and in contact. The first case examined for this model considers balls with 6 Copyright c\u00a9 2013 by ASME Downloaded From: http://proceedings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003434_12.2013591-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003434_12.2013591-Figure1-1.png", + "caption": "Figure. 1. Scheme of the laser dispersing process.", + "texts": [ + " However, the relatively short life-time of bulk light metals due to the disadvantageous wear, seizure and friction properties, still limits their wider application. In this case, a significant increase of wear resistance can be obtained by laser dispersing (known also as laser melt injection or laser embedding) of the ceramic, hard particles into the surface layer of the substrate material. In the process, the surface of the material is locally melted by the defocused laser beam and simultaneously the powder particles are injected into the molten material (Fig. 1). In results, the metal matrix composite (MMC) surface layer is formed with powder particles distributed in the top-layer of substrate material which serves as matrix, and of improved properties in comparison to the bulk material 1-7. Due to the flexibility of the laser treatment the MMC surface layer can be created at selected locations where the increased wear resistance is required. Moreover, the laser dispersing allows also for production of functionally graded materials (FGM) of spatially inhomogeneous properties and without sharp interface typical for laser cladding 1,4,8", + " The known problems appearing during laser dispersing in light-metal alloys are related to the low laser radiation absorption coefficient of that materials at wide range of wavelengths including the most common used laser sources. On the other hand, laser beam absorptivity of carbides is mostly much higher, but to avoid formation of undesirable phases, the powder particles should not be dissolved by the laser beam. Therefore, a lateral powder nozzle is preferred, because it allows for reduced contact between the powder particles and the laser beam, especially when the powder jet is situated behind the laser beam spot (see Fig. 1). However, in this case the laser formed molten pool should be extended behind the laser spot, and such situation can take place in the case of materials of relatively low thermal conductivity, e.g. Ti-6Al-4V alloy investigated here. Finally, usually poor wetting properties between carbide particles and the considered substrates can cause problems with injection of the powder to the matrix1,2. * rafj@imp.gda.pl; phone 48 58 6995193; fax 48 58 3416144 Laser Technology 2012: Applications of Lasers, edited by Wieslaw L" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002344_detc2013-12837-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002344_detc2013-12837-Figure7-1.png", + "caption": "Figure 7. SOLID MODEL OF A 30-TEETH MODIFIED CURVILINEAR PINION", + "texts": [ + " The unit normal vector of the generated modified curvilinear gear represented in coordinate system S2 was also obtained as follows: GG GGG GGG z y x n n n sincos coscoscossinsin sincoscoscossin 22 22 2 2 2 2n . (10) Geometry of the Modified Curvilinear Gear Generated by a Male Fly Cutter In this study, both pinion and gear of the modified curvilinear gear set are generated by a male fly cutter with a circular-arc normal section. Therefore, the tooth geometry of the pinion and the gear is the same. Figure 7 demonstrates a solid model of a 30-teeth modified curvilinear pinion with basic design parameters of RP = 500 mm and EP = 70 mm (Table 1). PRINCIPAL DIRECTIONS AND CURVATURES When a surface is generated by a rack cutter (generating) surface, the rack cutter surface and the generated tooth surface are in continuous tangency along a line during the generation process. In this section, the principal directions and curvatures of the generating surface according to the Rodrigues\u2019 formula are calculated first" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002011_2011-01-1691-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002011_2011-01-1691-Figure11-1.png", + "caption": "Figure 11. Vibration transmission characteristic by influence degree (Front pillar of passenger-side)", + "texts": [ + " The other paths whose vectors are almost orthogonal to the response are negligible to the front-side path. Therefore, the front-side path is focused to analyze. In order to analyze the front-side path effectively, the upper part of the body model is cut out from whole body model and is loaded on the cutting section as shown in Figure 10. The response of the cut body model is confirmed to be equal to that of the whole body model. For the reduction of the roof response, nine VT paths at the upper body structure shown in Figure 5 are focused. Figure 11, 12, 13 shows the VT characteristics that are calculated and visualized by the proposed techniques. Figures 11, 12 (a)-(d) and 13 (a) show the influence degree for translational forces in the directions of xyz-axis, and Figures 13 (b) shows the influence degree for moment around the yaxis because other influence degrees are so small to be omitted. As shown in Figure 13 (b), new local coordinates are defined for two-dimensional element: the normal direction is z-axis, the lateral direction is x-axis and the longitudinal direction is y-axis" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002344_detc2013-12837-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002344_detc2013-12837-Figure2-1.png", + "caption": "Figure 2. CIRCULAR-ARC NORMAL SECTION OF P FOR GENERATING MODIFIED CURVILINEAR PINION", + "texts": [ + " The proposed gear set is expected to exhibit point contact, and parabolic TE under ideal meshing condition. Figure 1(b) depicts the schematic formations of the rack cutter (generating) surface P . Herein, rack cutter surface P generates the involute pinion surface 1 , while surface G generates the modified circular-arc gear surface 2 . In the tooth contact analysis (TCA), the pinion surface generated by the left-side of rack cutter P is meshing with the gear surface generated by the right-side of rack cutter G . Equations of Rack Cutter P As shown in Fig.2, the normal section of the rack cutter surface P used for pinion surface generation is a circulararc. According to the coordinate relationship depicted in Fig.3, the position vector and unit normal vector of P represented in coordinate system ),,( )()()()( P c P c P c P c ZYXS are expressed as follows [16] PPP p nPP PPP P nPP nPP P c E 2 S R E 2 S [R R sinsin])cos(cos[ )cos1(cos])cos(cos )sin(sin )(R , (1) and PP PP P P c cos sin coscos sin )(n . (2) Parameters P and P are the surface parameters of rack cutter P " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002773_aim.2013.6584218-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002773_aim.2013.6584218-Figure1-1.png", + "caption": "Fig. 1. A typical robot-workcell setup", + "texts": [ + " This can be observed from the fact that several major robot-manufacturing companies such as ABB and KUKA have incorporated the motion/force control capability into their new product line in the past few years [3], [4]. With this additional capability, robots are now be able to handle more complicated tasks, like contact-type operations, such as polishing, grinding and so on. Since these tasks all require some level of interaction between the robot and the work-piece/environment, the 3D CAD model of the robot, its end-effector tools, the work-piece and the work station are now critical components as depicted in Figure 1. For example, the work-piece CAD model can be used to generate targets (Ttarget) along the desired path in such a way that the robots tool (TTCP ) can approach the work-piece surfaces *This work was supported by the research project (P12-R-024C) at SIMTech (Singapore Institute of Manufacturing Technology). Ngoc Dung Vuong1, Tao Ming Lim2 and Guilin Yang3 are with SIMTech, 71 Nanyang Drive, Singapore, 638075 (email: @SIMTech.a-star.edu.sg). at a certain angle depending on the process requirements" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001967_iet-cta.2014.0667-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001967_iet-cta.2014.0667-Figure1-1.png", + "caption": "Fig. 1 Single agent control for a multi-agent system", + "texts": [ + " As a matter of fact there have been steady attempts in control community in order to take advantages of powerful frequency domain analysis and design tools for dealing with linear multi-agent networked systems. For instances authors of [13] investigated on some performance limitations of linear multi-agent system in the frequency domain and the non-minimum property of a class of cyclic networked system was observed in [5, 14]. The essence of SAC scheme is that, for \u2018collective\u2019 stabilisation of a multi-agent system, it might be sufficient to stabilise only a single agent in the system by an external controller. This scenario is illustrated in Fig. 1 where a multi-agent system consists of eight identical agents and among them only one agent, labelled 1, is directly controlled by an exogenous controller. By the aforementioned collective stability, we mean the input\u2013output stability between the external reference and every agent output in {yi}. An initial idea of the SAC firstly appeared in [5] for a set of cyclic systems with identical integrator agents and later generalised to general undirected systems in [6] and cyclic (possibly directed) systems in [7]", + " On the contrary, in this paper, the SAC scheme is generalised to networked systems composed of \u2018arbitrary\u2019 linear dynamical agents and a \u2018general\u2019 linear controller. A key finding of this paper is that the SAC works for a wide class of linear multi-agent systems, including every undirected system, even without any consensus protocols. Furthermore, we will also substantiate that, for a generic multi-agent system, the SAC still works even when the output of a controlled agent is not measurable for the controller. Instead, either the sum or average of all agent outputs can be used for collective stabilisation under the SAC scheme, as illustrated in Fig. 1 with dashed lines. IET Control Theory Appl., 2015, Vol. 9, Iss. 6, pp. 929\u2013934 929 doi: 10.1049/iet-cta.2014.0667 \u00a9 The Institution of Engineering and Technology 2015 A set of polynomials {p1, . . . , pm} is called coprime if any two polynomials {pi, pj | j = i} are coprime. In addition, a polynomial is called \u2018monic\u2019 if its leading coefficient is equal to one. This paper considers a multi-agent dynamical system composed of identical agents whose dynamics is linear, time-invariant, possibly unstable and thus the agent dynamics can be described as a strictly proper rational transfer function yi(s) ui(s) = n(s) d(s) (1) where yi(s) and ui(s) denote the Laplace transform of agent output yi(t) and input ui(t)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003566_icems.2014.7013933-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003566_icems.2014.7013933-Figure3-1.png", + "caption": "Fig. 3. Torque-speed characteristics for operation point (P)", + "texts": [ + " 2 22 1m f f b bP ( s )( R I R I )= \u2212 \u2212 (2) 2 2 1 1Cus m m a aP r I R I= + (3) ( )2 2 22 2 22 ' m Cur mf a ' ' XR P I I R / s jX = + \u03b1 + + ( )2 2 2 2 22 m mb a ' ' X I I R / ( s ) jX + \u03b1 \u2212 + (4) ( )2 2 2 c Fe cf cb R P I I= + (5) In the event that the Rc in parallel as in Figure 2, the core losses equation can be written as ( )2 2 Fe Fe mf mbP K E E= + (6) where KFe is core losses coefficient, Ef and Eb are the amplitude of forward and backward induced voltage [7]. Note that in practice, there are other losses such as friction and windage losses and stray load loss which are not considered in the loss model. B. Effect of Low V/Hz on Motor Performance The torque-speed curves of adjustable speed induction motor drive with high and low V/Hz ratio is shown in Fig. 3. Each voltage-frequency pairing defines particular motor torque-speed characteristic passing through the specified operating point (P), but the efficiency may wary widely [8]. If the voltage is supplied to motor is high (high V/Hz) to find that the core losses is increased. But if the voltage supplied to the motor is low (low V/Hz), the core loss is decreased whereas the copper loss may rise unduly. Therefore, there is an optimum voltage and frequency (the core losses equal to the copper losses) which gives the specified torque and speed with maximum efficiency" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001889_dynamics.2014.7005647-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001889_dynamics.2014.7005647-Figure2-1.png", + "caption": "Fig. 2. The process of electro-mechanical processing 1 \u2013 Working tool; 2 \u2013 workpiece to be processed", + "texts": [ + " Thus, the total heat release\u0432 is: Q = Q1 + Q2 + Q3, (1) In most studies heat sources in calculations are considered flat, but in terms of thermo-physical analysis during schematization process heat sources must be considered threedimensional. Thus, the length of the arc corresponding to the length of the deformation area treated surface a working tool in the main plane can be divided into two portions A1E1 and E1B which correspond to the contact pads on the front and rear surfaces of the blade of the tool during machining. The working surface of the tool is in the form of a circular arc of radius r. Plot A1E1 is adjacent to the area in which the major work on the surface plastic deformation of the metal hsurf (Fig. 2). This plot is similar to the front surface of the tool blade. In the second part of the E1B tool interacts with a layer of elastic restoring helast workpiece material (Fig. 2). Plastic and elastic deformation of the material of the surface layer parts occurs in a certain volume AA1E1VE. In the production of a workpiece with a velocity V plastic deformation mainly undergoes volume AA1E1E, adjacent to the tool front surface. In the same volume there is elastic deformation. The volume of material adjacent to the rear surface of the instrument E1B mainly undergoes elastic deformation. Plastic deformation of the surface layer of the workpiece in this volume is negligible" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003777_amm.532.367-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003777_amm.532.367-Figure1-1.png", + "caption": "Figure 1 Structure of spiral groove DGS Figure 2 Geometry of mating ring face", + "texts": [ + " Influences of geometric parameters of seal face, namely, spiral angle, groove depth, groove-dam ratio and groove-land ratio, are analyzed and recommended values of the parameters are given, which one can refer to in turbine compressor DGS designs. All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 132.239.1.231, University of California, San Diego, La Jolla, USA-02/06/15,09:24:33) Structure of a typical spiral groove DGS is shown in Fig. 1. The mating ring is rigidly mounted to the shaft and rotates together with it. The primary ring is mounted flexibly to the housing by springs and O-rings and allow for small relative axial and angular motion for misalignment between the two parallel rings. While the DGS is operating\uff0ca quite thin gas film is formed between the rings as a result of hydrostatic and hydrodynamic effect. The gas film with stiffness supports the pressure load and keep the seal open. Geometry of the mating ring face is shown in Fig. 2. The seal face is divided into three parts, the groove region, the dam region and the land region. Firstly, the structure of DGS is simplified. As is shown in the Cartesian coordinate system in Fig. 3, the initial clearance between the primary ring and the mating ring is C0\uff0cwhich is equal to the film thickness h 0 at equilibrium condition. The tilts of primary ring about X and Y axes are X\u03b3 and Y\u03b3 . Figure 1 Kinetic model of DGS Assuming that the gas between seal rings is isothermal ideal gas. The fluid field of gas film is modeled using the compressible Reynolds equation ( )3 ph ph p 6 rph 12 t \u00b5\u03c9 \u00b5 \u2202 \u2207 \u22c5 \u2207 \u2212 = \u2202 , (1) where \u00b5 is gas viscosity and \u03c9 is rotating speed of shaft. The pressure distribution is obtained by solving the equation, and then is integrated over the seal face area to get the moments of gas film about the X and Y axes, M X and M Y . X X s X s X Y Y s Y s Y I M k c I M k c \u03b3 \u03b3 \u03b3 \u03b3 \u03b3 \u03b3 \u03b3 \u03b3 \u03b3 \u03b3 = \u2212 \u2212 = \u2212 \u2212 , (2) where sk and sc are angular stiffness and damping of springs and O-rings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001915_icnsc.2014.6819637-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001915_icnsc.2014.6819637-Figure4-1.png", + "caption": "Fig. 4. Manipulator adjustment", + "texts": [ + " And R?.Rmin\u2022 Evaluate the path L in Qb using the length and curvature of the path with the consideration of the constraints mentioned above, and the optimal path Lo can be represented as: s.t. v \ufffd V max (9) R?R Imn where kl is the weight of path length, and k2 is the weight of curvature of the path. 2) The posture adjustment o/the manipulator When the object is localized by vision, the manipulator should be adjusted in real time to lock the object in the center of the CMOS camera, as shown in Fig. 4. The angle change of the horizontal direction is \ufffd8s=atan2(P()m POly)-8s. The total angle change of the vertical direction is \ufffda=a '-a, and each joint change is based on the left margin amount and the inertia of each connecting rod. The left margin amount of each joint IS {82m:J[-82'!3m=-83,84:=-841\ufffda:o , and the adjust 82m - 82, \ufffdm -J[ +\ufffd, 84m -J[ + 841 \ufffda _ 0 amount of each joint is defined as (10) where Tm = I ke/J,m , kei is the coefficient related with the 1=2.3.4 inertia of each connecting rod" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001160_978-3-642-23147-6_13-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001160_978-3-642-23147-6_13-Figure1-1.png", + "caption": "Fig. 1. Illustration of riding an electric unicycle Fig. 2. Free-body diagrams of the body (including the rider) and the wheel", + "texts": [ + " The rest of the paper is organized as follows. Section 2 is devoted to establishing the dynamic model of the electric unicycle with one brushless motor. In Section 3, the sliding-mode controller is synthesized to achieve the design goals. Several simulations and experimental results are respectively performed in Section 4 and 5 to illustrate the effectiveness of the proposed control method. Section 6 concludes the paper. The section is aimed at deriving the mathematical model of an electric unicycle. As Fig.1 shows, the working principle of the unicycle is interpreted as follows. If the rider leans forward, the unicycle will move forward in order to maintain the rider\u2019s body without falling. Similarly, if the rider leans backward, then the unicycle will move backwards for balancing. After understanding the basic working principle of the vehicle, one desires to derive the dynamic model of the rider mounting the unicycle using Lagrangian mechanics. Note that the modeling process is based on a by two major components: the body and the wheel, as Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001649_icvrv.2014.24-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001649_icvrv.2014.24-Figure6-1.png", + "caption": "Figure 6. coordinate systems", + "texts": [ + " The top view of the tunnel is shown in figure 5, because there are certain changes in the horizontal and vertical directions, so we need to take both horizontal and vertical displacement into account during the motor process of the robot, which means we have to control operation process precisely. The space robot has six degrees of freedom, and moves by connecting rods, and it can be adapted to many kinematic models[5]. Because the shape of terminal probe is single, so we choose DH model. Since each degree of freedom revolves around a specific axis, so we choose rotation axis as Z axis, and each moving coordinate system of degrees of freedom is shown in figure 6. We can calculate kinematics model of the robot based on the connection rods coordinate systems. Transformation of connecting rods coordinate system {i} relative to connecting rods coordinate system {i-1} is represented as .Based on DH algorithm, we can have \u2212\u2212 \u2212 = \u2212\u2212\u2212\u2212 \u2212\u2212\u2212\u2212 \u2212 \u2212 1000 coscossincoscossin sinsincoscoscossin 0sincos 1111 1111 1 1 iiiiiii iiiiiii iii i i d d T \u03b1\u03b1\u03b1\u03b8\u03b1\u03b8 \u03b1\u03b1\u03b1\u03b8\u03b1\u03b8 \u03b1\u03b8\u03b8 (1) Transformation matrix of position of the robot\u2019s terminal relative to the foundation bed coordinate system is: 0 0 1 2 3 4 5 6 1 2 3 4 5 6T T T T T T T= (2) We can calculate the position of the robot\u2019s terminal relative to the origin of foundation coordinate system through the equations above" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002868_s12206-015-0143-9-FigureA.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002868_s12206-015-0143-9-FigureA.2-1.png", + "caption": "Fig. A.2. Tilt angle vs. ub.", + "texts": [ + " J. Astrom and T. Hagglund, PID Controllers: Theory, Design and Tuning, Second Ed. Instrument Society of America, Research Park, USA (1995). [25] K. J. Astrom and T. Hagglund, The future of PID control, Control Engineering Practice, 9 (11) (2001) 1163-1175. [26] A. Visioli and Q. C. Zhong, Control of integral processes with dead time, Springer-Verlag London Limited, London, UK (2011). Appendix A.1 Equation of motion Fig. A.1 shows the thrusts and torques acting on a quad rotor configuration and Fig. A.2 shows those acting on a quad tilt configuration. Two configurations are the same if the tilt angle \u03b3 is zero degree, thus all equations are equal for VTOL mode and forward flight mode by substituting \u03b3 = 0\u00b0. The body fixed reference frame B: ( bO , bx , by , bz ) and the earth fixed inertial reference frame W: (O, x, y, z) are shown in Fig. A.1, where the Euler angles \u03d5, \u03b8 and \u03c8 are called the roll, pitch, and yaw angles, respectively. Four motors create thrusts ( 1T ~ 1T ) and torques ( 1Q ~ 4Q ) with arm length sl " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001312_icuas.2015.7152309-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001312_icuas.2015.7152309-Figure5-1.png", + "caption": "Figure 5: Vane movement to induce a rolling moment on the device.", + "texts": [ + " A Pitching moment is induced by the retraction or extension of the two rear vanes, and the opposite movement of the forward vane (shown in figure 4). Thus the differential of lift and drag forces acting on the rear and fore of the device increases, causing the device to keel in the desired direction. To induce a moment in the roll axis, the vanes are adjusted by an amount \u03b2d\u03d5 which is mapped so that the device will roll in the direction of the retracted vane, and away from the extended vane, as is shown in figure 5. collective and differential vane deflections. By mechanism design, the mapping is effectively linear across the operating range, [\ud835\udefd1, \ud835\udefd2, \ud835\udefd3]\ud835\udc47 = \ud835\udc45[\ud835\udefd\ud835\udc50, \ud835\udefd\ud835\udc51\ud835\udf19 , \ud835\udefd\ud835\udc51\ud835\udf03] \ud835\udc47 (1) where R is a transformation matrix. III. AERODYNAMICS ANALYSIS Computational Fluid Dynamics (CFD) simulations in Ansys were used to understand the behavior of the device under the airspeeds experienced during a sky dive. The results were then parameterized to form functions which could be used in a simulation of the system. Figure 6 denotes the CFD domain used" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002198_amm.644-650.763-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002198_amm.644-650.763-Figure1-1.png", + "caption": "Figure 1.Geometric parameters of gear Figure 2 Normal gear meshing stiffness", + "texts": [], + "surrounding_texts": [ + "2.1 Calculation of time-varying meshing stiffness In this paper, the energy method is used to calculate the time-varying meshing stiffness of gear pair. First of all, gear stiffness the potential energy in the meshing gears can be divided into four components: Hertz energy \ud835\udc48\u210e , bending energy \ud835\udc48\ud835\udc4f , radial compression energy \ud835\udc48\ud835\udc4e and shear energy \ud835\udc48\ud835\udc60. The stiffness corresponding to each energy is respectively Hertz stiffness \ud835\udc58\u210e, bending stiffness \ud835\udc58\ud835\udc4f , radial compression stiffness \ud835\udc58\ud835\udc4e and shear stiffness \ud835\udc58\ud835\udc60 .These stiffness combined together in series becomes the meshing stiffness of gear tooth. The stiffness corresponding to each energy can be respectively expressed as: \ud835\udc58\u210e = \ud835\udf0b\ud835\udc38\ud835\udc3f 4(1\u2212\ud835\udc632) (1) 1 \ud835\udc58\ud835\udc4f = \u222b 3{1+cos \ud835\udefc1[(\ud835\udefc2\u2212\ud835\udefc) sin \ud835\udefc\u2212cos \ud835\udefc]}2(\ud835\udefc2\u2212\ud835\udefc) cos \ud835\udefc 2\ud835\udc38\ud835\udc3f[sin \ud835\udefc+(\ud835\udefc2\u2212\ud835\udefc) cos \ud835\udefc]3 \ud835\udc51\ud835\udefc \ud835\udefc2 \u2212\ud835\udefc1 (2) 1 \ud835\udc58\ud835\udc60 = \u222b 1.2(1+\ud835\udc63)(\ud835\udefc2\u2212\ud835\udefc) cos \ud835\udefc cos2 \ud835\udefc1 \ud835\udc38\ud835\udc3f[sin \ud835\udefc+(\ud835\udefc2\u2212\ud835\udefc) cos \ud835\udefc] \ud835\udc51\ud835\udefc \ud835\udefc2 \u2212\ud835\udefc1 (3) 1 \ud835\udc58\ud835\udc4e = \u222b (\ud835\udefc2\u2212\ud835\udefc) cos \ud835\udefc sin2 \ud835\udefc1 2\ud835\udc38\ud835\udc3f[sin \ud835\udefc+(\ud835\udefc2\u2212\ud835\udefc) cos \ud835\udefc] \ud835\udc51\ud835\udefc \ud835\udefc2 \u2212\ud835\udefc1 (4) Where E is the elastic modulus; L is the axial thickness of gear, and \u03bd, the Poisson's ratio. They are the four components of gear stiffness. The meshing stiffness of gear pair should also include fillet-foundation stiffness kf, which can be expressed as follows. 1 \ud835\udc58\ud835\udc53 = cos2 \ud835\udefc \ud835\udc38\ud835\udc3f {\ud835\udc3f\u2217 ( \ud835\udc62\ud835\udc53 \ud835\udc60\ud835\udc53 ) 2 + \ud835\udc40\u2217 ( \ud835\udc62\ud835\udc53 \ud835\udc60\ud835\udc53 ) + \ud835\udc43\u2217(1 + \ud835\udc44\u2217 tan2 \ud835\udefc)} (5) Where the coefficientsL\u2217, M\u2217, P\u2217 and Q\u2217 can be represented by polynomial functions. \ud835\udc4b\u2217 = \ud835\udc34\u2217 \ud835\udf03\ud835\udc53 2 + \ud835\udc35\u2217\u210e\ud835\udc53 2 + \ud835\udc36\u2217\u210e\ud835\udc53 \ud835\udf03\ud835\udc53 + \ud835\udc37\u2217 \ud835\udf03\ud835\udc53 + \ud835\udc38\u2217\u210e\ud835\udc53 + \ud835\udc39\u2217 (6) X\u2217 represents the coefficients L\u2217, M\u2217, P\u2217 and Q\u2217; hf = rf /r; where rf is the radius of root circle; the meanings of uf, \u03b8fand Sf are shown in Figure4; and the values of A\u2217, B\u2217, C\u2217, D\u2217, E\u2217 and F\u2217 can be seen from Table 1 below. The Hertz stiffness kh, bending stiffness kb, radial compression stiffness ka, shear stiffness ksand fillet-foundation stiffness kf combined in series give the meshing stiffness of gear pair. \ud835\udc58\ud835\udc61 = \ud835\udc58\ud835\udc61,1 + \ud835\udc58\ud835\udc61,2 = \u2211 1 1 \ud835\udc58\u210e,\ud835\udc56 + 1 \ud835\udc58\ud835\udc4f1,\ud835\udc56 + 1 \ud835\udc58\ud835\udc601,\ud835\udc56 + 1 \ud835\udc58\ud835\udc4e1,\ud835\udc56 + 1 \ud835\udc58\ud835\udc531,\ud835\udc56 ++ 1 \ud835\udc58\ud835\udc4f2,\ud835\udc56 + 1 \ud835\udc58\ud835\udc602,\ud835\udc56 + 1 \ud835\udc58\ud835\udc4e2,\ud835\udc56 + 1 \ud835\udc58\ud835\udc532,\ud835\udc56 2 \ud835\udc56=1 (7) Where i is the i teeth pair of gear teeth for a meshing gear pair. Equations (1), (2), (3), (4) and (5) are solved separately before we substitute the results into Equation (7) to get each stiffness and meshing stiffness of gear pair. It is not given in detail about the calculation of meshing damping coefficient as the calculation method has been specified in literature [9] . The meshing stiffness and damping coefficient of the gears with the module m = 5, and the gear teeth number z1 z2\u2044 = 22 30\u2044 are shown in Figure 2 as below. 2.2 Calculation of the collision force based on the absolute nodal coordinate For collision detection of the two gear, if found to exist intrusive between the two gear, returns the number of entry points and each impact point coordinate and the normal direction, as shown in Figure 3, according to these information applying a and intrusion into resistance function role in the invasion of the point, and according to the action and reaction principle, but also in the other side of the surface applied an equal and opposite force, equivalent to placing the normal spring in all detection geometry. If the surface is a surface of analytical rigid body, the equivalent force and moment applied to the mass center. If the surface is composed of unit surface, the collision force is discrete to node of contact element. Using the principle of virtual work, the collision friction force of the unit surface can be transformed into the generalized nodal force f cQ (p)f T cQ N f Where p is the element coordinate parameters corresponding unit collision point, determined by the position of collision point, f is the force that roll in the collision point. Including the collision force and friction force of the collision plane, Calculated by Hertz contact model. t t nf u v f Where n is a vector to unit collision law, nf for the law to the collision force, Use the following formula: 2 31 . . . m mm nf k c Where K and C are spring and damping coefficient, the value measured by experiment method to determine. Index m1 and m2 to produce nonlinear collision force. the m3 index indentation damping effect. General steel material, m1=1, M2 between 2.2 and 2.5, these coefficients are relate to the collision shape. and . were between two geometry of intrusion rate and the time derivative of invasion rate. When the invasion depth is very small, because the damping force of negative influence that the collision force may be negative, This can be done by the index m3 is greater than 1 to solve. t is tangential unit vector, defined as t tt v v among them the tangential velocity tv defined as .t r rv v v n n among them rv is the relative velocity between two detection body corresponding to the collision point.. t t nf u v f according to the Kulun friction among them nf the friction coefficient which is determined by the relative tangential velocity tv . 2.3 Establish the rigid flexible coupling model The SOLIDWORKS set up the model into ABAQUE, and will import the ADAMS of the shaft and the box flexible. Parameters for the value into the model and simulation .To solve the model shown in Figure 4." + ] + }, + { + "image_filename": "designv11_84_0001497_978-3-319-07572-3_6-Figure6.1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001497_978-3-319-07572-3_6-Figure6.1-1.png", + "caption": "Fig. 6.1", + "texts": [ + " Tensile and compressive stresses are normal stresses developed within a body as a result of external forces. These stresses are axial stresses and act along the longitudinal axis of the member. Their action is in the same direction as the applied force. In the tensile or compressive stress, the force is applied normal to the cross section under consideration, however if the transverse force is applied to a member the internal forces develop in the plane of the cross section and they are called shearing forces (Fig. 6.1). Since shear distribution in the cross section of the member is not necessarily uniform, an average shear stress is used. Examples of shear stress are found in bolts, rivets and pins that are being used for connecting of various structural members. The relationship between average shear stress in the cross section and the shear (F) can be expressed as: \u03c4ave \u00bc F=A Figure 6.2, shows the rivet connection under tension force F and the shear stress will be developed in the rivet section connecting the plates" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001126_2015-01-2515-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001126_2015-01-2515-Figure11-1.png", + "caption": "Figure 11. Horizontal inject injector cutaway view", + "texts": [], + "surrounding_texts": [ + "This new style of rivet injector is in production use on a variety of fastening machines used by major aircraft manufacturers. We have implemented the parallel gripper in both vertical axis and horizontal axis riveting applications. It is equally effective in both orientations. We have implemented the parallel gripper rivet injector on headed rivets, threaded bolts, ribbed swage bolts and unheaded (slug) rivets. The use of the parallel gripper to constrain and control the injector guide chute enables the feeding of multiple fastener types through the same guide geometry while minimizing jam opportunities and providing a built in purge function. It enables reliable injection of a wide range of fastener types with minimal hardware reconfiguration between types. Contact Information Contact Electroimpact at 425-348-8090 or the authors for more information. Adlai Felser adlaif@electroimpact.com Peter B. Zieve peterz@electroimpact.com Bryan Ernsdorff bryane@electroimpact.com The Engineering Meetings Board has approved this paper for publication. It has successfully completed SAE\u2019s peer review process under the supervision of the session organizer. The process requires a minimum of three (3) reviews by industry experts. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of SAE International. Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE International. The author is solely responsible for the content of the paper. ISSN 0148-7191 http://papers.sae.org/2015-01-2515" + ] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.17-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.17-1.png", + "caption": "Fig. 7.17 An example of 6\u2013UPS PKM with simplified FGM", + "texts": [ + " This result was first shown inRonga andVust (1992), and then confirmed through different approaches proposed in Husty (1996), Lazard (1993), Mourrain (1993), Raghavan (1993) and Wampler (1996). The first researchers who were able to give the expression of the univariate polynomial of degree 40 whose roots correspond to the assembly modes of the 6\u2013UPS PKM were Husty in Husty (1996) and Wampler in Wampler (1996). The number of solutions considerably decreases for special arrangement of the legs. For example, with the design proposed in Fig. 7.17 for which the legs 1, 2 and 3 (4 and 5, resp.) are linked to the same platform point A16 (A46, resp.), the number of solutions is decreased to 8 and all of them can be obtained in a closed-form using the following method (Hunt and Primrose 1993; Nanua and Waldron 1991): 1. Knowing the lengths q13, q23 and q33 of the legs 1, 2 and 3, compute the position of point A16 (that will be considered here as the platform controlled point with coordinates (x, y, z)) which is at the intersection of the three spheres centers in A11, A21 and A31 of radius q13, q23 and q33, respectively; thus the translational part of the vector x is found, 2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002639_amr.314-316.653-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002639_amr.314-316.653-Figure5-1.png", + "caption": "Fig 5 The direction of the Z axle Fig 6 The direction of the x-y axle", + "texts": [], + "surrounding_texts": [ + "In the most compression condition, the most stress of the connection rod emerged in the unilateral internal surface of the rod\u2019s small end anear big end. In the most tension stress condition, the most stress emerged in the connected arc of the small end and the rod body. So we applied stress to the connection rod\u2019s small end, the emulator results is shown as Fig3-Fig8 : The results are mapped path and the result graph of the path is shown as Fig 9:" + ] + }, + { + "image_filename": "designv11_84_0001077_978-1-4419-7979-7_6-Figure6.61-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001077_978-1-4419-7979-7_6-Figure6.61-1.png", + "caption": "Fig. 6.61 Ring-type inductor", + "texts": [ + "12: Flux concentration within air gap of a C-core and application of the maximum energy product for permanent magnets It is desired to achieve a magnetic flux density Bg\u00bc 2.0 T in the air gap of the magnetic circuit of Fig. 6.59. The field is to be created by a samarium-cobalt (SaCo) permanent magnet. For the air-gap dimensions of Fig. 6.59 find the magnet length \u2018m and the magnet area Am that will achieve the desired air gap flux density and minimize the magnet volume Vmagnet_minimum. Calculate Vmagnet_minimum. Problem 6.13: Analysis of an inductor with given geometrical dimensions An inductor is designed using the magnetic core of Fig. 6.61. The core and gap have uniform cross-section areas Ac\u00bcAg\u00bc 5 cm2 (neglect fringing and stacking lamination effects), the core average length is \u2018c \u00bc 40 cm; the excitation current is i\u00bc 4.5 A, the air-gap length is g\u00bc 0.17 mm, the constant permeability of the core is mr\u00bc 3,500 and the number of turns is N\u00bc 100 turns. Calculate: (a) The flux density B and the inductance L, (b) The magnetic energy stored in air gap Wg, (c) The magnetic energy stored in the core Wc, and (d) The total magnetic energy stored in the inductor WT" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001497_978-3-319-07572-3_6-Figure6.14-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001497_978-3-319-07572-3_6-Figure6.14-1.png", + "caption": "Fig. 6.14", + "texts": [ + "00-in. diameter rivet as shown in Fig. 6.12. Determine the shear stress in the rivet. Solution A \u00bc \u03c0D2=4 \u00bc \u03c0 1\u00f0 \u00de2=4 \u00bc 0:785 in:2 \u03c4 \u00bc F=A \u00bc 400 lb=0:785 in:2 \u00bc 509:3psi Example 6.8 Calculate the shearing stress in the pins with diameter of\u00bd-in. shown in Fig. 6.13. A tensile load of 250 lb is applied. A \u00bc 2 \u03c0D2=4 \u00bc 2 \u03c0 1=2\u00f0 \u00de2=4 h i \u00bc 0:393 in:2 \u03c4 \u00bc F=A \u00bc 250 lb=0:393 in:2 \u00bc 636:6psi Example 6.9 Calculate the shearing stress in the bolts with the diameter of \u00bc in. for connecting the plates shown (Fig. 6.14). A tensile load of 10,000 lb is applied. A \u00bc 4 \u03c0D2=4 \u00bc 4 \u03c0\u00f0 \u00de 0:25 in:\u00f0 \u00de2=4 \u00bc 0:196 in:2 \u03c4 \u00bc F=A \u00bc 10, 0001b=0:196 in:2 \u00bc 5:09 104 psi Shear stress can also be calculated when a force F is applied to punch a hole through a plate. In this case, the shear stress is resisted by a cylindrical area with a diameter equal to the punch diameter and a height equal to the thickness of the plate. Example 6.10 A punching machine punches a 1-in. diameter hole through a \u00bc-in. plate. Calculate the force needed to punch through the plate" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002496_s10846-013-9971-y-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002496_s10846-013-9971-y-Figure3-1.png", + "caption": "Fig. 3 Distances between the external forces to the helicopter center of gravity", + "texts": [ + " On other hand, the external torque depends on the thrust of the main and tail rotors, and on the stiff at the joint of the hub and the blade, together with the reaction torque of the main and the tail rotor. The mathematical expression for such a torque is given by [7] \u23a1 \u23a3 L M N \u23a4 \u23a6 = \u23a1 \u23a3 LM MM + MT NM \u23a4 \u23a6 + \u23a1 \u23a3 YMlh + Z M yM + YT hT \u2212XMlh + Z MlM \u2212YMlM \u2212 YTlT \u23a4 \u23a6 (9) The terms (lM, yM, hM) and (lt, yt, ht) are distances from the center of gravity to the point of application of the thrust of the main and tail rotors (see Fig. 3). The torques LM, MM, and NM are given by [7] LM = k\u03b2b \u2212 QMsa MM = k\u03b2a + QMsb NM = \u2212QMcacb MT = \u2212QT where k\u03b2b and k\u03b2a are the torques induced by the rotor stiffness, QM is the torque of the main rotor and QT is the torque due to tail rotor. 3.1 Reduction of the Dynamic Model A dynamic model reduction is carried out in order to obtain a model restricted to the plane xIzI . As a result, the angular motions corresponding to roll and yaw and the lateral motion are considered null, that is, y = 0 m, v = 0 m/s, Y = 0 N, \u03c6 = \u03c8 = 0 rad, p = r = 0 rad/s, and L = N = 0 N" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003606_amm.339.510-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003606_amm.339.510-Figure5-1.png", + "caption": "Figure 5 Equivalent stress nephogram of gear", + "texts": [ + " Because the surface contact pressure stress near is high, but the surface relative sliding speed is low.The product of them is low. So the friction heat is low;The temperature at the pitch point is the lowest,because the surface relative sliding speed here is zero. Study on thermal and contact Properties of Cylindrical Gear Contact question is solved afer the temperature and thermal deformation is solved, then the contact state show the cooperation effect of thermal deformation and mechanical deformation as fig.5. It is shown from figure 5 that under the cooperation of thermal load and mechanical load, there are some characteristics about gear surface contact stress distribution. (1) In the tooth width direction, gear surface contact stress approximately symmetrical distributes along the middle section surface of tooth width. The biggest contact stress on the capstan and driven gear appears in the middle of theoretical contact line. Symmetrical distribution style of temperature field of gear tooth effects this. And \"drum shape\" appears in the middle of tooth width" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001961_icra.2014.6907673-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001961_icra.2014.6907673-Figure6-1.png", + "caption": "Fig. 6. Input constraint", + "texts": [ + " In previous section, we assumed that the attitude of the helicopter is near hovering state. Accordingly the desired attitude angle should be limited for satisfying the assumption. We deal with such limit as an inequality constraint, and it is derived as g(u\u0304) = F\u0304 ix 2 + F\u0304 iy 2 \u2212 (mg sin\u03b1)2 (10) Here, suffix i represents i th estimate in evaluation interval, \u03b1 [rad] is permissible maximum tilt angle of the helicopter. The relation between the external force and the tilt angle of the helicopter is shown in Fig.6. We introduce such input constraints to criterion as the penalty function Pu(u\u0304(t+ \u03c4)), and it could be obtained as Pu(u\u0304) = (max{0, g(u\u0304)})2. (11) Fourth and fifth term in (7) are state constraints for collision avoidance. In order to avoid the collision, the prohibited area is established around each helicopter (Fig.7). The inequality constraints could be obtained as gx(x\u0304) = r2 \u2212 { (x\u0304ij \u2212 x\u0304ik) 2 + (y\u0304ij \u2212 y\u0304ik) 2 } (12) Here, suffix j\uff0ck represent j th and k th helicopter, r is the radius of no entry area" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002117_2013-01-2011-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002117_2013-01-2011-Figure9-1.png", + "caption": "Figure 9. CVM-D model used to calculate the forces entering the cab mounts for Automatic Side Loading operation", + "texts": [ + " Figure 5 shows the displacement of the cab rear mount in the vertical direction for the 5 configurations. In addition to the string pots, the cab was instrumented with accelerometers as well. Figures 6, 7, 8 show the locations of the accelerometers. CVM-D (Complete Vehicle Modeling - Diamondback), which is a Volvo in-house developed tool, was used to calculate the Forces entering the cab mounts based on the displacements measured by the string pots. CVM-D model used for the analysis is shown in Figure 9. A fully trimmed cab model as shown in Figure 10 was used for the analysis. Forces extracted from CVM-D were applied at each cab mount and the stresses were extracted while giving support cards using PARAM, INREL,-2. A single Automatic Side Loading operation takes 10 seconds from start to finish. The vertical global displacements with the 75Lb garbage can, measured at the four locations are shown in Figure 11. At 5seconds time slice, the vertical global displacements show a roll mode of the cab" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000926_gt2008-50641-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000926_gt2008-50641-Figure3-1.png", + "caption": "Figure 3 \u2013 The six DOF model", + "texts": [ + "s-1] Superscript \u2022 Derivative with respect to time Coordinate systems ( )zyx rrr ,, Cylindrical coordinate system ( )ZYX rrr ,, Cartesian coordinate system Copyright \u00a9 2008 by ASME rms of Use: http://www.asme.org/about-asme/terms-of-use Dow THEORETICAL ANALYSIS The Foil Bearing Structural Model The mathematical model of the elastic structure is first presented for a strip made of two bumps (Fig. 2) and then it is extended to an arbitrary number of bumps. The model considers the continuous foil structure as an equivalent discrete one with a restricted number of nodes linked by linear springs. For instance, the model depicted in Fig. 3 has six degrees of freedom, namely two vertical displacements, 1v and 3v , and four horizontal displacements 1u \u2026 4u ; this is the minimum required for describing the coupling between two successive bumps. The elementary characteristics of the bump strip model are the stiffnesses ik and the transmission force angle d\u03b8 that can be analytically expressed for each type of bump. The elementary stiffnesses and transmission force angles are determined by using energetic methods for two kinds of boundary conditions (Fig", + " The complete mathematical development and the resulting relations for ik and d\u03b8 are nloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 12/20/2017 T given in [17]. The global stiffness matrix of the whole structure is derived from the expression of the potential energy. To determine this potential energy, the elongation of the elementary springs needs to be expressed. For more clarity these elongations will be first expressed for the strip constituted of two bumps having the same elementary characteristics ik and d\u03b8 shown on Fig. 3. Then the results will be extended to the general case of n bumps with the last bump having different elementary characteristics ibisk and dbis\u03b8 . The elongations of the eight springs are depicted in Fig. 3. The model is considered to be linear, i.e. the final configuration is close from the initial configuration. Hence the eight elongations are: \u23aa \u23aa \u23a9 \u23aa \u23aa \u23a8 \u23a7 =\u0394 \u2212\u2212=\u0394 \u2212=\u0394 =\u0394 24 1123 112 11 ul svcucul svcul ul \u23aa \u23aa \u23a9 \u23aa \u23aa \u23a8 \u23a7 \u2212=\u0394 \u2212\u2212=\u0394 \u2212\u2212=\u0394 \u2212=\u0394 248 3347 2336 135 uul svcucul cusvcul uul (1) where dc \u03b8cos= and ds \u03b8sin= . The potential energy is: ( ){ ( ) }2 14 2 53 2 8 2 42 2 7 2 6 2 3 2 212 1 lklkllk llllkEpot \u0394+\u0394+\u0394+\u0394+ \u0394+\u0394+\u0394+\u0394= (2) This leads to: { ) ( ) ( ) }2 14 2 331 2 13 42 2 4 2 2243 2 4332 2 3221 2 21 22 4 22 3 22 3 22 2 22 1 22 11 2 222 22222 22222( 2 1 ukuuuuk uuuukcsuv cuucsvucuucsuvcuu cusvcucusvcukEpot ++\u2212+ \u2212++\u2212 \u2212+\u2212\u2212\u2212 +++++= (3) The foil is represented by a linear elastic system" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001427_intmag.2015.7157491-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001427_intmag.2015.7157491-Figure1-1.png", + "caption": "Fig. 1 3D motor configuration.", + "texts": [], + "surrounding_texts": [ + "A 2D-3D hybrid finite element method (FEM) approach is proposed for fast analysis of eddy current loss in magnets of permanent magnetic synchronous motors (PMSM) driven by a pulse width modulation (PWM) inverter . A voltage-source PWM inverter circuit coupled 3D FEM model of a PMSM was built with the commercial FEM code JMAG and the eddy current loss in the magnets of the PMSM was calculated which was extremely time-consuming . As the eddy current loss in magnets is relatively small compared with the power capacity of the motor, a voltage-source PWM inverter circuit coupled 2D FEM model is built to calculate the current waveforms ignoring the division of permanent magnets along the axial direction . Applying the calculated current waveforms from the 2D calculation to a current-source circuit coupled 3D FEM model, the eddy current loss in the magnets can be obtained which needs much less computation time . The result is very close to that obtained directly from the voltage-source PWM inverter circuit coupled 3D FEM model . II . VOLTAGE-SOURCE PWM INVERTER COUPLED 3D FEM MODEL The research target is a Y-connection 30 kW surface mounted PMSM with a rated rotation speed of 1500 rpm . The configuration of the motor is shown in Fig . 1 . The division number of the permanent magnets along the axial direction is 15 . The permanent magnetic material of the motor is Nd2Fe14B of type N35UH, with a conductivity of 6 .944\u2019105 S/m and residual flux density of 1 .21T . The core material is DW470-50 anisotropic steel sheet, which thickness is 0 .5 mm and iron loss density is 4 .7 W/kg at 50 Hz with the magnetic flux density of 1 .5 T . The motor has a 2-layer winding structure with 3 phase 4 poles 48 slots and coil pitch of 11/12 . Considering the division of permanent magnets along the axial direction, the 3D FEM model of the motor is built with the commercial FEM code JMAG, coupled by a voltage-source PWM inverter circuit . The PWM carrier frequency is 3 .15 kHz, with a modulation index of 0 .8 and the fundamental frequency of 50 Hz . The 3D FEM analysis is run on a computer with the Core 2 Duo 3 .0GHz CPU and 4GB memory . The whole analysis takes 289 hours and the eddy current distribution in the magnets is obtained and the eddy current loss calculated is 472 .2W . To evaluate the accuracy of the eddy current loss, a 3D FEM thermal analysis is performed using the results of the eddy current loss calculation . The predicted temperature rise on the surface of the magnets is 100 .86 while the measured one by an infrared thermometer sensor embedded through the stator core to the air-gap of the motor is 94 .38, the relative error is 5 .4% . III . VOLTAGE-SOURCE PWM INVERTER COUPLED 2D FEM MODEL As the results of the 3D analysis indicate that the eddy current loss of 472 .2W in magnets is relatively small compared with the power capacity of 30kW of the motor, a voltage-source PWM inverter coupled 2D FEM model is built to calculate the current waveforms where the division of permanent magnets along the axial direction is ignored . The calculated current waveforms are shown in Fig . 2 that agree with both measured results and those obtained from the voltage-source PWM inverter coupled 3D FEM analysis . IV . CURRENT-SOURCE CIRCUIT COUPLED 3D FEM MODEL The calculated currents of the motor by the voltage-source PWM inverter coupled 2D FEM are applied to a the current-source PWM inverter coupled 3D FEM the eddy current loss can be calculated with the computation time reduced significantly . The result agrees very well with that obtained from the voltage-source PWM inverter coupled 3D FEM analysis . 1) Wenyi Liang et al, \u201cAnalytical Modeling of Current Harmonic Components in PMSM Drive With Voltage-Source Inverter by SVPWM Technique\u201d, IEEE Trans . Energy Conversion, 29(3) 673-680, 2014 . 2) A . M . Knight et al, \u201cIntegration of a First Order Eddy Current Approximation With 2D FEA for Prediction of PWM Harmonic Losses in Electrical Machines\u201d, IEEE Trans . Magn . 49(5) 1957-1960, 2013 . 3) Katsumi Yamazaki et al, \u201cInvestigation of Locked Rotor Test for Estimation of Magnet PWM Carrier Eddy Current Loss in Synchronous Machines\u201d, IEEE Trans . Magn ., 48(11) 3327-3330, 2012 . 4) Mircea Popescu et al, \u201cA General Model for Estimating the Laminated Steel Losses Under PWM Voltage Supply\u201d, IEEE Trans . Ind . Appl ., 46(4) 1389-1396, 2010 . 5) K . Yamazaki and S . Watari, \u201cLoss Analysis of Permanent-Magnet Motor Considering Carrier Harmonics of PWM Inverter Using Combination of 2-D and 3-D Finite-Element Method\u201d, IEEE Trans. Magn., 41(5) 1980-1983, (2005) . 6) K . Yamazaki and A . Abe, \u201cLoss Investigation of Interior Permanent-Magnet Motors Considering Carrier Harmonics and Magnet Eddy Currents\u201d, IEEE Trans. Ind. Appl., 45(2) 659-665, (2009) ." + ] + }, + { + "image_filename": "designv11_84_0003273_time-e.2014.7011622-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003273_time-e.2014.7011622-Figure4-1.png", + "caption": "Figure 4. Mechanical structure designs by using SolidWorks.", + "texts": [ + " In designing the mechanical structure of miniature CNC milling machine will use the SolidWorks, the dimension of the work area is 210 x 300 x 50 mm. Each axis will use a permanent magnet DC motor through a reduction gear box to driven ball screw. For detecting of position and speed of each axis be use an absolute encoder. In addition, there are others parts that are designed to assemble the machine. However, this design process must consider the real parts and accessories that are used in the industry and can find devoid difficult. The result of the design is shown by Fig. 4. 2) Control Designs: the concept of position and speed control of each axis of the machine is shown by Fig. 5. However, in the simulation of any systems, we need to know the mathematical model of those systems. As we have discussed in the past topic, the design of mechanical structure, we need to select and configure devices and accessories as fact and sometimes necessary to provide for use in testing of important parameters. For this project to be of calculating the size of motor is 24 watts, and select a permanent magnet DC 24 volts motor" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001092_j.proeng.2015.07.168-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001092_j.proeng.2015.07.168-Figure4-1.png", + "caption": "Fig. 4 Vectors demonstrate each position on two-link model. r is defined as vector for the direction from the shoulder to the grip end. \u03c1 is also defined as vector for the direction from the grip end to the origin of arbitrary node on the shaft coordinate. We then define displacement vector n and position vector for the direction from the shoulder to the arbitrary node", + "texts": [], + "surrounding_texts": [ + "2.1. Displacement of club grip and shaft A golf club is divided into three parts: the grip, shaft, and club head. In this study, the grip and shaft are modelled with a multistage beam. This multistage beam is devised using FEM with Euler-Bernoulli beam-type element [7-9]. The grip consists of six elements and the shaft consists of 16. We then divide a golf club into two areas. The area which includes the grip and the element which connects to the grip is defined as the physical area. Another area is defined as the elastic deformation area (Fig. 1). Regarding with the i-th element, the i-th node and the i+1-th node are located on either end. The origin of shaft coordinate system is put on either node of the i-th element and constitutes the nondeformation condition. On this shaft coordinate system, the negative direction of the y axis is defined as the toe direction and the positive direction of the z axis is the face direction (Fig. 2). Defining displacement as u, v, and w for x, y, and z directions, these displacements are formulated by the following equations: [ ] dNwvu T ][==n (1) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )[ ]Tiziyixiiiiziyixiii zyxzyxd 111111 ++++++= \u03b8\u03b8\u03b8\u03b8\u03b8\u03b8 (2) [N] indicates shape function and d indicates node displacement (Eq. 1). In Eq.2, x, y, and z indicate the node displacement of each direction. \u03b8x is the angle of twist (rotation of x axis) and \u03b8y, \u03b8z show the slope of each axis. In Eq. 2, each index shows each axis and each node. 2.2. Motion equation for grip and shaft The motion equation for the grip and shaft is led by a two-link pendulum model (Fig. 3). On this pendulum model, the origin point of inertial coordinate system [a] is placed on the golfer\u2019s shoulder. We then define vector for the direction from the shoulder to the grip end as r and local joint coordinate system [b] which the origin point is the grip end. We also define vector for the direction from the grip end to the origin on the shaft coordinate system as \u03c1. Vector u, which shows the direction from the origin of the inertial coordinate system to the i-th node, is obtained by: The club face Grip Face Toe The i-th element The i+1-th node The i-th node z y xL Fig. 2 Coordinate system on the i-th element. The origin of this coordinate system is put on either end. The length between either end is indicated as L. The negative direction of the y axis is defined as the toe direction and the positive direction of the z axis is the face direction n\u03c1ru ++= (3) [ ] [ ]\u03c1b\u03c1ar == ,r\u0302 (4) r\u0302 indicates the translation component on the inertial coordinate system and \u03c1 is the non-time variable component. Then, the relationship of each coordinate system (Fig. 3) is obtained by: [ ] [ ]Sab = (5) S is the coordinate transform matrix. Using Eq. 5, the motion equation for the grip and shaft is formulated as follows using d\u2019Alembert\u2019s principle: 0=\u22c5un\u03b4 (6) [ ][ ] dN \u03b4\u03b4 bn = (7) \u03b4n is the virtual displacement of deformation and angle bracket is the mass integral. Expanding Eq. 6, Eq. 6 is deformed by: [ ] [ ] [ ] ( )rSNdNN TTT \u02c6~~~ ++\u2212= \u03c1\u03c9\u03c1\u03c9\u03c9 (8) \u03c9~ indicates the antisymmetric tensor of the angle rate. Considering potential energy, we can obtain the motion equation for the grip and shaft of each element. [ ] [ ] [ ] ( )( )grSNdKdM TT \u02c6\u02c6~~~ +++\u2212=+ \u03c1\u03c9\u03c1\u03c9\u03c9 (9) g\u0302 is the gravity component on the inertial coordinate system and [M] is the mass matrix that is guided by kinetic energy. [K] is the stiffness matrix that is guided by strain energy. Finally, combining the motion equation for the grip and shaft of each element and the motion equation for the torsion, the complete motion equation is obtained by: [ ] [ ] [ ] [ ] ( )( )grSNdKdCdM TT ttttttt \u02c6\u02c6~~~ +++\u2212=++ \u03c1\u03c9\u03c1\u03c9\u03c9 (10) Index t shows the total of each matrix and vector. [Ct] indicates the total modal damping matrix [10] that is guided onto the elastic deformation area (Fig.1). Moreover, considering the angle of twist and slope as small, the motion equation for the club head is obtained from the follow two equations: ( )( )grSmdm T headehead \u02c6\u02c6~~~ +++\u2212= \u03c1\u03c9\u03c1\u03c9\u03c9 (11) [ ] [ ] [ ]( )\u03c9\u03c9\u03c9\u03b8 JJJ e +\u2212= ~ (12) ( ) ( ) ( ) ( ) ( ) ( )][][ eeeeeeee zyxd \u03b8\u03b8\u03b8\u03b8 == (13) de demonstrates the displacement of the shaft tip\u2019s node and \u03b8e is the angle of twist and slope of the shaft tip\u2019s node. J in Eq. 12 indicates the inertia moment tensor around of the club head\u2019s centroid. mhead is the club head\u2019s mass. 2.3. Gripping model To take the golfer\u2019s grip into consideration [11], we modelled the grip using potential grip energy during the swing. This potential energy is calculated by the grip displacement (Eq. 1). Using Eq. 1, each direction\u2019s potential energy is formulated by the following equations: dxkKdxwkKdxvkKdxukK L tw L zz L yy L xx \u222b\u222b\u222b\u222b ==== 0 2 0 2 0 2 0 2 2 1, 2 1, 2 1, 2 1 \u03b8\u03b8\u03b8 (14) L indicates the length between either end of the i-th element (Fig.2) and \u03b8tw is the angle of twist on the i-th element [8]. ,kx, ky, kz, and k\u03b8, are the stiffness parameters for each direction, which are estimated by the latter noted parameter estimation method. The grip force of the i-th node on the grip part is then calculated by: ( ) ( ) ( ) ( ) ( ) ( ) ( ) \u23a5 \u23a5 \u23a6 \u23a4 \u23a2 \u23a2 \u23a3 \u23a1 \u2202 \u2202 \u2202 \u2202 \u2202 \u2202 \u2202 \u2202 \u2202 \u2202 \u2202 \u2202 \u2212= iz y iy z ixi z i y i x iG KKK z K y K x K F \u03b8\u03b8\u03b8 \u03b8 (15)" + ] + }, + { + "image_filename": "designv11_84_0001960_012020-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001960_012020-Figure3-1.png", + "caption": "Figure 3. Phase portrait based on the twist in the first shaft (q2 \u2212 q1).", + "texts": [ + " The largest error is at the rotation speed of the corresponding shaft, for example due to errors in the location of the gear centre. The next highest component is at the harmonic corresponding to the number of teeth and represents tooth meshing errors. The other harmonic terms are significantly smaller. From the Fourier series of the gear errors it is clear that the excitation contains many frequencies. Furthermore the parametric excitation terms means that there will also be responses at many combination frequencies. Figure 3 shows the phase portrait of the twist in the first shaft, given by q2 \u2212 q1. Figure 4 shows the FFT of the time response of the twist in the first shaft. It is clear that many frequencies are excited, including many of the higher frequencies. The interest in this paper is the effect on the low frequency dynamics. Figure 5 shows a zoomed plot for the twist in the first shaft and clearly shows many combination frequencies. There are large discrete responses at \u21262 = 45Hz and \u21263 = \u221222.7Hz. For comparison Figure 6 shows the zoomed FFT plot when the parametric excitation is switched off" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002962_j.ijmecsci.2014.03.030-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002962_j.ijmecsci.2014.03.030-Figure5-1.png", + "caption": "Fig. 5. Lagrangian's coordinates on the shell.", + "texts": [ + " If the sizes of the trapezoids a and b are not big compared to the length of the whole strip, then the middle lines of the fractal system trapezoids approximate the surface of revolution, the metric and curvature of which are dependent on the function of the angle \u03c8 . The natural parameterization and homogenization of the surface are reached if the Lagrangian's coordinates along the lines \u03b12 \u00bc const are equal to the geometrical length of the middle line of the strip while the distance between the strips is taken equal to a along the lines \u03b11 \u00bc const (Fig. 5). The first and the second quadratic form of the surface look as follows ds2 \u00bc A2 1 d\u03b1 2 1\u00feA2 2 d\u03b1 2 2 A1 \u00bc\u03981\u00f0\u03a8 \u00de; A2 \u00bc\u03982\u00f0\u03a8 \u00de B\u00f0d\u03b11; d\u03b12\u00de \u00bc A2 1 d\u03b1 2 1 \u03c11 \u00feA2 2 d\u03b1 2 2 \u03c12 \u03c11 \u00bc s1 \u03bb1 \u03c611 ; \u03c11 \u00bc s2 \u03bb2 \u03c622 \u00f021\u00de The functions s1; s2;\u03981;\u03982; \u03bb1; \u03bb2 in the formulas (21) are identified by the shape of the cell. They depend on the length a, the relation a=b, the sharp angle by the base of the trapezoid and the angle \u03c8 . The functions \u03c611;\u03c622 are selected from the relations of the structure, which is the Gauss-Codazzi relation for the two- dimensional surface" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002344_detc2013-12837-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002344_detc2013-12837-Figure5-1.png", + "caption": "Figure 5. CIRCULAR-ARC NORMAL SECTION OF G FOR GENERATING MODIFIED CURVILINEAR GEAR", + "texts": [ + " Angle 1 is the pinion\u2019s rotational angle during the generation process, and 1r denotes the pitch radius of the pinion. (a) (b) Figure 4. (a) RACK CUTTER P ; (b) PINION GENERATED BY RACK CUTTER P The unit normal vector of the generated modified curvilinear pinion represented in coordinate system S1 was obtained as follows: PP PPP PPP z y x n n n sincos coscoscossinsin sincoscoscossin 11 11 1 1 1 1n . (5) Equations of Rack Cutter G The normal section of the rack cutter surface G used for gear surface generation is a circular-arc, as Fig.5 depicts. According to the coordinate system relationship depicted in Fig.5 and Fig.3, an imaginary three-dimensional rack cutter surface can be obtained, as shown in Fig.6(a). The position vector and unit normal vector of G represented in coordinate system )Z,Y,X(S )G( c )G( c )G( c )G( c are expressed as follows [16] : GGG G GnG GGG G GnG GnG G c E 2 S R E 2 S R R sinsin])cos(cos[ )cos1(cos])cos(cos[ )sin(sin )(R , (6) and GG GG G G c - sincos coscos sin )(n . (7) Symbols G and G are the surface parameters of rack cutter G . n is the normal pressure angle of the gear" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002731_icma.2013.6618005-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002731_icma.2013.6618005-Figure1-1.png", + "caption": "Fig. 1 A multi-link robot.", + "texts": [ + " The goal of our research is to develop a method for the coordination of multiple actuators to appropriately distribute energy to each body of a robot through the kinetic chain according to a desired motion. For this goal, it is important to clarify how to coordinate multiple actuators in order to generate proper energy flow. In this paper, in terms of energy flow in the kinetic chain, we evaluated robot motion generated by the coordination of multiple actuators using simulation. In this paper, we deal with a multi-link robot comprised of n rigid bodies (see Fig. 1 and 2 and Table I). The motion of the thi link is described by the Newton- Euler equations (see Fig. 2 (a)): gFv iGiim (the Newton equations), (1) iii T\u03c9I (the Euler equations). (2) where g is the gravitational acceleration vector. The total mechanical energy of the thi link iE is given as follows: Gi T iii T iGi T Giii mmE pg\u03c9I\u03c9vv 2 1 2 1 . (3) where Gip is the position of the center-of-gravity of the thi link. In a musculoskeletal robot, each artificial muscle actuator is attached around some certain joints, and the joint torque is generated by a group of artificial muscle actuators attached around the joint" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002054_amm.117-119.347-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002054_amm.117-119.347-Figure2-1.png", + "caption": "Fig. 2 Global analysis of system (7) with (a) 75.0=r , 0=\u00b5 and (b) 75.0=r , 0775.0=\u00b5", + "texts": [ + "1 show the global bifurcation diagram versus the various value r , which is obtained by keeping 300 period points after discarding the first 500 period points for each initial point. From Fig.1 it can be seen that the chaotic transition occurs as r pass through the critial value 7855.0\u2248cr , which dues to the boundary crisis. On the other hand, when r varies during the interval [ ]7855.0,6855.0=\u0398 , the coexistence phenomenon of period attractor and chaotic attractor arises. In order to unfold clearly the location of two coexisting attractors in Fig. 1, the generalized cell mapping using diagraph is utilized to show the global structure when \u0398\u2208= 75.0r , see Fig. 2. In Fig. 2 (a), the interested region D is divided into 200\u00d7400 uniform rectangular small grid of cells, and 5\u00d75 sampling points are evenly chosen in each cell. There are two attractors denoted by A(1) and A(2), and the corresponding basin of attraction are denoted by B(1) and B(2). The basin boundary of B(1) and B(2) is denoted by B(1,2). The period saddle on B(1,2) is denoted by S(1,2). Obviously, B(1,2) is the regular basin boundary and represents the stable manifold of period saddle S(1,2), which separates the period attractor A(1) from the chaotc attractor A(2). As the intensity of noise \u00b5 increases, the escape from period saddle S(1,2) occurs and the chaotic transition arises. Fig. 2 (b) shows that the global analysis after the chaotic transition. Comparing with Fig. 2 (a), the attractor A(1) becomes diffuse for the larger noise. But more significantly, larger noise results that the chaotic attractor A(2) (see Fig. 2 (a)) losses its atractive character and eventually becomes a chaotic saddle denoted by S(2) (see Fig. 2 (b)). The unstable solution denoted by S(1) fills the gap of escape. When 0\u2260\u00b5 , we are focus on the study of noise-induced chaotic transition. When r varies from 0.7 to 1, we choose the initial condition ( ) ( )5.0,5.1, 0 2 0 1 =xx in the basin B(2). Fig. 3 (a) shows that the top Lyapunov exponent (TLE) diagram for different \u00b5 . For each value \u00b5 , the step-size h is taken as 0.01 and the total time is 50000105 6 =\u00d7\u00d7= ht , the detailed procedure refers to [10]. It is clear that the chaotic transition occurs as r increases" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003006_1.3552414-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003006_1.3552414-Figure3-1.png", + "caption": "FIGURE 3. Tensile specimens design and laser cladding strategies", + "texts": [ + " The powder feeder is a Sulzer Metco Twin 10-C, which can control the mass flow injected to the coaxial nozzle by controlling the rotating speed of a metering disc. The test parts consist on two types of tensile specimens. First, a specimen that combines substrate and deposited material, which allow studying the bonding between laser cladded and base material. Second, a specimen that is built completely by laser cladded material, which allows the characterization of the deposited material (See Figure 3). Each test part has been built using two different strategies, resulting on four different tensile specimens. The laser cladding conditions have been obtained from previous results and a 1,100 W laser power and 700 mm/min feed have been used. The mass flow was set to 5.2 g/min. Once the laser cladding operation is finished, the tensile specimens are obtained by wire EDM cutting using an ONA PRIMA E 250 machine. Although EDM process can damage the surface of the test parts, minimal damage on the structure is introduced; in fact it is one of the recommended processes to obtain the tensile test probes The specimens have been tested following the EN 10002-1: 2002 standard on a universal testing machine" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001551_icma.2015.7237777-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001551_icma.2015.7237777-Figure1-1.png", + "caption": "Fig. 1 The structure of test-bed and the arrangement of measuring points", + "texts": [ + " When ( ) ( )( ) ( ) 1 2 11 2 2 1 2 0 , , ( , ) 1 3 , ,Bp N N \u03c9 \u03c9 \u0394 \u03c9 \u03c9 \u03c9 \u03c9 \u0394 \u2212 \u2209 = \u2212 + \u2208 (the discrete two-dimensional array of bispectral estimation is N\u00d7N, \u2018 \u2019 is bottom integral function, \u0394 is bifrequency triangle definition domain), bispectral entropy has a maximum ( )( )2 max =ln 1 3BH N N\u2212 + . So the closer to uniform distribution, the greater bispectral entropy value is. Conversely bispectral entropy value is smaller. Designed a gear test-bed. Its structure and arrangement of measuring points were shown in Figure 1. The test-bed consisted of two helical gear reducers, with the same structure and mirror image arrangement. And the shaft of two gear reducers was connected by one coupling. The gears had an involute helical gear structure. The larger one had 75 teeth, and the smaller one had 17 teeth. Two ends of the gear shafts were supported by rolling bearing in the gearbox. It took about 149 hours for a test on the test-bed. The motor with load had a constant speed in the test. Shut down the motor and inspected, a crack failure was occured at the small gear in reducer A" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002275_s1064230714040042-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002275_s1064230714040042-Figure1-1.png", + "caption": "Fig. 1. Schematic diagram of the machine.", + "texts": [ + " In [16], this problem was solved for a simple model of the machine consisting of two rigid bodies connected by a linear servodrive. In the present paper, we consider the model of the machine consisting of two pivotally connected bodies. This model can be considered as a simplified model of motion of a hopping machine, a high board diver, or an acrobat under the assumption that only the hip joint is used in the supportless phase. 1. STATEMENT OF THE PROBLEM The schematic diagram of the machine is shown in Fig. 1. The mass of the ith body (i = 1, 2) is mi, the moment of inertia of the ith body about its center of mass Ci is Ji, and the distance from the joint A to the center of mass of the ith body is . Let C be the center of mass of the machine. The state of the sys tem relative to the absolutely fixed reference frame Oxy (the axis Oy is directed vertically upwards) is deter i il AC= DOI: 10.1134/S1064230714040042 JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL Vol. 53 No. 4 2014 OPTIMAL ATTITUDE CONTROL OF TWO PIVOTALLY CONNECTED BODIES 611 mined by the coordinates x and y of the joint A, the pitch of the first body (the angle between the hori zontal axis Ox and the line ) and the angle between the bodies , which is controlled by the servodrive" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003767_j.apm.2013.04.046-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003767_j.apm.2013.04.046-Figure1-1.png", + "caption": "Fig. 1. Configuration of a thrust bearing with one porous wall.", + "texts": [ + " [15] but the problem of a porous squeeze film bearing lubricated by the same lubricant was presented by Walicka [16]. From various models of fluids with a yield shear stress the Voc\u030cadlo fluid (sometimes called Robertson and Stiff [17] fluid) flow frequently appears in many industrial branches: food processing, metal processing, petroleum industry. It also appears in tribology modeling semi-fluid lubricants [18,19]. The purpose of this study is to investigate the pressure distribution in a clearance of squeeze film bearing formed by two surfaces of revolution, having common axis of symmetry, as shown in Fig. 1; the lower one of these surfaces is connected Nomenclature C\u00f0n\u00de\u00f0x; t\u00de;Ck\u00f0n\u00de\u00f0x; t\u00de auxiliary functions given by formulae (3.10)1 and (3.12)1, respectively H thickness of a porous layer ~H non-dimensional thickness of a porous layer h bearing clearance thickness K non-dimensional capillary radius of a porous matrix n exponents in a Voc\u030cadlo fluid p pressure Q flow rate R;R\u00f0x\u00de local bearing radius rc capillary radius of a porous matrix SV Saint\u2013Venant plasticity number V squeeze velocity tx; ty components of the flow velocity e non-dimensional axial bearing clearance thickness Kyx component of shear stress s shear stress s0 yield shear stress sw shear stress on the bearing clearance wall U\u00f0n\u00de auxiliary function in the plane flow of a Voc\u030cadlo fluid given by Eq", + " The analysis is based on the assumption that the porous matrix consists of a system of capillaries of very small radii restricting the lubricant flow through the matrix in only one direction. The lubricant is assumed to be the Voc\u030cadlo type for which the constitutive equation has a following form [20,21]: s \u00bc s 1 n 0 \u00fe l _c n ; \u00f01:1\u00de where s is the shear stress, s0 is the yield shear stress, l is the coefficient of plastic viscosity, _c is the rate of deformation, m is the exponent in the Voc\u030cadlo fluid. The flow configuration is shown in Fig. 1. The upper boundary of a porous layer is described by function R(x), which denotes the radius of this boundary. The fluid film thickness is given by function h(x,t). An intrinsic curvilinear orthogonal coordinate system (x, #, y) is also depicted in Fig. 1. By using the assumptions typical for the flows in a thin layer the equations of motion for the Voc\u030cadlo fluid for axial symmetry one can present in the form [22\u201324]: 1 R @\u00f0Rtx\u00de @x \u00fe @ty @y \u00bc 0; \u00f02:1\u00de @p @x \u00bc @Kyx @y ; \u00f02:2\u00de @p @y \u00bc 0; \u00f02:3\u00de where the non-zero component of stress tensor is [23,24,18] Kyx \u00bc s 1 n 0 \u00fe l @tx @y n s 1 n 0 l @tx @y n for y 6 h0; y P h h0: 8>< >: \u00f02:4\u00de In the flow of a fluid with yield shear stress there exists a quasi-solid core bounded by surfaces laying at y \u00bc h0 or y \u00bc h h0 for which the shear stress is : jKyxj \u00bc s0: \u00f02:5\u00de The problem statement is complete after specification of boundary conditions which are tx\u00f0x;0; t\u00de \u00bc 0; tx\u00f0x;h; t\u00de \u00bc 0; \u00f02:6\u00de ty\u00f0x;0; t\u00de \u00bc V ; ty\u00f0x;h; t\u00de \u00bc @h @t ; \u00f02:7\u00de @p @x x\u00bc0 \u00bc 0; p\u00f0xo\u00de \u00bc po; \u00f02:8\u00de here V is the lubricant velocity on the upper boundary of a porous matrix, po is the outlet pressure" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002146_rvsp.2011.54-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002146_rvsp.2011.54-Figure3-1.png", + "caption": "Figure 3. The waypoint in the cone of the follower robot.", + "texts": [ + " 22 2 2 1 2 01 02 01 2 2 2 1 2 1 01 02 ( ) ( ) ( ) 2 ( ) ( ) sgn( ) 2 i i i i iw t t iw i it i i i t t t d L d x dx L y d L d y y y L \u2212 \u2212 \u2212 \u2212 \u239b \u239e\u239b \u239e+ \u2212\u239c \u239f\u00b1 \u2212 \u239c \u239f\u239b \u239e \u239c \u239f\u239c \u239f \u239d \u23a0\u239c \u239f = \u239c \u239f\u239c \u239f \u239c \u239f\u239d \u23a0 + \u2212\u239c \u239f+ \u2212\u239c \u239f \u239d \u23a0 (6) 2 2 2 1 2 1 01 02 22 2 2 1 2 01 02 01 ( ) ( ) sgn( ) 2 ( ) ( ) ( ) 2 i i i i i t t tiw t iw i i i it t d L d x x x Lx y d L d y d L \u2212 \u2212 \u2212 \u2212 \u239b \u239e+ \u2212 + \u2212\u239c \u239f \u239b \u239e \u239c \u239f \u239c \u239f = \u239c \u239f\u239c \u239f \u239b \u239e+ \u2212\u239c \u239f\u239d \u23a0 \u00b1 \u2212 \u239c \u239f\u239c \u239f\u239c \u239f\u239c \u239f\u239d \u23a0\u239d \u23a0 (7) The instantaneous waypoint bearing angle of follower robot is depicted as: ( ) ( )( )1 1tan 2 ,iw iw i iw i t t t t ta y y x x\u03b8 \u2212 \u2212= \u2212 \u2212 (8) III. WAYPOINT IN CONE METHOD In this section, the proposed waypoint in cone method guarantees the follower robot arrive at its waypoint correctly. Fig. 3 shows the relationship between the waypoint and the follower robot. The linear velocity and the angular velocity of the follower robot are calculated as: 2 2 1 1( ) ( ) , iw i iw ifw t t t ti t x x y ylv dt dt \u2212 \u2212\u2212 + \u2212 = = (9) 1 1 1 1 arctan . iw i it t ti iw i i t t t t y y x x dt dt \u03b8 \u03b1 \u03b8\u03c9 \u2212 \u2212 \u2212 \u2212 \u2212 \u2212 \u2212 \u2212 = = (10) In Fig. 3, there is a cone made up of \u03b4 and \u03b4\u2212 , the center of which is the orientation of the follower robot. The angle \u03b1 from waypoint to the follower robot needs to be within this cone. If not, the follower has to turn its orientation until 1 i t\u03b1 \u03b8 \u03b4\u2212\u2212 \u2264 using the following velocities. Then the follower robot recalculates its velocity based on the waypoint to move. 0,iv = (11) 1 maxsgn( ) .i i t\u03c9 \u03b1 \u03b8 \u03c9\u2212= \u2212 (12) The follower robot velocity has the following constraints: max0 ,i tv v\u2264 \u2264 (13) max max ,i t\u03c9 \u03c9 \u03c9\u2212 \u2264 \u2264 (14) " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003871_amc.2014.6823303-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003871_amc.2014.6823303-Figure3-1.png", + "caption": "Fig. 3. Sample of Link Confiburation.", + "texts": [ + " CONFIGURATION OF LEG-EXTENDER EXOSKELETON Figure 2 shows the two types of exoskeleton. The left one is similar shape with human leg. This configuration, however, has some problem because the knee of exoskeleton is located in front of operator and the usability become smaller. Therefore the right one was proposed in [1]. The right one has a counter type link mechanism and there are two knee joints, forward and backward. The shape around ankle looks like a bird. The operational feeling of right one seems worse. Figure 3 illustrates the posture of human and exoskeleton with various knee angles. The red stick is for human, and the green stick is for exoskeleton. The left one is the case with long back thigh and the right one is the case with short back thigh. The knees are located at same place, but the location of foot is not on the similar shape point especially in case of long back thigh. The range from 0 degree to 120 degree are used in the following discussion because the exoskeleton will be used with the knee joint stretched posture in most of the time" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001968_2013-01-0424-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001968_2013-01-0424-Figure1-1.png", + "caption": "Figure 1. Rigid body template", + "texts": [ + " ADAMS/Car is a professional module for simulation and analysis of the whole vehicle and various assemblies. Its abundant modeling capabilities and powerful kinematics and dynamics solver capabilities can help to establish a systemlevel simulation model with large scale and complex structure [6]. The integrated chassis control system studied in this paper is mainly made up of ABS, ESP and ARC, so there are some special structural requirements for the vehicle's front and rear suspension model, brake and body system. The rigid body template represents the base frame of a car (Fig. 1). The rigid body template defines a series of variables, most of which are used to compute the aerodynamic forces acting on the body. For a detailed description of the force function, see [7]. To feed the ABS ESP and ARC controller with input signals from the vehicle model, we will add virtual transducer signals in the body templates(see Table 1.). The disc-brake system template represents a device that applies resistance to the motion of a vehicle. The caliper part is mounted to the suspension upright, while the rotor is mounted to the wheel" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.20-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.20-1.png", + "caption": "Fig. 7.20 Case 1 of serial singularity: in that case, the axes of actuators 1 and 3 are aligned and if q\u03071 = \u2212q\u03073 (q\u0307i being the velocity of the actuator i), actuator 1 and 3 motions do not lead to a platform motion", + "texts": [ + " Let us rewrite this basis into two sub-bases written under a matrix form as: Bi = {Bia Bid} : [ 0Jai 0Jdi ] , where Bia : 0Jai groups the columns of 0Ji mi corresponding to the active joints and Bid : 0Jdi the columns corresponding to the passive joints. Three cases can then be met: \u2022 Case 1: the sub-system 0Jai is rank-deficient (this also corresponds to the degeneracy of the matrix Jtai in (7.87) and, as a result, to the degeneracy of the matrix Jta in (7.91))\u2014in that case, a displacement of the active joints of the leg does not necessarily bring a displacement of the end-effector along one given direction (Fig. 7.20). We will call them Leg Active Joint Twist System (LAJTS) singularities. \u2022 Case 2: the sub-system 0Jdi is rank-deficient (this also corresponds to the degeneracy of the matrix Jtdi in (7.87) and, as a result, to the degeneracy of the matrix Jtd in (7.91))\u2014in that case, a displacement of the passive joints of the leg does not necessarily bring a displacement of the end-effector along one given direction. Moreover, in such configuration, the leg instantaneously gains an uncontrolled motion (Fig. 7" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002681_amr.328-330.186-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002681_amr.328-330.186-Figure3-1.png", + "caption": "Fig. 3 Tooth profile of rack cutter Fig. 4 Coordinate systems for external gear generation", + "texts": [ + " Moreover, this method cannot evaluate the design of the tooth profiles, which usually use operational indexes of the gear couple, such as kinematics, kinetics and tribology. At present, mathematical analysis is the common approach for the study of conjugated tooth profiles, which generally derives the formula with the known formula of source profile or line of action. In this paper, tooth profile of the external gear is first derived with the given rack cutter and then tooth profile of the internal gear is obtained as the conjugated curve to the external one. Tooth Profile of Rack Cutter. Rack cutter is the common tool to generate external gears. As shown in Fig.3, the tooth profile of a rack cutter, which is designed for generating involute external gears, is composed of line ab, arc bc, line cd and line de. Coordinate system tS ( tx , ty ) is rigidly connected to the rack cutter with the tx axis and the ty axis coinciding with the pitch line and the midline of the tooth respectively. Here, p is the tooth thickness with a value of m\u03c0 , ah is the addendum with a value of * ah m , fh is the dedendum with a value of * * a( )h c m+ , \u03b1 is the pressure angel, m is the module, * ah is the addendum coefficient, and *c is the dedendum coefficient" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002291_s1064230713020056-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002291_s1064230713020056-Figure6-1.png", + "caption": "Fig. 6. ND and AD methods.", + "texts": [ + " 2 2013 AUTOMATION OF CONSTRUCTION OF CHARACTERISTIC CURVES 261 or UM = {\u2013UPOW, +UPOW, }) depending on the load torque and the PWM duty cycle. Therefore, the expression for the speed\u2013torque characteristics of the D method (Fig. 5) is a combination of descriptions of the speed\u2013torque characteristics for impulse modes II and IV: 7.4. ND and AD Methods The possible presence of impulse modes I and V on the winding is a specific feature of the ND and AD methods. Therefore, the mathematical description of the speed\u2013torque characteristics of the ND and AD methods (Fig. 6) is a combination of the corresponding expressions for impulse modes I and V: 7.5. SN and SA Methods When the SN and SA methods are used, there may be impulse modes I\u2013IV on the winding depending on the load torque, the PWM duty cycle, and the sequence of alteration of voltage pulse signs within the PWM period. Thus, the speed\u2013torque characteristics (Fig. 7) are described by the expression Tavg* 2\u03b3 1\u2013 \u03a9avg* for UM\u2013 +UPOW UPOW\u2013,{ },= 2\u03b3 1 \u03a9avg*+ \u03c4a \u2013 2e \u03b3\u03c4a 1\u2013 \u03a9avg*+ 1 \u03a9avg*+ for UMln +UPOW UPOW\u2013 , ,{ },= 2\u03b3 1 \u03a9avg*\u2013 \u03c4a + 2e \u03b3\u03c4a 1\u2013 \u03a9avg*\u2013 1 \u03a9avg*\u2013 for UMln +UPOW UPOW\u2013 , ,{ },=\u2013 2\u03b3\u2013 1 \u03a9avg* for UM\u2013+ UPOW\u2013 +UPOW,{ }" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001377_acc.2015.7172274-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001377_acc.2015.7172274-Figure3-1.png", + "caption": "Fig. 3. Schematic of inverted pendulum system.", + "texts": [ + " linear quadratic Gaussian (LQG) control, then after obtaining optimal gains, we may then compute the corresponding coprime factorization using the LQG controller, KLQG(s) = Y (s)\u22121X(s). The transfer function of the LQG controller is then used to construct initial Controller (CRCBode) or Youla (QBode) Robust Bode plots. An iterative loop-shaping design procedure may then be performed in order to \u201crobustify\u201d the initial controller by attempting to minimize the robust performance metric, \u0393, over all frequencies. The inverted pendulum system considered here, shown in Fig. 3, consists of a rigid link with total mass m and moment of inertia, Icm. The length from the pendulum pivot point to the center of mass is Lcm. The pivot point is allowed to move along the horizontal axis, x. The objective is to design a feed-back control law such that the link remains balanced in a upright position, \u03b8 = 0. The dynamics of the inverted pendulum system are given by Eq. (11) assuming the entire mass is at the location of the bob. Note that this transfer function has a pair of non-minimum phase zeros at s = 0 and an unstable/stable pole pair at s = \u00b1 \u221a g/L" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001887_humanoids.2014.7041377-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001887_humanoids.2014.7041377-Figure1-1.png", + "caption": "Fig. 1: A humanoid robot makes contacts c1 and c2 with the ground plane. The gravity wrench wg and the inertia wrench wi are applied to the robot. The contact wrenches w1 c and w2 c can have values in their friction cone. The robot is stable when w1 c +w2 c +wg +wi = 0.", + "texts": [ + ", a L i ] , the vector of contact variables of L potential contact points for phase i. In our objective function (4), CStability(qk) is the stability cost for the waypoint qk, which is defined in Equation (2). Though [28] uses a simplified physics model to make animated characters move naturally, we compute CPhysics(qk) accurately based on Newton-Euler equations. A key issue in our formulation is computation of physicsviolation cost CPhysics(qk) for maintaining dynamic stability (as shown in Equation (2)). We first describe our physicsbased formulation. Fig. 1(a) illustrates a high-DOF humanlike robot, which makes contacts with the ground plane using its feet. Let \u03a3R be the global coordinate frame, J be the number of links in the robot, and c1, ..., cL be the positions of L contact points. There are several wrenches (forces and torques) exerted on the robot. The robot is dynamically stable when all wrenches on the robot constitute an equilibrium [12]. 1) Contact wrench : The sum of contact wrenches wl c applied to the robot from contact points cl with respect to \u03a3R is given by wc = L\u2211 l=1 wl c = L\u2211 l=1 [ fl rl \u00d7 fl ] , (5) where fl is the contact force of cl and rl is the position vector of cl in the frame \u03a3R", + " First we formulate the combination of contact forces, which can be defined as: f = [fT1 , ..., f T L ]T . (11) Equation (5) can be represented as wc = Bf , where B is the corresponding 6\u00d73L matrix. Using this formulation, we solve an inverse dynamics computation problem, which computes f such that it satisfies the Coulomb friction constraints of Equation (6): f = arg min f\u2217 (\u2016Bf\u2217 + wg + wi\u2016+ f\u2217TRf\u2217). (12) The Coulomb friction constraint is usually converted to an inequality constraints, using a pyramid to approximate a friction cone Fi (shown in Fig. 1(b). The constraint for fi is reduced to \u2212 \u00b5fln \u2264 flt \u2264 \u00b5fln \u2212 \u00b5fln \u2264 flo \u2264 \u00b5fln. (13) In (12), we add the contact variable penalty term f\u2217TRf\u2217 as it is used in [28]. It increases the difference between f from (12) and the actual optimal force that satisfies (10), when contact variable al is small for a large contact force fi. The matrix R is a 3L \u00d7 3L diagonal matrix, and its diagonal elements correspond to Rjj = k0 (al\u03c1(k)) 2 + k1 , (14) where 3l\u2212 2 < j < 3l. k0 and k1 are constants that control the weight of the penalty cost" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003148_j.euromechflu.2011.08.006-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003148_j.euromechflu.2011.08.006-Figure1-1.png", + "caption": "Fig. 1. Sketch of two sheets swimming parallel with equal velocityU due to a plane wave displacement with phase velocity c = \u03c9/k on each sheet with phase shift \u03d5 = 0.", + "texts": [ + "2) In the low Reynolds number limit the flow velocity and pressure satisfy the Stokes equations \u03b7\u2207 2v \u2212 \u2207p = 0, \u2207 \u00b7 v = 0. (2.3) The fluid follows the motion of the sheets instantaneously. In particularwe consider the runningwave transverse displacement \u03be(x, t) = Aez sin(kx \u2212 \u03c9t) (2.4) with amplitude A, positive wavenumber k, and positive frequency \u03c9 for the lower sheet, and \u03be\u0302(x, t) = Aez sin(kx \u2212 \u03c9t \u2212 \u03d5) (2.5) with phase shift \u03d5 for the upper sheet. In a coordinate framewhere the fluid is at rest at z \u2192 \u00b1\u221e the sheets are caused to move in the negative x-direction. In Fig. 1 we show a sketch of two sheets swimming parallel with phase shift \u03d5 = 0. To second order in the amplitude A the swimming velocity of either sheet is U2 = \u2212U2sex with U2s given by Felderhof [2]: U2s(\u03ba, \u03d5) = 1 2 \u03c9kA2 [F(\u03ba) + H(\u03ba) cos\u03d5], (2.6) where F(\u03ba) = sinh2 \u03ba + \u03ba2 sinh2 \u03ba \u2212 \u03ba2 (2.7) with \u03ba = kL, and where H(\u03ba) = \u2212 2\u03ba sinh \u03ba sinh2 \u03ba \u2212 \u03ba2 . (2.8) Unfortunately the preceding minus sign was omitted in a previous publication [2]. We showed that to second order in the amplitude A the rate of dissipation in the fluid, averaged over a period of time T = 2\u03c0/\u03c9 and per unit area of sheet, is D2s(\u03ba, \u03d5) = 2\u03b7\u03c92kA2 [G(\u03ba) \u2212 2K(\u03ba) cos\u03d5], (2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003259_gt2013-94105-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003259_gt2013-94105-Figure5-1.png", + "caption": "Figure 5: Proposed sensor orientations (static)", + "texts": [ + " In their application to damage detection, a defect in or below the surface of the material presents a discontinuity in the conduction path, thereby disrupting the flow of eddy currents in the material, and altering their effect on the primary field and sensing coil. The depth to which defects can be detected depends on the depth of penetration of the induced eddy currents into the material. In the case of the active eddy current sensor, this is governed by the following equation: Therefore, varying the driving frequency of the electric coil generating the primary magnetic field can be used to control the depth to which the material (for example, a gear tooth), can be monitored. Figure 5 shows some orientations considered for static eddy current sensors for gear damage detection. These are as follows: A \u2013 Sensor located over gear tooth tips for tip timing and \u2018birdseye\u2019 view 4 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/asmep/76989/ on 07/08/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use B \u2013 Side view, various radial positions out from the gear hub C \u2013 Root inspection D \u2013 Cross-mesh sensor, for monitoring the mesh region (gear teeth under load) In Figure 6, sensors are positioned either radially (as shown) or axially offset, or a combination of both to determine irregularities in velocity transfer between mating gears" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001197_12.2189458-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001197_12.2189458-Figure1-1.png", + "caption": "Figure 1. Deformation.", + "texts": [ + " The temperature is considered staying stable which means the influence of temperature can be ignored, so the central Bragg wavelength shifts to the stain can be expressed as: ( )1B e B P\u03bb \u03b5 \u03bb \u0394 = \u2212 \u0394 (3) The paper supposes that all stain cause bending deformation, so the curvature to the stain can be expressed as: 2 s s s \u03b8 \u03b4\u03c1 \u03c1 + \u0394 = = + (4) where s is length of infinitesimal surface, \u03c1 is the radius of curvature, \u03b4 is the thickness of infinitesimal surface, s\u0394 is the length of infinitesimal surface shift, \u03b8 is the radius angle of infinitesimal surface shift. The deformation and all parameters are shown in the Figure1. The relation between stain and curvature can be expressed as: 0.5 2 s k s \u03b4\u03b5 \u03b4 \u03c1 \u0394 = = = (5) Proc. of SPIE Vol. 9524 95241H-2 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 11/13/2015 Terms of Use: http://spiedigitallibrary.org/ss/TermsOfUse.aspx where k is the curvature of infinitesimal surface. Considering the central Bragg wavelength shifts to the stain, the relation between central Bragg wavelength shifts to the curvature can be expressed as: ( ) 2 1 e B k P \u03bb \u03bb \u03b4 \u0394 = \u2212 (6) It shows that relation between central Bragg wavelength shifts to the curvature is linearity" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003649_cdc.2014.7039664-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003649_cdc.2014.7039664-Figure3-1.png", + "caption": "Fig. 3. Flexible joint robot link", + "texts": [ + " 2b respectively associated to the nonlinear R model and linear model with unknown input in resistance causality and Fig. 2c and Fig. 2d respectively associated to the nonlinear R model and linear model with unknown input in conductance causality. The nonlinearity which may affect the storage components (I, C) in integral causality can be redefined as proposed for the fault detection problem with a linear part and an unknown input. The system is linear time-invariant. Consider the nonlinear system defined as a flexible joint robot link, Fig. 3 studied in [24]. \u03b8m, \u03b8l , \u03c9m, \u03c9l denote the motor and link angular positions and velocities respectively. The dynamic equations are given by (8) with \u03b8\u0307l = \u03c9l , \u03b8\u0307m = \u03c9m. { \u03c9\u0307m =\u2212 1 Jm \u03c4\u2212 B1 Jm \u03c9m + 1 Jm u \u03c9\u0307l =\u2212B2 Jl \u03c9l + 1 Jl \u03c4 (8) Jm is the inertia of the DC motor, Jl is the inertia of the link, B1 and B2 are the viscous frictions. u the control input is considered as a torque (otherwise an amplifier gain is added) and \u03c4 is the torque due to the stiffening spring and is defined in equation (9), where \u03b31 and \u03b32 are positive constants" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003659_20131120-3-fr-4045.00042-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003659_20131120-3-fr-4045.00042-Figure3-1.png", + "caption": "Fig. 3. Quadrotor in the X-Y plane", + "texts": [ + " The strategy, as it is shown in figure 2, is based on the design of a control scheme for the translational dynamics so that tracking trajectory can be achieved; the resulting controller then provides an output which is the desired orientation to be fed in the orientation control scheme. A third control scheme is used for the quadrotors synchronization in the X-Y plane. The assignments qi = [xi yi] T , q\u0304i+1 = ei+1 and q\u0304i\u22121 = ei\u22121, are then made. The translation dynamics of the i-th quadrotor in the X-Y plane can be described by (see also Fig. 3) x\u0307i = vxi cos(\u03c8i)\u2212 vyi sin(\u03c8i) y\u0307i = vxi sin(\u03c8i) + vyi cos(\u03c8i) (12) \u03c8\u0307i = \u03c9i where vxi and vyi are the velocities in the xi and yi components of the body fixed coordinates frame B of the i-th quadrotor, and \u03c9i is the angular velocity of the yaw angle. As mentioned above, it is assumed that the dynamics of \u03c8i evolve much faster than those of xi and yi to a desired value (for example, by means of a proportional control); in fact, the dynamics of \u03c8i are decoupled from the dynamics of xi and yi in (12)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001792_9781118899076.ch5-Figure5.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001792_9781118899076.ch5-Figure5.2-1.png", + "caption": "FIGURE 5.2 The three-electrode cell configuration typically used for protein film electrochemistry experiments with a PGE working electrode surface. The electrode\u2013 potentiostat interface is denoted by asterisks.", + "texts": [ + " Instead, we focus here on the wealth of biochemical information to be gained purely from the electrochemical experiment and illustrate this with examples drawn from our own research interests. However, first, we outline some of the practicalities to be considered when performing PFE of enzymes. PFE is typically performed using a cell configuration incorporating reference, counter, and working electrodes. One example of a glass electrochemical cell frequently used for PFE in our laboratories is illustrated in Figure 5.2. The reference electrode is housed in a side arm connected to the sample chamber by a Luggin capillary that minimizes physical mixing of the solutions in the sample and reference chambers while defining a sensing point for the reference electrode near the surface of the working electrode. The sample chamber contains the platinum wire counter electrode and is shaped to minimize solution turbulence during rapid working electrode rotation as is frequently required during studies of redox enzyme catalysis", + " The working electrode is prepared from a cylinder (3 mm diameter) of the desired electrode material mounted onto a brass rod with silver-loaded epoxy and subsequently sealed into a nylon sheath with epoxy resin. The sheath allows the working electrode to be mounted onto the shaft of a rotor that makes electrical contact with the working electrode via the brass rod. During experiments, the working electrode is positioned in the sample chamber at the level of the Luggin capillary and with its exposed face in contact with a 3.5 ml solution (Fig. 5.2). The entire electrochemical apparatus is placed inside a Faraday cage within a N2-filled chamber to minimize electrical noise and the presence of oxygen, respectively. The temperature of the sample chamber is maintained by means of a jacket filled with circulating water under thermostat control. Several electrode surfaces have been found to support nondestructive yet robust protein adsorption [1\u20133]. We favor either the edge or basal plane of pyrolytic graphite. These surfaces have been successful in adsorbing a variety of enzymes such that they retain similar redox and/or catalytic properties to those described in solution, which are the usual criteria for confirming the functional integrity of adsorbed protein", + " The optimum composition of the solution for enzyme adsorption is usually found by exploring a range of ionic strengths, pH, buffer, and electrolyte identity in addition to enzyme concentration and where the inclusion of a coadsorbate, for example, positively charged neomycin, may help to secure an electroactive protein film. The time invested in exploring these conditions is rewarded when it comes to performing and analyzing experiments that probe enzyme function since both are greatly facilitated by the reproducible formation of stable films. Once the film is prepared, its catalytic performance can be readily assessed in a range of conditions using the cell illustrated in Figure 5.2. The design allows aliquots of reagents to be injected directly into the experimental solution from concentrated stock solutions, or the sample solution can be removed completely and replaced with a fresh solution of choice. Modification of the basic cell design allows for rapid equilibration of the sample chamber with gaseous agents of varying partial pressures [10]. In essence, the enzyme film can be subject to instantaneous dialysis by a variety of means and to great advantage as we illustrate later" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001918_bf00536604-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001918_bf00536604-Figure1-1.png", + "caption": "Fig. 1. Scheme of the system.", + "texts": [ + " The solution to be presented in this paper is carried out on a simplified system; several examples are used for elUcidatingthe effect Of the foundation tuning upon~the stability of the ~otor motion. 2. Solution of the Problem. Let us assume a one-disk rotor with mass m mounted on a massless shaft whose section at the place of the disk a t tachment has stiffnesses c 1 and c 2. The rotor rotates with a constant angular velocity ~o. For the sake of simplicity let us neglect the elasticity of the bearing, supports. Let us further assume tha t the mounting of the rotor upon the foundation can be represented in a simplified form by the scheme of Fig. 1, in which the foundation can be replaced by a mass mounted elastically upon massless springs whose total stiffness in the vertical direction is %1 and in the horizontal direction c02. Let us also assume t h a t the disk is located on the shaft midway between the bearings and that the foundation is completely symmetrical relative to the disk plane so far as both distribution of the mass of the base plate and the rigidity of the mounting is concerned; thus in standstill position the disk centre of gravity (assuming perfect balance) lies x A" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003293_chicc.2014.6896713-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003293_chicc.2014.6896713-Figure1-1.png", + "caption": "Fig. 1: Planar interception geometry.", + "texts": [ + " m is missile mass and Iyy is moment of inertia. T\u03b1 = \u03b1 \u03b3\u0307M is assumed as a known variable varying slowly [31], chosen as 3.75 seconds similar to [10] in our research. Fx, Fz andM are aerodynamic forces and moment, given by Fx = kF \u03c1V 2cx(\u03b1), Fz = kF \u03c1V 2cz(\u03b1,Mm), M = kM\u03c1V 2cm(\u03b1,Mm, \u03b4e), where kF and kM are constants determined by the missile geometry, \u03c1 is atmospheric density, cx(\u03b1), cz(\u03b1,Mm), cm(\u03b1,Mm, \u03b4e) are the aerodynamic coef cients which are approximated by well de ned af ne functions and can be found in [30]. In Fig. 1 a schematic view of planar interception geometry is shown. M denotes the missile and the target is denoted by T . VM , nL and \u03b3M denote the speed, normal acceleration and ight-path angle of missile, respectively. The speed, acceleration and ight-path angle of target are denoted by VT , AT and \u03b3T , respectively. \u03bb is the line of sight (LOS) angle between the LOS and the reference line. The planar missile-target engagement kinematics with gravity neglected are given by [10] r\u0307 = Vr, V\u0307r = V 2 \u03bb r +ATr \u2212 sin(\u03bb\u2212 \u03b3M )nL, \u03bb\u0307 = V\u03bb r , V\u0307\u03bb = \u2212V\u03bbVr r +AT\u03bb \u2212 cos(\u03bb\u2212 \u03b3M )nL (15) where r, Vr, \u03bb, V\u03bb denote the relative range, the relative velocity along LOS, LOS angle, the relative velocity perpendicular to LOS, respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002582_0954406211404854-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002582_0954406211404854-Figure1-1.png", + "caption": "Fig. 1 Typical layout of portion of a spatial linkage containing only turning pairs; incorporated are conventional Cartesian link-frames of reference", + "texts": [ + " While this development is welcomed and pleasing, it is also the case that such practitioners sometimes misunderstand concepts or findings in the literature of linkage kinematics, owing to related mathematical means adopted in their own disciplines. Consequently, particular care is taken below to explain the approaches used, even if some may be regarded as routine by scholars of linkage analysis. 2 BASIC KINEMATIC TOOLS The primary analytical techniques applied here have been used many times in the literature and so a brief recapitulation should suffice. The linkage to be examined, consisting of six revolutes, is definable by its joint variables i and fixed parameters Ti, ai i \u00fe 1, i i\u00fe 1, which are readily comprehended by recourse to Fig. 1. Relationships among these quantities in the closed loop are expressible in various ways, and provided in the Appendix, for the reader\u2019s convenience, is a universal set of displacement\u2013closure equations (11) to (22). A full explanation of their origin may be found, for example, in reference [9]. Trigonometric functions cosine and sine are denoted by c and s, respectively, and it will be convenient to signify 1 Proc. IMechE Vol. 225 Part C: J. Mechanical Engineering Science at MCMASTER UNIV LIBRARY on July 1, 2015pic" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002048_j.jappmathmech.2013.07.003-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002048_j.jappmathmech.2013.07.003-Figure1-1.png", + "caption": "Fig. 1.", + "texts": [ + " The axes of the first system f coordinates are constantly parallel to the corresponding O axes, and the axes of the second system are rigidly connected with the ody. The Gz axis is directed along the axis of dynamic symmetry of the body and is the principal central axis of inertia. We will define the osition of the body and the system of coordinates Gxyz rigidly connected to it relative to the translationally moving system G 1 1 1 by he three Euler\u2013Krylov angles: , and . We will specify the position of the rod O1O2 with respect to the system of coordinates O1 by he angles 1 and 1. The body, the corresponding axes and the angles 1 and 1 are shown in Fig. 1. We will assume that the suspension point moves periodically, and the acceleration of the motion is described by a piecewise-constant unction, so that Prikl. Mat. Mekh., Vol. 77, No. 2, pp. 202\u2013208, 2013. \u2217 Corresponding author. E-mail address: ocheretnyukeugen@ukr.net (Ye.V. Ocheretnyuk). 021-8928/$ \u2013 see front matter \u00a9 2013 Elsevier Ltd. All rights reserved. ttp://dx.doi.org/10.1016/j.jappmathmech.2013.07.003 b i a w h g It was shown in Ref. 1 that, when there are no parametric perturbations (k1 = k2 = 0), steady rotation of a dynamically symmetrical rigid ody suspended on a string will be stable for rotation with a high velocity, irrespective of the value of a \u2013 the z coordinate of the point O2 n the Gxyz system" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure4.17-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure4.17-1.png", + "caption": "Fig. 4.17. The Full Body Extender design (PERCRO, 2006)", + "texts": [ + " Arm Extender (1996) and BLEEX (Berkeley Lower Extremity EXoskeleton, 2004, [20]), both developed at Berkeley University, are separate upper- and lower-limb-strengthening devices. The control algorithm is based on an inner stabilizing position or velocity controller. A full body exoskeleton system is currently under development at PERCRO laboratory in the framework of a research project funded by the Italian Ministry of Defense. A schematic view of the actual mechanical design of the system is shown in Fig. 4.17. This paper has introduced the concept of robotic exoskeletons to be used as haptic interfaces for the control of manipulation tasks in teleoperation conditions or for the interaction with virtual environments. General definitions and functionalities of exoskeleton systems have been given by pointing out the advantages of such devices when there is the need of tracking the spatial configuration of the human arm for specific gesture recognition tasks or rehabilitation procedures. Control schemes for exoskeleton systems have been described" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000823_s13369-021-05594-8-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000823_s13369-021-05594-8-Figure6-1.png", + "caption": "Fig. 6 a Representation of bearing damage, b Representation of rotor bearing damages, c Representation of foot damages of motor", + "texts": [ + " In our study, it was considered to place screws of different geometries at the shaft, respectively, to represent any damages that may occur in the bearings and rotor bearings of single phase capacitor start motor. For this, a flywheel was made from cast polyamide assembled on the rotor; and 2 screw holes of 3\u00a0cm were drilled on the same to input the screws. It was considered that the bearing damages would vibrate the motor less than the damages in rotor and therefore, that the small screws placed on the flywheel would represent such a damage (Fig.\u00a06a, b). Reducing the quantity of part inside the flywheel would mean increasing the size of same kind of damage because the vibration would be reduced as the screw enters the flywheel. Furthermore, it was considered that removing the nut bolts of motor foot, respectively, would represent the damages that may occur at Fig. 4 Locations of VB83 vibration sensors on the motor 1 3 the connection points of motor to the chassis (foot damage) (Fig.\u00a06c). As seen in Fig.\u00a07, the length of screw placed in the flywheel is 3\u00a0cm (socket head cap screw) and 2.5\u00a0cm (set screw), respectively. The screw of 3\u00a0cm was defined as \u201c1st Screw\u201d, and the screw of 2.5\u00a0cm was defined as \u201c2nd Screw\u201d. In our study, the screws were first placed one by one and then in pairs in 35 different positions; and the vibration signals created for each position were acquired using VB83 vibration sensors in intervals of 1\u00a0ms for one minute. In the next stage, screws inside the flywheel were removed; and four nut bolts connecting the motor to the chassis were removed firstly one by one and then in pairs in 10 different ways; and the data obtained from vibration sensors in intervals of 1\u00a0ms were recorded" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002084_j.sbspro.2013.09.160-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002084_j.sbspro.2013.09.160-Figure1-1.png", + "caption": "Fig. 1. Bicycle\u2019s stand.", + "texts": [], + "surrounding_texts": [ + "At the end of the year, the students have to make an assembly drawing of a machine as well as the detail drawings of its component parts. Initially, students may opt between two choices. The first one is choosing which object they are going to work on: the stand of a bicycle or a ball-point pen (these mechanisms are provided by the teacher as material for the practical sessions). Besides, they have to decide which software applications they are going to use for their work (AutoCAD or Inventor). Once decisions have been made, the students create the drafts by analyzing the mechanism that they were given. The students also create a multimedia presentation of the work done. 2.1. Reverse engineering work: bicycle stand To make the drawings, the students take a series of steps in the reverse order to the process of design and assembly of a product. Instead of using an idea as the starting point for turning it into reality, the students work with a real product for analyzing its value according to improving possibilities (reverse engineering). The aims of this premise is getting students involved at work and motivating them to create a new, improved version of the product. Step 1. Each student takes the mechanism apart and numbers its component parts so that he/she can assemble it back correctly, using both real parts and the parts drawn in the software\u2019s assembly simulator (Figure 2). In this paper, we describe the work done using the Autodesk Inventor software that works with parametric design.Step 2. The students take the measurements of each part (Figure 3) and then sketch them by hand to study their forms and plan the graphic sketching using CAD software. Through the Inventor software, the students are able to draw each component part of the mechanism directly in 3D. Normally, students believe they save time and efforts when they measure and draw data from the real part directly into the computer. It is extremely enlightening seeing how they find out for themselves that\u2019s not the proper methodology, as it is way better planning the job from a sketch or a contour drawing. Step 3. Once the previous step has been completed, the students draw each part in the computer using Autodesk Inventor 2012 software (Su\u00e1rez et al, 2006). Step 4. Once that all parts have been drawn in 3 dimensions, the next step is assembling them in sequential order, seeing that if they fit together properly while checking that all measurements have been correctly acquired. If they are not correct, the students should make any necessary modification for solving the problem. Step 5. Once the mechanism has been drawn and assembled in 3 dimensions, the students use the same software for automatically generating the parts\u2019 drawing as well as the isometric drawing of the whole piece separated into its parts, the assembly\u2019s drawing and the detailed drawings of every component part. This is a critical point for the students because they believe that computer programmers always perform these processes correctly, but they soon realize that all software-generated drawings have to be revised so they may study and setup the graphic sketch of the original product correctly, which explains the need to learn all procedures as well as language of Engineering Graphics and, what is even more important, that it is not enough just owning this knowledge but knowing how to apply it properly making sense. Step 6. Finally, the drawings are printed at the right standardized scale and in the right format by following the rules of technical drawing. Once all drawings have been revised, they are printed in PDF files. The students make a multimedia presentation where they explain the whole process through hyperlinks to the PDF files." + ] + }, + { + "image_filename": "designv11_84_0003729_detc2011-48019-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003729_detc2011-48019-Figure5-1.png", + "caption": "FIGURE 5: STRAIN GAUGE POSITION.", + "texts": [ + " When the steering operation was given to the racing kart, the vertical load of each wheel varies according to the movement of the contact point of wheels, which brings the deformation of the frame. The actual deformation of the kart frame according to the steering operation was examined by an experimental way to validate the FEM analysis results. The kart with a driver was positioned on a level plane and vertical load transfers at different steering angle from -80 to 80 degree were measured by four load cells located under the wheels, as is shown in Fig. 4. In addition, a set of strain gauge was attached at point (a) in Fig. 5 to measure the principal strain of the frame. The experimental result of vertical load transfer is reported in the Fig. 6 (a). This measurement data of vertical load was used as the force input in FEM analysis, In addition, the mass of the driver was also considered as a force input. The maximum and minimum principal strain values at the measurement point are compared between the experimental and simulation result in Fig. 6 (b). The excellent agreement on the principal strain confirms that the FEM model is representative enough of the actual frame", + "org/about-asme/terms-of-use A modal analysis with the FEM model was performed by using ANSYS to evaluate the frequency characteristics of the kart frame. Eight eigenmodes in the frequency range up to 200 Hz are shown in Fig. 7. Next, the frequency response of the actual kart frame was evaluated by hammering test. The experiment was conducted assuming free-free constraint condition by hanging the frame in three points. In this experiment, the frame has been excited with the hammer at point (1) in Fig. 5, and the strain at point (b) was measured. The experimental layout is shown in Fig. 8. The power spectral density of measured strain was shown in Fig. 9. The frequency resonance about 45 Hz and 138 Hz is well observed, which is considered to correspond to the 1st and the 6th mode in Fig. 7. In addition, the peaks at the frequency of the 2nd and the 7th modes can also be found. On the other hands, other eigenmodes could not be found in this test. The hammering position, the sensing position, and the mode shape may effect the observability of the eigenmodes" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002011_2011-01-1691-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002011_2011-01-1691-Figure2-1.png", + "caption": "Figure 2. Overview of transfer paths", + "texts": [ + " In this study, element force is adopted for accurate estimation of transmitted force and is used the contribution of body structure. Visualization technique of contribution on the body structure is proposed for efficient contribution analysis. An SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 4 | Issue 2 1283 application is shown to be booming noise for the validation of the usefulness and effectiveness of the proposed VTA. This section describes the procedures for applying TPA to the FE model for a body-in-white. A cut section at the upper part of automotive body structure is configured on VT paths as shown in Figure 2. This structure is separated into two substructures at the boundary of this cut section. Internal forces transmitting the boundary (equivalent to transmitted force) and transfer functions defined from boundary points to an output point are calculated. A response at the output point x is expressed by two factors: transmitted force vectors R and the transfer function matrix H. In the Equation, \u201ci\u201d is the number of paths, and C is the contribution of the paths. (1) In TPA, product of transfer function and transmitted force indicates contribution of transfer path", + " It takes notice that each element adjoining the focused cutting section of body structure to be investigated plays the part of a transfer path. Calculation of contribution by element requires accurate and numerous transmitted forces in the contribution analysis. Transmitted force is obtained by the use of element force of which calculation is used the vibration response xc, xa of the FE model for the elements between the cut section and the adjacent section shown in Figure 3 that zooms in the red circle in Figure 2. The cut section \u201cc\u201d is expressed by dotdashed lines and the adjacent section \u201ca\u201d is expressed by dashed lines. Transmitted force vectors Rc, Ra in each DOF are derived by the following equation of motion for elements. (2) where m is mass matrix, k is stiffness matrix and x is response vector of the assembly. Based on the reciprocity theorem, transfer functions defined from boundary points to an output point are derived at once by frequency response results when analysis gives a unit impulse input to a response point [7]", + " Contribution vector of DOF of each nodal point is orthographically projected onto the response vector. The projected parts of these contribution vectors are replaced by the response vector of corresponding DOF and are represented by the replacement to displacement in the body deformation shape. Influence degree I of each DOF is expressed by the magnitude of the contribution A and an angle \u03b8 between the response vector and the contribution vector as follows. (3) Focusing on one cutting section, as shown in Figure 2, we can derive only partial characteristics of VT paths. As mentioned in the previous section, cutting sections for calculating the contribution are shown continuously. The proposed approach is applied for cutting sections of all focused paths, and the VT characteristics of all paths are visualized. Considering the body deformation diagram indicated by influence degree (VT characteristic diagram), we are able to take more appropriate measures and efficient derivation of measures by the VT characteristics and spatial trends of all paths offers", + " Analysis efficiency is raised by two-step application of the proposed techniques. The accuracy of the contribution calculation was verified by comparison of the frequency response result and the summation of the contribution by Equation (1). The summation approximately corresponds to the response near reviewing frequency as shown in Figure 8. This shows that the accuracy of the contribution calculation is sufficient to analyze on TPA or VTA. SAE Int. J. Passeng. Cars - Mech. Syst. | Volume 4 | Issue 21286 The contribution on the cutting section, as shown in Figure 2, is calculated to three VT paths. The front-side vibration path is on the windshields and front pillars. The rear-side vibration paths are in right- and left- hands on the rear quarter windows, center pillars and rear pillars. The vector diagram of the contributions of each path at 118Hz is shown in Figure 9. The front-side path is principal. The other paths whose vectors are almost orthogonal to the response are negligible to the front-side path. Therefore, the front-side path is focused to analyze" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002263_iccas.2014.6987984-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002263_iccas.2014.6987984-Figure6-1.png", + "caption": "Fig. 6 Geometrical relation for the guidance law", + "texts": [ + "2 Estimation method of wake trajectory If the target does evasion maneuver, the real wake trajectory can be a curve shape which is changing continuously. The basic concept of the trajectory estimation is to generate a virtual wake boundary trajectory by using the wake boundary points which the vehicle are passed on. A circle can be defined by using at least three points. It is defined to the circle as the virtual wake trajectory. If the vehicle moves through the wake boundary enough, the error of the real boundary and the estimated boundary can be small like Figure 6. The center and radius of the circle are calculated by the as follow: [;:J = A-I E, r = \ufffd( xk - xc l2 + (Xk - X, )2 where (4) If there is no inverse matrix of A, three points are located in the one single straight line. Numerically, it is considered that the virtual trajectory is the straight line at the condition of det (A) < E' \u2022 Because the target is also a ship, even though it does evasion maneuver, the curvature of the circles is not too large. Whenever the vehicle detects a new wake boundary point, the circle is updated" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001965_978-3-658-05978-1_7-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001965_978-3-658-05978-1_7-Figure5-1.png", + "caption": "Figure 5: Front and rear engine torque characteristics.", + "texts": [ + " As shown above, one main aspect when operating a split axel hybrid is the on demand torque enforcement. Compared to a classical four-wheel drive vehicle, the four-wheel features are present all the time in the BMW i8, but are strongly dependent on the driver inputs in relation to the vehicle road circumstances. This is realized by an adaptive control algorithm that estimates vehicle and road parameters while driving the vehicle. The records of drives on two different road conditions are shown in figure 5. On the left side a snowy track in northern Sweden with a friction level of about 0.35 is shown. The right side of figure 5 depicts a recording from a profile of a free and winding road in southern France under high mu conditions. Both rides were driven as quickly as possible. As it is demonstrated in figure 8 the all-wheel part of the snowy track is 25% compared to about 8% on high mu conditions. When operating the torque enforcement of vehicle on-demand, the non four-wheel parts of the track can be used for regenerating the battery\u2019s SOC by load point shifting instead of forcing the electric axle to ineffective electric boosts" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002155_00368791311292765-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002155_00368791311292765-Figure2-1.png", + "caption": "Figure 2 (a) Tilted and (b) oriented two lobe journal bearing", + "texts": [ + " (2010) also, in their recent work have studied the diagnosis of rolling bearings using MRA and neural networks. In the present work, keeping in view the efficiency and capabilities of ANNs, performance of gas lubricated noncircular five lobe journal bearing considering tilt and mount effects, is investigated. For this purpose, suitable structures of ANNs are designed and the needed computer programming is done. Figure 1 shows the schematic of a five lobe bearing system with other common lobed bearing systems. Also in Figure 2, as a sample, a tilted and oriented two lobe journal bearing, is shown. The difference in a symmetric and tilted configuration may be understood from Figure 2(a). In a symmetric configuration, the line joining the bearing geometric center and the center of each lobe passes centrally through the lobe, while in a tilted configuration, the lines of centers pass eccentrically through the lobes. This change on the location of lobe line centers, is called tilting and the corresponding angle in referred to as tilt angle (uT). The bearing mounting which is referred to as mount angle (uM), is the orientation of a bearing with respect to a fixed load direction" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001895_amm.288.208-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001895_amm.288.208-Figure3-1.png", + "caption": "Fig. 3 Simplified machine model machining Right-hand formatting gear", + "texts": [ + " 1 correspond to generation of concave and convex of gear tooth surface. Formatted method can be used in different machines, the traditional mechanical milling machine and the advanced CNC milling machine, whose mechanical structure and parameters are different, but the processing adjustment parameters can be obtained by the conversion of the adjustment parameters of a basic machine model. The gear tooth surface using different machines in the same kind of processing method is almost the same [3]. Fig. 3 illustrates the dextrose gear basic machine model in formatted method. In Fig. 3, \u03c3t [Ot; xt, yt, zt] is the cutter coordinate system; plane xtOtyt coincides with cutter plane; zt-axis is along the cutter axis; \u03c3m [Om; xm, ym, zm] is the machine coordinate system; plane xmOmym coincides with the cutter plane; \u03c3w [Ow; xw, yw, zw] is the gear blank coordinate system. According to the forming cutting principle, the double cutter blade cut directly the gear blank. The shape of profile and the track formed by the cutting edge of the tool are exactly same, so the gear tooth surface equation rw(u, \u03b8)can be obtained from the cutting surface equation: ),(),( 12 \u03b8\u03b8 urTTur tw \u2192\u2192 \u2022\u2022= " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003606_amm.339.510-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003606_amm.339.510-Figure4-1.png", + "caption": "Figure 4 Temperature field distribution nephogram of gear", + "texts": [ + " As is shown in figure 3, the calculation of boundary condition is shown in reference [5] . In the figure 3, GM area is the meshing surface;GT1, GT2, GT3 are separately top, root and no meshing surface of tooth; GD area is end surface; Gw area is bottom surface; Gj, GJ are sections surfaces of tooth. Besides, load average heat flux density and convection heat transfer coefficient of meshing surface on the pitch cylindrical surface. Solve the thermal analysis model and get the temperature field distribution nephogram of capstan and driven gear, as is shown in figure 4. From the figure, the temperature field distribution of standard involute cylindrical gear follows the rules: (1) It is shown from the result that the high speed and heavy load on the gear teeth produces the bigger temperature gradient, and the distribution style of each tooth on the same gear is the same. (2) On the direction of the teeth width, temperature field of gear approximately symmetrical distributes along the middle section surface of tooth width. (3) The high temperature area on the meshing surface is near the root and top of gear, the reason is that the product of surface relative sliding velocity here and contact pressure stress is too high, so more friction heat is needed" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.32-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.32-1.png", + "caption": "Fig. 2.32. Manipulator with spherical wrist", + "texts": [ + " Inspired by the previous solution to a three-link planar arm, a suitable point along the structure can be found whose position can be expressed both as a function of the given end-effector position and orientation and as a function of a reduced number of joint variables. This is equivalent to articulating the inverse kinematics problem into two subproblems, since the solution for the position is decoupled from that for the orientation. For a manipulator with spherical wrist, the natural choice is to locate such point W at the intersection of the three terminal revolute axes (Fig. 2.32). In fact, once the end-effector position and orientation are specified in terms of pe and Re = [ne se ae ], the wrist position can be found as pW = pe \u2212 d6ae (2.93) which is a function of the sole joint variables that determine the arm position17. Hence, in the case of a (nonredundant) three-DOF arm, the inverse kinematics can be solved according to the following steps: \u2022 Compute the wrist position pW (q1, q2, q3) as in (2.93). \u2022 Solve inverse kinematics for (q1, q2, q3). \u2022 Compute R0 3(q1, q2, q3)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003929_cyber.2013.6705474-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003929_cyber.2013.6705474-Figure1-1.png", + "caption": "Fig. 1. 2DTORA system on a slop", + "texts": [ + " In section II, the dynamics of the TORA is developed based on the Lagrange method, and the equilibriums is analyzed in detail. Based on the passivity of system\u2019s dynamics, total energy of the system is included in the control Lyapunov function design to yield a simple PD controller for the system in section III. In section IV, simulation studies on the dynamics and controller design are presented. Conclusions are presented in the last section. Comparing to the benchmark system TORA on the horizontal plane, TORA system on the slope is depicted as in Fig. 1. The cart of mass Mx is connected to a fixed base by a linear spring of stiffness kx. The cart is constrained to have one-dimensional translational motion with x denoting the travel distance. The actuated rotor attached to the cart has mass m and moment of inertia I about its center of mass, which is located a distance r from the point about which the rotor rotates. In Fig. 1, \u03c4 denotes the control torque applied to the rotor. Let x and x\u0307 denote the translational position and velocity of the cart, respectively. Define the rotor rotating away from the negative y-axis as the angular position \u03b8 and let \u03b8\u0307 denote rotating velocity of the rotor. The motion occurs on a slope having an angle \u03b2 with respect to the horizontal plane; therefore, gravitational forces has to be considered. Let g be the gravity constant and \u03b3 denote the heading angle of the TORA system. Since the TORA on a slope in the paper is a multi-body dynamical system, the Lagrange method is employed to derive its dynamics", + " We choose x, y, and \u03b8 as the generalized coordinates and \u03c4 as the generalized force. The Lagrangian L can be calculated as L = T \u2212 P (2) where T and P are the kinetic energy and potential energy of the system, respectively. Total kinetic energy T is a sum of kinetic energy Tx corresponding the equivalent mass of Mx and kinetic energy Tm of the rotor m. T = Tx + Tm (3) = 1 2 (Mx +m)x\u03072 +mr cos \u03b8x\u0307\u03b8\u0307 + 1 2 (mr2 + I)\u03b8\u03072 Let zero potential energy of the gravity be the center of the cart when the spring is with free status. According to Fig. 1, the mass points of cart o1 and rotor A on the slope can be depicted as in Fig. 2, where ol \u22a5 os, o1l1 \u2016 ol, F1o1F2 \u2016 os and a is the projection of point A on line os. The coordinates oxy is the same as the motion coordinates oxyz in Fig. 1. The circle with center o1 and radius r denotes the oscillating trajectory motion of rotor. Because point o is the zero potential energy of the spring, potential energy of gravity is positive at right side of line ol. The total potential energy is the sum of the energy stored in the spring Ps and gravity energy Pg: P = Ps + Pg = 1 2 kxx 2 +\u2212Mxg |op|+mg |oa| = 1 2 kxx 2 + (Mx +m)gx cos \u03b3 sin\u03b2 +mgr sin(\u03b3 + \u03b8) sin\u03b2 (4) where | \u00b7 | denotes the length of the line segment. Remark1 Normally, the angle \u03b2 of the slope is caused by the assembly or inherent slope of the installation plane" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002554_12.873101-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002554_12.873101-Figure2-1.png", + "caption": "Figure 2. Finite element discretization of eyelet-to-ferrule weld configuration.", + "texts": [ + " Different from the experiment, where the laser beam was fixed and the parts were rotating to make a circumferential welding, in the FE simulation, the parts were fixed and the laser beam was made rotating along the eyelet-ferrule interface. The heat input from the laser beam was simulated by a heat flux, applied normal to the individual laser welds. 3-D tetrahedral solid elements were generated in the mesh of the model. The total numbers of nodes and elements in the model were around 99,000 and 60,000, respectively. A FE discretization of the welding structure is shown in figure 2 below. A bond contact between the heat sink and the eyelet was assumed. An initial temperature of 300K was applied to each of the components. The boundary conditions were applied as follows: 1) on each of the exposed surfaces, heat convection was applied. The convection coefficient was 25 W/m2K and the bulk ambient temperature was 300K; 2) heat flux was applied on each of the welds, and was time dependent. The working time of each single heat flux was 2 ms \u2013 equal the laser pulse duration in the experiment" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003208_amm.446-447.678-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003208_amm.446-447.678-Figure1-1.png", + "caption": "Fig. 1. Main transmission system and mathematical model of linkage force increasing mechanism", + "texts": [ + " The velocity-displacement curve is inputted by user to meet specific technology requirements, which is one of the research hotspots about servo press [4]. In this paper, structural form of main servo motor-worm and worm gear commutation reducer-transmission screw-linkage force increasing mechanism is used as main transmission system, and how to implement velocity-displacement curve inputted by user should be discussed deeply. Mathematical Model of Linkage Force Increasing Mechanism. As shown in Fig. 1(a), structural form of main servo motor-worm and worm gear commutation reducer-transmission screw-linkage force increasing mechanism is used as main transmission system. In Siemens configuration software, main servo motor can be configured as linear axis by setting reduction ratio of worm and worm gear commutation reducer, diameter and lead of transmission screw such as mechanical system parameters, All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www", + "199, Purdue University Libraries, West Lafayette, USA-11/07/15,08:30:58) so angular displacement, angular velocity, angular acceleration of main servo motor are directly transformed into displacement, velocity, acceleration of nut [10,11]. Because linkage force increasing mechanism belongs to nonlinear transmission mechanism [12,13], the relationships of the velocity and displacement between nut and slide need to be researched. The mathematical model of linkage force increasing mechanism is shown in Fig. 1(b). With the point O as the coordinate origin and the direction of the coordinate system shown in Fig. 1(b), the original coordinates of the nodes are shown in Table I. Assuming that the edge OA rotates counterclockwise around the point O by angle \u03b8 to the edge OA', so the point B, C, D correspond to the point B', C', D'. However, subject to mechanical structure, the point B and B' are the same coordinates of X axis, as well as the point D and D', so the coordinates of the point A' is )cos360,sin360( \u03b8\u03b8 \u2212 , and the coordinates of the point B' is ))sin360(640cos360,0( 22 \u03b8\u03b8 \u2212\u2212\u2212 . shown in Eq. 1 and Eq" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003477_jjap.53.07kf26-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003477_jjap.53.07kf26-Figure1-1.png", + "caption": "Fig. 1. (Color online) 13-element linear array probe: (a) transducer layout and (b) photograph.", + "texts": [ + " The position of the needle tip was observed by B-mode imaging of a blood vessel phantom and animal experiments. Using the 128- element linear array probe, the position of the needle tip was also observed by B-mode imaging in animal experiments. The resolution of the image obtained with the 128-element probe was finer than that obtained by the 13-element probe. The human median cubital vein was observed, and from the results it is expected that the puncture of the vein while observing the needle tip in humans will be realized. Figure 1(a) shows the structure of the 13-element ultrasonic probe, and Fig. 1(b) shows a photograph of the probe. Six linear array elements are placed on both sides of a central ring element. The central through hole is 1.6mm in diameter, the inner and outer diameters of the ring transducer are 3 and 6mm respectively, and the element pitch and width of the six-element linear array transducers are 1 and 6mm, respectively. The transducer is made of a lead zirconate titanate (PZT)-epoxy 1\u20133 composite material (0.22mm thick) and is fixed on a damper using an epoxy adhesive. The driving frequency is 7MHz" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001396_romoco.2015.7219740-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001396_romoco.2015.7219740-Figure1-1.png", + "caption": "Fig. 1: The unicycle", + "texts": [ + " In this way (31) provides a sort of the estimate of the joint distance between the current trajectory and control in the system (2) and the initial trajectory and control. It might be expected that, by a skilful choice of the matrices Q(t) and R(t), one could possibly control either of these distances, especially as long as \u03b8 remains relatively small. This possibility will be confirmed by computer simulations presented in the next section. In order to verify the hypotheses stated in the previous section, we shall study a number of motion planning problems for the unicycle mobile robot that does not slip laterally, drawn in Figure 1. The robot kinematics are represented by the driftless control system q\u0307 = cosq3 0 sinq3 0 0 1 ( u1 u2 ) , y = k(q) = q. (32) The meaning of the state and control variables follows directly from the figure, and from (32). The output function equals the identity. It is assumed that the robot\u2019s initial state q0 = (0,0,0)T , the horizon of motion T = 2, and the convergence coefficient \u03b3 = 0.1. The motion planning problem will consist in transferring the robot to the desired point yd = (1,1,0)T . Controls are chosen in the form of truncated trigonometric series of order 5" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001968_2013-01-0424-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001968_2013-01-0424-Figure4-1.png", + "caption": "Figure 4. Virtual prototype model of the vehicle", + "texts": [ + " When the control system is connected (active), controller in Matlab/Simulink will return real-time anti-roll torque and this anti-roll torque becomes an active roll controller. When the control system is not connected (inactive), it becomes a passive roll controller (anti-roll bar), and the anti-roll torque is determined by torsion stiffness of anti-roll bar [8\u223c9]. The virtual prototype model of active roll controller and rear suspension are shown in Fig. 3. The final virtual prototype model of vehicle created using ADAMS/Mechatronics in Adams/car is shown in Fig. 4. ADAMS/Mechatronics is a plug-in to ADAMS which can be used to easily incorporate control systems into mechanical models. ADAMS/Mechatronics has been developed based on the ADAMS/Controls functionality and contains modeling elements which transfer information to/from the control system. The ABS, ESP and ARC integrated control systems are created using ADAMS/Mechatronics. Then each subsystem's simulation model is built in MATLAB/Simulink, respectively. At last, the ADAMS/Mechatronics and MATLAB/Simulink co-simulation is carried out for the integrated control system" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002428_amr.308-310.1893-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002428_amr.308-310.1893-Figure3-1.png", + "caption": "Fig. 3 Diagram of test-bed", + "texts": [ + " From table 2, the efficiency formula of differential gear train has been obtained: ( ) ( ) 0 1 1 1 1 H AB BH H AB BH i i i i \u03b2\u03b1\u03b7 \u03b7 \u2212 + = \u2212 + Among: 10 3 9 4 27 30 1.25 27 24 H AB Z Z i Z Z \u00d7 = = = \u00d7 Take all parameters into the efficiency formula, can be obtained: \u03b7=0.9529. Test Investigation of the Differential Gear Train Through the front theoretical derivation, the computation of the differential gear train efficiency has been known. In the actual circumstances, the relationship between efficiency and dynamic load has been analyzed by the test. The test has been adopted the JLCL-\u2161 type gear train innovation comprehensive test-bed, as show in figure 3. The test-bed is constituted by mechanical platform, electric parts and computer control. It could realize the measurement of torque and ratio of input and output. The transmission ratio and efficiency can be calculated base on the measurement data. The test drive system diagram as figure 2 shows, the gear train structure parameters for: Z1=Z12=20, Z2=Z3=Z6=30, Z4=Z5=24, Z7=60, Z8=82, Z9=Z10=Z11=27, Z13=29, m6=m7=m8=2.5, m1=m2=m3=m4= m5=m9=m10= m11=m12=m13=2. Motor 1 input is used by a fixed axis gear train which composed by gear 1 and 2, and motor 2 is also used by a fixed axis gear train which composed by gear 6 and 7", + " When transmission ratio is definite of the differential gear train, and assigns the size and the direction of torque of a basic component, then the torque's size and the direction of other both are also determined only, have nothing to do with the ambient condition. 2. The efficiency of the differential gear train is not a fixed value, and change along with the change of the torque and rotational speed. 3. In the situation of fixing torque and changing rev, when the rotational speed of differential gear train is higher, the efficiency is lower. The system composition diagram of the fluctuating loading test-bed is show in figure3. The test efficiency changes along with torque and rotational speed. According to the highest mechanical efficiency of test results, 61.16% has been used to calculate test-bed efficiency. Choose the 10th group data of fluctuating loading under fixing torque and changing rev, as show in table 6. This test power transfer is both parallel and series connection, and there is series in parallel connection, so the power transfer mode is mix league. Two electric motors power first through parallel gear input, then in 2K-H differential gear train to mixing, finally through a fixed axis gear train output to the magnetic powder brake as end load" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002241_1.4007860-Figure17-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002241_1.4007860-Figure17-1.png", + "caption": "Fig. 17 Contours of (a) total deformation and (b) von-Mises stress distribution in the workpiece", + "texts": [], + "surrounding_texts": [ + "We have successfully developed a numerical model for the coupled field analysis of GTA welding process. The present study integrates the weld pool dynamics field with the other fields of a welding process. The present model takes into account all the major weld pool dynamics driving forces in the weld depth, width, and shape predictions. All the arc parameters affecting the heat flux and current density distributions have been incorporated into the model. The coupled weld pool dynamics-thermal fields provide a more accurate thermal energy distribution. Through material modeling the effects of microstructure evolution on mechanical behavior is also indirectly incorporated. The arc heat flux and current density distribution on the workpiece, and the weld pool depth/width ratio have been compared with existing 011010-10 / Vol. 5, MARCH 2013 Transactions of the ASME Downloaded From: http://thermalscienceapplication.asmedigitalcollection.asme.org/ on 03/09/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use literature. They agree well. Thermally induced stress distribution and workpiece deformations have been reported. The mathematical framework developed here provides us with the tools to study the multiphysics problem of welding. A future goal is to study in detail the effects of welding parameters on the weld D/W ratio. The obtained trends can then serve as guidelines for better choice of welding parameters for achieving the desired weld D/W ratio. It is desired to conduct parametric studies of the effects of welding direction and speed with reference to workpiece clamping on the stress evolution. Studying the effect of preheat and other stress relief methods on welded joints is also being planned." + ] + }, + { + "image_filename": "designv11_84_0003641_peds.2013.6527115-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003641_peds.2013.6527115-Figure5-1.png", + "caption": "Figure. 5. Flat magnet structures.", + "texts": [ + " The rotor structures are, for example, a flat structure that has circumferentially arranged magnets, a V-shaped structure that has an opening angle of magnets at thier center position, and a circular arc magnet structure. 1.62 0.982 0.516 0.628 0.0 0.5 1.0 1.5 2.0 2.5 S-F1 P-F1 Tr Tm To rq ue T [N m ] T TIe = 3 A2.14 1.61 Figure. 4. Maximum torque characteristics. Figure. 1. Analysis model of the IPMSM. 0.53 Permanent magnet (b) Basic Rotor Structure 0.7 Flux-barrier (a) Analysis model Stator core Rotor core 112 The rotor structures with flat magnets are shown in Fig. 5. The P-F2 model and the P-F3 model are two-layer and threelayer structures, respectively, with flat magnets. The thickness of the magnets is 3 mm in the P-F1 model, 1.5 mm in the P-F2 model, and 1.0 mm in the P-F3 model. In this regard, the PM volume is the same in each layer. The width of the flux path between PMs is 4.5 mm in the P-F2 model and 4.0 mm in the P-F3 model. The increase in the number of layer is expected to cause a decrease in Ld and an increase in Tr The motor parameters and the torque characteristics of the flat structures are shown in Figs" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.9-1.png", + "caption": "Fig. 2.9. Representation of Roll\u2013Pitch\u2013Yaw angles", + "texts": [ + " In fact, if \u03d1 = 0, \u03c0, the successive rotations of \u03d5 and \u03c8 are made about axes of current frames which are parallel, thus giving equivalent contributions to the rotation; see Problem 2.2.4 Another set of Euler angles originates from a representation of orientation in the (aero)nautical field. These are the ZYX angles, also called Roll\u2013Pitch\u2013 Yaw angles, to denote the typical changes of attitude of an (air)craft. In this case, the angles \u03c6 = [\u03d5 \u03d1 \u03c8 ]T represent rotations defined with respect to a fixed frame attached to the centre of mass of the craft (Fig. 2.9). The rotation resulting from Roll\u2013Pitch\u2013Yaw angles can be obtained as follows: \u2022 Rotate the reference frame by the angle \u03c8 about axis x (yaw); this rotation is described by the matrix Rx(\u03c8) which is formally defined in (2.8). \u2022 Rotate the reference frame by the angle \u03d1 about axis y (pitch); this rotation is described by the matrix Ry(\u03d1) which is formally defined in (2.7). \u2022 Rotate the reference frame by the angle \u03d5 about axis z (roll); this rotation is described by the matrix Rz(\u03d5) which is formally defined in (2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002494_amr.1028.105-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002494_amr.1028.105-Figure5-1.png", + "caption": "Fig. 5 Total deformation diagram of the drive shaft", + "texts": [ + " 1) The cylindrical surface constraints are applied to the contact surfaces of the bearings and the shaft. 2) The gravity of synchronous belt and belt pulleys is imposed on the central drive shaft. Synchronous belt is chose as PU belt whose density is 1.25g/cm 3 , meantime, belt pulleys are chose as aluminum alloy whose density is 2.7g/cm 3 . The total gravity is 96.7N. 3) The fixed constraint is applied to the keyway of the shaft. By the analysis of ANSYS Workbench, the stress contours (shown in Figure 4) and overall deformation contours (shown in Figure 5) can be got under the gravity of belt and pulleys. The static structural analysis of the drive shaft is used for pre-stressed modal calculation in Model module, therefore the natural frequencies and corresponding vibration modes of the shaft can be got. Normally we do not have to find all the natural frequencies and mode shapes. Low order natural frequencies and vibration modes have bigger impact on the vibration of the stepper motor [4],so the first six natural frequencies and vibration modes obtained in ANSYS Workbench are concerned, as shown in Figure 6-11 and Table 1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002594_kem.462-463.780-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002594_kem.462-463.780-Figure3-1.png", + "caption": "Fig. 3 The traction distributions within the contact area for radius of", + "texts": [ + " The devised model consists of a flat-ended cylinder, whose height/diameter ratio is \u00be. Note that other aspect ratios had been used for the simulation studies but no significant differences were found. Assume that 50% of the area is in contact with the substrate, which means \u03b1 = 0.7. In addition, the values employed for other parameters are the same as those stated above, i.e., The distributions of traction, \u03c3zz, within the contact region at the critical pull-off force are obtained for different cylinder radii. Fig. 3 shows that the traction becomes more uniform with decreasing radius. Below the critical size of 135nm obtained from various simulation results, the stress concentration near the edge of the contact area completely vanishes and the adhesive structure maintains a state of uniform contact stress in spite of the crack-like flaw around the outer edge. The fact that the theoretical strength can be achieved regardless of surface imperfection may be used to explain why gecko\u2019s spatula is evaluated at the scale of nanometer" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002336_icsd.2013.6619912-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002336_icsd.2013.6619912-Figure2-1.png", + "caption": "Fig. 2. Detailed model of three phases", + "texts": [ + " With the purpose of improving calculation accuracy, the model is established consistent with the actual situation as far as possible, the tower, tension string, suspension string, fittings, etc were all involved in the simulation. The tower using split structure; it is composed of two main towers and one vice tower. The main tower, with a height of 100 m, is on the side, while the vice tower is in the middle, with a height of 98 m. The distance between main tower and vice tower is 33 m. The jumper used in this tower is eight bundle, the subconductor diameter is 33.26 mm and the bundle 978-1-4673-4461-6/13/$31.00 \u00a92013 IEEE diameter is 400 mm. The jumper is modeled as smooth conductors. As shown in Fig. 2, the three phases were named with A, B and C from top to bottom for convenience. Phase A, which arranged outside the main tower, contains 5 sections of jumper. Phase B and C are arranged between vice and main tower, both of them contain 3 sections of jumper. Insulator strings are used to connect jumper and tower. All electromagnetic field problems can be regarded as starting from Maxwell\u2019s equations, and solved as boundary value problems described by the partial differential equations and boundary conditions [9]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure2-1.png", + "caption": "Figure 2 CYLINDRICAL HOB CUTTER", + "texts": [ + " The methods of manufacture associated with bevel and hypoid gears do not allow these gears to be treated with the same type of geometric considerations that currently exist for cylindrical gears. To illustrate, spur cylindrical gears are helical gears with a zero helix angle and both gear types are produced using the same machine. Spur hyperboloidal gears cannot be produced using existing fabrication techniques for spiral hyperboloidal gears. Depicted in Figure 1 is a cylindrical hob in mesh with a spur gear. Depicted in Figure 2 is a cylindrical hob cutter with gashes or cutting flutes. Here, only the motion of the hyperboloidal hob cutter is considered. 1 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use The majority of hypoid and bevel gear manufacture today is the focus of The Gleason Corporation and KlingelnbergOerlikon. Today, the following three companies provide the machines and machine tools for the production of crossed axes gear pairs: The Gleason Works\u00f1 Klingelnberg-Oerlikon\u00f1 Yutaka Seimitsu Kogyo, LTD" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001223_med.2015.7158762-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001223_med.2015.7158762-Figure2-1.png", + "caption": "Fig. 2. Quadrotor model", + "texts": [ + " Then one can define the dilation matrix Dr(\u03bb) = diag{\u03bbri} for i = 1, n and Dr(\u03bb)x = (\u00b7 \u00b7 \u00b7 , \u03bbrixi, \u00b7 \u00b7 \u00b7 ) represents a mapping x 7\u2192 Dr(\u03bb)x called a dilation. Definition 1 A function h : Rn \u2192 R (vector field f : Rn \u2192 Rn) is r-homogeneous of degree m if h(Dr(\u03bb)x) = \u03bbmh(x) (f(Dr(\u03bb)x) = \u03bbmDr(\u03bb)f(x)) for all x \u2208 Rn and all \u03bb > 0. Theorem 1 [21] Let continuous r-homogeneous with a negative degree vector field f : Rn is defined on Rn and the origin of the system (1) is locally attractive then it is globally finite-time stable. Schematic representation of quadrotor is shown in Fig. 2 is obtained using Newton\u2019s laws and the Lagrange equations. Quadrotor dynamics is described by the system of six second order nonlinear differential equations [1]: x\u0308 = (sin\u03c8 sin\u03c6+ cos\u03c8 sin \u03b8 cos\u03c6) U1 m , y\u0308 = (\u2212 cos\u03c8sin\u03c6+ sin\u03c8 sin \u03b8 cos\u03c6) U1 m , z\u0308 = \u2212g + (cos \u03b8 cos\u03c6) U1 m , \u03c6\u0308 = Iyy \u2212 Izz Ixx \u03b8\u0307\u03c8\u0307 \u2212 U2 Ixx , \u03b8\u0308 = Izz \u2212 Ixx Iyy \u03c6\u0307\u03c8\u0307 \u2212 U3 Iyy , \u03c8\u0308 = Ixx \u2212 Iyy Izz \u03c6\u0307\u03b8\u0307 \u2212 U4 Izz , (2) where x, y, z \u2013 Cartesian coordinates of quadrotor, \u03c6, \u03b8, \u03c8 \u2013 Euler angles (\u03c6 \u2013 pitch, \u03b8 \u2013 roll, \u03c8 \u2013 yaw), Ixx, Iyy, Izz \u2013 diagonal elements of quadrotor inertia tensor, m \u2013 mass of quadrocopter, g - gravitational acceleration, U = (U1, U2, U3, U4) \u2013 virtual control forces, constrained with forces generated by UAV actuators U1 = b(\u21262 1 + \u21262 2 + \u21262 3 + \u21262 4), U2 = b(\u2212c1\u21262 2 + c2\u21262 4), U3 = b(\u2212c3\u21262 1 + c4\u21262 3), U4 = d(\u2212c5\u21262 1 + c6\u21262 2 \u2212 c7\u21262 3 + c8\u21262 4), (3) where \u2126 = (\u21261,\u21262,\u21263,\u21264) \u2013 rotor velocities of the motors (propeller thrust is proportional to square of its angular velocity), b, d \u2013 some physical constant, that generally identified experimentally, ci, i = 1, 8 \u2013 moment arms" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001979_amm.284-287.461-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001979_amm.284-287.461-Figure5-1.png", + "caption": "Fig. 5 Non-linear stiffness model with mesh period Fig. 6 Shaft element and coordinate system", + "texts": [ + " (5) The pressure angle and the gear position angle in this paper are affected by the translational motions of the gear pair while the previous studies neglected these effects. According to the above assumption that the gear mesh is modeled as the equivalent stiffness and damping with transmission error, the gear mesh deformation along the pressure line can be written as tddpddpdd eRRWWVVt )()cos()()sin()()( 22111212 . (6) Thus, the gear mesh force along the pressure line can be expressed as mmh cktF )( . (7) The mesh stiffness model in Kahraman et al. [11] is used in this paper, and is described as a periodic function with mesh period mT in Fig. 5. The succeeding pair of teeth should contact immediately when one pair of teeth lose the contact with each other. The contact ratio pm is presented to measure this overlapping action. There are two pairs of contacting teeth when time t is in the range from mTn )1( to mp Tnm )2( , and uk is the maximum mesh stiffness in this mesh process. There is only one pair of contacting teeth if t is in the range from mp Tnm )2( to mnT , and lk is the minimum mesh stiffness. The definition of mT and pm are defined as 2211 /2/2 NNTm " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure5.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure5.2-1.png", + "caption": "Fig. 5.2. Angular disparity from A and B", + "texts": [ + " In other experiments carried out by Ian Howard [12] limits between 4\u25e6 \u2212 7\u25e6 for crossed disparity and between 9\u25e6 \u2212 12\u25e6 for uncrossed disparity are given. According to Howard, variability of these findings result from numerous factors affecting the proper image blending, such as scene lighting conditions, contrast between objects or image exposure duration. To evaluate the effect of human disparity perception limit in the teleoperation working range, disparity is calculated as a function of object to camera distance. Fig. 5.2 shows a situation where cameras are fixating on point A at a distance d, in a symmetric vergence configuration, where \u03b1 is the angle of camera axes. Let point B be located in the cyclopean axis at a distance x under an angle \u03b2. In the right eye the optical ray from B forms a \u03b81 angle and in the left eye it forms a \u03b82 angle, with opposite values (\u03b81 = \u2212\u03b82 = \u03b8). If the distance between the eyes is represented by o, then: tan \u03b1 2 = o 2d (5.1) tan \u03b2 2 = o 2x (5.2) 5 Stereoscopic Image Visualization for Telerobotics 81 The angular diference is thus given by: \u03b8 = \u03b1\u2212 \u03b2 2 = arctan( o 2d )\u2212 arctan( o 2x ) (5" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.14-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.14-1.png", + "caption": "Fig. 7.14 Coupler surface for a 3\u2013RPR PPM when leg 1 is disconnected and actuators of legs 2 and 3 are fixed", + "texts": [ + " It is necessary to mention that the present equations miss some degenerated kinematic conditions that appear when R V \u2212 S U = 0. This problem is investigated in Wenger et al. (2007) and the interested reader should have a look at the mentioned paper. From the geometric point of view, one could observe that, when the first leg of the robot is disconnected from point A13 and for constant values of the actuated variables q22 and q32, the resultingmechanism is equivalent to a passive four-bar linkagewhose legs can freely rotate around the R joints located at A21 and A31 (Fig. 7.14). As a result, the curve drawn by the point A13 (also called the coupler curve) when the four-bar linkage is freely moving is a sextic curve S , i.e. an algebraic curve of degree 6. As point A13 also belongs to the first leg, it moves on a circle C centered in A11 and of radius q12 when the leg 1 actuated joint is fixed. The circle C is the vertex space of the leg 1. As a result, the solutions of the FGM of the 3\u2013RPR PPM are at the intersection of the sextic curve S and the circle C and at most 6 solutions may exist (the proof is given in Merlet 2006b) (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000821_amm.10-12.900-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000821_amm.10-12.900-Figure1-1.png", + "caption": "Fig. 1 The structure of motorized spindle Fig. 2 The FEM model in dynamic characteristics", + "texts": [ + " Along with the increasing request of design quality, in order to set up a proper dynamic model to obtain the analytical result that meets motorized spindle, every influencing factor must be taken into account, especially the influence of centrifugal forces and gyroscopic moments on the motorized spindle shaft. In this paper, the FEM model of motorized spindle is set up to research on its dynamic characteristics, which is effected by the axial preload on the natural frequency, the motorized spindle\u2019s natural frequencies and corresponding vibration shapes, centrifugal forces and gyroscopic moments on the motorized spindle shaft. The mass distribution of rotor, the non-linearity of bearing rigidity and the whirling motion of spindle are taken into account in modeling. Fig. 1 shows the structure of motorized spindle. The spindle parts are supported by two sets of angular contact ceramic ball bearing in front and back which are installed back to back. The front bearing is fixed, and the back bearing floats along the axial direction. The motor is set between two bearings, All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 169.232.111.167, UCLA EMS Serials, Los Angeles, USA-05/04/15,06:44:15) and the rotor connects with spindle through shrink fitting. As one whole body, the stator and collant jacket are installed to the shell of spindle. The spindle in Fig.1 is a kind of multi-diameter with the characteristics of hollow and multi-support. There are many kinds of load on spindle, including the cutting force and moment of flexion applied to the front end of spindle, the forque transferred from rotor to spindle, etc. Supported by four bearings, the spindle rotates at a high speed. Therefore, it is regarded as a more complicated structure of statically indeterminate beam. For the structure of spindle is symmetry and its shape is simple, some measures are done to simplify the calculation when modeling in FEM, including treating the spindle as the space spring beam, treating the angular contact ceramic ball bearing as a radial compression spring mass unit, treating the rotor, jacket and socket as the axial material with same density, equaling to node in relevant unit as the additional distributing mass in spindle" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.12-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.12-1.png", + "caption": "Fig. 2.12. Conventional representations of joints", + "texts": [ + " Analogously to what presented for the rotation matrices, it is easy to verify that a sequence of coordinate transformations can be composed by the product p\u03030 = A0 1A 1 2 . . .A n\u22121 n p\u0303n (2.47) where Ai\u22121 i denotes the homogeneous transformation relating the description of a point in Frame i to the description of the same point in Frame i\u2212 1. A manipulator consists of a series of rigid bodies (links) connected by means of kinematic pairs or joints. Joints can be essentially of two types: revolute and prismatic; conventional representations of the two types of joints are sketched in Fig. 2.12. The whole structure forms a kinematic chain. One end of the chain is constrained to a base. An end-effector (gripper, tool) is connected to the other end allowing manipulation of objects in space. From a topological viewpoint, the kinematic chain is termed open when there is only one sequence of links connecting the two ends of the chain. Alternatively, a manipulator contains a closed kinematic chain when a sequence of links forms a loop. The mechanical structure of a manipulator is characterized by a number of degrees of freedom (DOFs) which uniquely determine its posture" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002782_ecj.10179-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002782_ecj.10179-Figure2-1.png", + "caption": "Fig. 2. Experimental model of inverted pendulum.", + "texts": [], + "surrounding_texts": [ + "A simplified motion diagram of a rotary inverted pendulum and its experimental model are shown in Figs. 1 and 2, respectively. The rotary arm of the first link is coupled directly with a DC motor (100 V, 250 W) and the pendulum portion can be extended or contracted by a mini DC motor (24 V, 6.4 W) via a rack-and-pinion mechanism. The equation of motion for the controlled object in Fig. 1 can be derived by using the Euler\u2013Lagrange equation as follows: Here \u03b8, \u03b8 . , and \u03b8 .. are respectively the angle, angular velocity, and angular acceleration, \u03c4 is the torque, M(\u03b8) is the inertial term, C(\u03b8, \u03b8 . )\u03b8 . is the nonlinear term, G(\u03b8) is the gravity term, B\u03b8 . is the viscous friction term, and D(\u03b8 . ) is the Coulomb friction term: Here J1 = I1 + m1a1 2 + m2l 2, l = l1/2, J2 = I2 + m2a2 2, r = m2a2, S2 = sin\u03b82, C2 = cos\u03b82. In addition, g is the gravitational acceleration, \u03c41 is the input torque from the motor, b1 and b2 are viscous friction coefficients, and d1 is the Coulomb fiction coefficient of the arm. The Coulomb friction in the pendulum part is assumed negligibly small." + ] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure9.34-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure9.34-1.png", + "caption": "Fig. 9.34 Folding mechanism", + "texts": [ + " The evaluation of the lumbar curve using the sagittal arrow test was performed by determining the distance between the vertical plumb line with the back, as a final result a straight line is obtained, which is one of the most used tools to determine if the postural reference points of a person are aligned, as shown in Fig. 9.33. According to the values obtained from the measurement, lordosis can be classified as: \u2022 Hyperlordosis, if the value is greater than 35 mm. \u2022 Normal lordosis, if the value is between 20 and 35 mm. \u2022 Hypolordosis, if the value is less than 20 mm. According to these results, it was possible to determine the operation of the sensor and the necessary length for its placement. Based on the design of the folding mechanism shown in Fig. 9.34, the idea of designing an articulated system that can perform vertical movements to cover the entire back of the patient was developed [25\u201330]. 9 Design of an Auxiliary Mechanical System for the Diagnosis \u2026 251 This mechanism was chosen due to its useful features such as portability, low cost, easy handling and its wide range of dimensions for different patients. Likewise, with a digital post-processing method, it is expected to achieve a spine deformation result considerably close to the obtained result through the use of medical images [31\u201333]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure13.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure13.2-1.png", + "caption": "Fig. 13.2. Standard linear models: Kelvin-Voigt (a) and Zener (b)", + "texts": [ + " In this case, a number of mechanical models, composed by linear springs and linear viscous dampers arranged in different configurations (e.g. Maxwell and Kelvin models), can be adopted. These models are widely used because of their simplicity and their clear physical interpretation [13], [4]. In this case, a general representation of the force-displacement relation is given by ordinary differential equations with constant coefficients, which depend on the 214 L. Biagiotti and C. Melchiorri model structure as well as on the considered material. The simplest example of linear model is the Kelvin-Voigt contact model (see Fig. 13.2.a), composed by the mechanical parallel of a linear spring and a damper. If F is the force exerted by the material on a probe during contact, the linear model is expressed by: F (t) = { \u03ba\u03b4(t) + \u03bb\u03b4\u0307(t) \u03b4 \u2265 0 0 \u03b4 < 0 (13.1) where \u03b4\u0307 is the penetration velocity of the probe and k and \u03bb are the elastic and viscous parameters of the contact. When the visco-elastic behavior of the material/object under analysis is remarkable, more complex models are available. A standard linear viscoelastic solid model (Zener model) is shown in Fig. 13.2.b. Such a model is frequently adopted to represent behaviors in which instantaneous and delayed elasticities arise. Force and displacements are related by the following differential equation: F + \u03bb \u03ba1 + \u03ba2 F\u0307 = \u03ba1 \u03ba1 + \u03ba2 \u00b7 (\u03ba2\u03b4 + \u03bb\u03b4\u0307) \u03b4 \u2265 0 (13.2) where the meaning of the parameters \u03ba1, \u03ba2 and \u03bb is shown in Fig. 13.2.b. Lumped Parameters Systems with Non-linear Coefficient: The Hunt-Crossley Case When the linear hypothesis is not applicable, non linear models must be considered. For the sake of simplicity, many authors developed dynamic models of contacting systems drawing inspiration from classical linear models and introducing nonlinearities in the definition of the spring stiffness and of the dashpot viscosities [14, 15]. A noteworthy nonlinear model has been proposed by Hunt and Crossley [16] and adopted by several authors [17,18,19] to describe the dynamic behavior of the interface between a robot and the environment when a contact occurs" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003172_sii.2014.7028011-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003172_sii.2014.7028011-Figure3-1.png", + "caption": "Fig. 3. Coordinate systems of the wrist.", + "texts": [ + " B AR is the rotational matrix between reference coordinate A\u03a3 and wrist coordinate B\u03a3 , iP Br is iP vector based on B\u03a3 , )=( 41ii\u03c9 is wire length between iP and iQ , iQ Ar is a iQ vector based on A\u03a3 . From the equations (10) and (11) inverse kinematics can be obtained. 978-1-4799-6944-9/14/$31.00 \u00a92014 IEEE 59 Conversely, it is difficult to find analytical solutions of forward kinematics, hence the numerical Newton-Raphson method is employed to solve the forward kinematics [7]. A geometrical model is shown in Fig. 3 Considering the interference of joint, the motor angle related to the wrist joint can be described as follows: 4422 1 5 \u03b8k\u03b8k r \u03c9 f m --=)(\u03b8 (12) 4422 2 6 \u03b8k\u03b8k r \u03c9 f m -+=)(\u03b8 (13) 4422 3 7 \u03b8k\u03b8k r \u03c9 f m +=)( -\u03b8 (14) 4422 4 8 \u03b8k\u03b8k r \u03c9 f m ++=)(\u03b8 (15) C. Effect of twist joint for single wire To reduce length fluctuation of wire caused by twist joint, the wire route is located close to the center of rotation by using a troidal shaped PTFE (Poly Tetra Fluoro Ethylene) at shoulder joint 1\u03b8 and twist joint 3\u03b8 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003678_1.4024704-Figure19-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003678_1.4024704-Figure19-1.png", + "caption": "Fig. 19 Total deformation (in meters) of the workpiece. Welding toward the fixed face CC0D0D.", + "texts": [ + " The total deformation of the workpiece is shown in Fig. 15. For the second constraint configuration, face CC0D0D is fixed. The other end of the workpiece is free to deflect. Other faces are free to deform. This configuration corresponds to welding toward the fixed face. Figures 16\u201318 highlight the x-, y-, and z-directional deformation of the workpiece. Welding parameters are: welding current 150 A, arc length 3 mm, welding speed 2.5 mm/s, and 100 ppm of oxygen. The total deformation of the workpiece is shown in Fig. 19. These results show that the free end deflection can be minimized when welding toward the fixed end. Various other constraint configurations and their effects on the developed stresses are simulated but results are not highlighted here. It is to be noted that the developed stresses obtained from the FE solver are influenced by the nature of the constraints applied. Additionally, we note that the welding bead is symmetrical about the welding axis, and normal stress components prevail in the weld" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001550_chicc.2015.7259633-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001550_chicc.2015.7259633-Figure2-1.png", + "caption": "Fig. 2: Thrust state of heavy movement", + "texts": [ + " It includes the translational motion of buoyant center and pitch of circling the oy axis. Submarine don\u2019t produce yawing force yF , yawing moment zM and roll moment xM under vertical motion condition, so all the parameter of lateral movement and roll. Now we can interfere the horizontal movement of submarine, but it nearly couldn\u2019t affect the vertical movement. So the interference effectiveness of the ocean current is neglected when we study the vertical motion. The vertical motion of submarine with VVP includes heave motion and pitching motion, heave motion is shown as in Fig. 2. When the two VVP is in sine cycling pitch state and rotary direction is reverse, or the phase contrast of cycling pitch angles of the two VVP is 180 , that is to say two rotary incline plane of VVP transverse the same angle circling Z axis, VVP produce upward or downward thrust, so the submarine can move alone up-down direction. Because the characteristic of VVP is the pitch angle of blade is variable, the thrust magnitude of VVP is affected by the actual speed (i.e. u ), relative water velocity (i" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002641_s12206-013-0504-1-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002641_s12206-013-0504-1-Figure1-1.png", + "caption": "Fig. 1. Schematic diagram of the seal test rig.", + "texts": [ + " When the vibration of the cylinder appears under the action of air force, the cylinder displacements are measured under different inlet/outlet pressure ratios and rotor speeds, and the equivalent seal force can be identified. To enlarge the influence of the seal force, the rotordynamic coefficient test is conducted near the resonance region of the cylinder system. The unbalanced mass distribution on the rotor is used to provide synchronous frequency excitation force to the seals. Four independent equations are formulated under the two sets of rotor unbalance conditions. Fig. 1 depicts our rotordynamic test rig. A slender steel shaft is supported by two steel oil journal bearings lubricated by ISO VG32 turbine oil. The middle of the shaft holds a steel sleeve coupled with labyrinth seals with a diameter of 180 mm. Two steel balance disks near the bearings are used to regulate the original vibration of the rotor and provide unbalanced exciting forces. The magnitudes and angles of unbalanced excitation forces can be changed by changing the location of unbalance blocks, which are fixed on the balance disks" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002460_ichqp.2014.6842802-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002460_ichqp.2014.6842802-Figure2-1.png", + "caption": "Figure 2. PMSG axial section.", + "texts": [ + " Considering that the magnets will have the same area and the same thickness, generators with round-shaped magnets will have a bigger diameter in comparison with the generators that will use trapezoidal magnets in the rotor construction. In figure 1 is presented the cross section of the model. The three phases of the stator are disposed as above, with three coils connected in series on each phase. Coreless stator implies that the magnetic flux on each phase is zero when the magnets are in the center of the coils, phase C in figure 1. The flux path is represented with green color in figure 2. II. 2D MAGNETIC FLUX SIMULATION The model of the generator has circular coils. Because of this, the analytical determination of the magnetic flux through the stator can be quite complicated. For this reason the magnetic flux was determined from the 2D simulation of the generator. The simulation was performed in Quick Field program [6]. For this analysis we have considered three pairs of poles of the generator and the distances: the distance between two magnets of a pair of poles is 15 mm and the distance between two poles situated on the same disc is equal with 17" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002526_cjme.2014.0717.119-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002526_cjme.2014.0717.119-Figure3-1.png", + "caption": "Fig. 3. Separate FE Models", + "texts": [ + " (4) The heat generated between the sliding surfaces is replaced by heat fluxes, which are directed to the surfaces of the sliding surface elements. (5) The heat convection coefficient is a constant parameter and is not related to the fluctuations of the temperature distribution and the axial translation feed rate of the mechanism. The FE model for thermal behavior analysis was developed into two parts: the guide part model(G-Model) and the slide part model(S-Model), both of which were established in ANSYS V12 workbench as shown in Fig. 3. The heat source in the model consists of three parts: mechanical, electrical and convection heat. The mechanical heat source is mainly from frictional energy loss between the guide way and wagons during the motion. Part of the heat flows to the guide way, in which the flux is qg, and the remaining part flows to wagon, in which the flux is qbl. The electrical heat source, which is simplified as a thermal action on the contact surface with heat flux qm to the sliding table, originated mainly from the primary coil of the linear motor" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003298_s1052618813040158-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003298_s1052618813040158-Figure2-1.png", + "caption": "Fig. 2. Projections of forces and displacements on the disk plane.", + "texts": [ + " The calculation scheme of the construction (a view in the projection on the plane passing through the bearing axis and the disk mass center) is shown in Fig. 1: epr = ecos(\u03d5 \u2013 \u03c8\u00b7 DOI: 10.3103/S1052618813040158 282 JOURNAL OF MACHINERY MANUFACTURE AND RELIABILITY Vol. 42 No. 4 2013 VOLOKHOVSKAYA, BARMINA \u03c80), \u03b4pr = \u03b4cos\u03c80 + \u03b4st, mgpr = mgcos(\u03c8 + \u03c80) are the eccentricity, total shaft flexure, and rotor weight pro jections on the specified plane given a fixed rotor; \u03b4st is the static deflection of the shaft in this plane; the meaning of the angles \u03d5 and \u03c80 is evident from Fig. 2. Figure 2 conditionally shows the scheme of the rotating disk at some arbitrary time (a side projection). In this case, it is assumed that the bent shaft is in its initial state, i.e., not deformed by the forces P1 and P2 of the bearing reaction. With rigid bearings, the point O (Fig. 1) is the projection on the disk plane of the fixed axis of the bearings; the point O1 is the projection on the same plane connecting the ends of the non deformed bent shaft; the point O2 is the geometrical center of the disk (the attachment point of the disk on the shaft); O3 is the mass center of the disk having the vector\u2014eccentricity e; the vector \u03b4 = cor responds to the projection of the cantilever of the bent shaft on the disk plane and is equal to bending vec tor of the nondeformed shaft", + " The force Pax causes only the cantilever bending, and the corresponding displacement uax along the axis O2O1 is expressed by (2) O1O2 Pax P1 \u03c8cos P2 \u03c8, Ppasin+ \u2013P1 \u03c8sin P2 \u03c8.cos+= = uax 1 2 Pax\u03b4bend, \u03b4bend\u2013 l 3 / 3EI( ),= = JOURNAL OF MACHINERY MANUFACTURE AND RELIABILITY Vol. 42 No. 4 2013 FEATURES OF OSCILLATIONS OF AN INITIALLY DEFLECTED ROTOR 283 where \u03b4bend is the bending compliance of the canti lever beam of length l; EI is the stiffness of the shaft section at flexure. The force Ppa/2 causes both bending and torsion of the cantilever in the direc tion perpendicular to O2O1 (Fig. 2): (3) Let us determine the displacements under tor sion utors, which are induced by Ppa/2. The displace ment differential dutors in the direction of Ppa under upa ubend utors+ 1 2 Ppa\u2013 \u03b4bend \u03b4tors+( ).= = rotation of the current section z due to the torque Mtors = \u23afPpa y/2 on the length dz and the displacement utors will be the following (Fig. 3): (4) During integration, it was taken here that y = \u03b4cos(\u03c0z/2l) and GItors is the stiffness of the shaft section under torsion. Let us determine the projections of displacements u1 and u2 at axis Ox1 and Ox2: By adding the bearing compliances in the direction of axes Ox1(1/k1) and Ox2(1/k2) with use of (1), (2), and (3), we find the displacements in the directions of these axes for the rotor installed on the anisotropic compliant bearings: (5) (6) Note that the correlation \u03b4tors/\u03b4bend ~ (\u03b4/l)2 1 is of a sufficiently small value, so the values \u03b4tors/\u03b41 1, \u03b4tors/\u03b42 1 are small according (6)", + "+= = P1/m u1\u03a91 2 4\u03bc\u03c71 u1\u03a91 2 \u03c8sin 2 u2\u03a92 2 \u03c8 \u03c8cossin\u2013[ ],\u2013= P2/m u2\u03a92 2 4\u03bc\u03c72 u2\u03a92 2 \u03c8cos 2 u1\u03a91 2 \u03c8 \u03c8cossin\u2013[ ];\u2013= \u03a91 2 = 1/ m\u03b41( ), \u03a92 2 = 1/ m\u03b42( ), \u03c71 = \u03b4bend/ 2\u03b41( ), \u03c72 = \u03b4bend/ 2\u03b42( ), \u03bc = \u03b4tors/ 2\u03b4bend( ). mu\u00b7\u00b710 \u20132h1mu\u00b7 1 P1, mu\u00b7\u00b720\u2013 \u20132h2mu\u00b7 2 P2\u2013 mg\u2013= = 284 JOURNAL OF MACHINERY MANUFACTURE AND RELIABILITY Vol. 42 No. 4 2013 VOLOKHOVSKAYA, BARMINA and the equation of change of the kinetic moment of the system with respect to the fixed axis Oz perpen dicular to the disk plane (9) where Mext is the moment of external forces with respect to the axis Oz; Mtor and Mbr are torque and the braking moments; e' = |e + \u03b4| (Fig. 2); IO3 is the inertia moment of the disk with respect to O3z. Substituting the expression for reactions P1 and P2 (7) into (8) and (9) and calculating the time deriv ative from the kinetic moment KO, taking into account the correlations (Fig. 2) , and making the labor consuming transformations, we find the system of three equations of motion for the physical model in the following form: (10) In (10), besides the designations used above, the following were additionally adopted: Let the angle \u03d5 between the disk eccentricity vector e and the bending arrow of the bent shaft \u03b4 be known (Fig. 2). Then, the trigonometrical functions of the angle \u03c80 and the value e' are determined by the following formulas: The initial conditions of the problem depend on the conditions of the turbounit startup. To obtain the solution of (10) corresponding to the startup of the cold rotor with a formed initial deflection, we can write the following: (11) In order to solve system (10) at startup, with a thermally unstable rotor, we need to know the relation ship between the modulus of the bending arrow \u03b4 and the time [0, twork] corresponding to the time required K\u00b7 O Mext,\u2211= KO IO3\u03c8\u00b7 m u10u\u00b7 20 u20u\u00b7 10\u2013( ), K\u00b7\u00b7 O+ IO3\u03c8\u00b7\u00b7 m u10u\u00b7\u00b720 u20u\u00b7\u00b710\u2013( ),+= = Mext 2h1mu\u00b7 1 P1+( )u2 2h2mu\u00b7 2 P2+( )u1\u2013 mg u1 e' \u03c8 \u03c80\u2013( )cos+[ ]\u2013 Mtor Mbr,\u2013+= u10 u1 e' \u03c8 \u03c80+( ), u20cos+ u2 e' \u03c8 \u03c80+( ),sin+= = u\u00b7\u00b710 u\u00b7 1 e' \u03c8\u00b7\u00b7 \u03c8 \u03c80+( )sin \u03c8\u00b7 2 \u03c8 \u03c80+( )cos+[ ],+= u\u00b7\u00b720 u\u00b7 1 e' \u03c8\u00b7\u00b7 \u03c8 \u03c80+( )cos \u03c8\u00b7 2 \u03c8 \u03c80+( )sin+[ ]+= u\u00b7\u00b71 2h1u\u00b7 1 \u03a91 2 1 2\u03bc\u03c71\u2013( )u1 2\u03bc\u03c71 \u03a91 2 u1 2\u03c8cos \u03a92 2 u2 2\u03c8sin+[ ]+ + + = e' \u03c8\u00b7\u00b7 \u03c8 \u03c80+( )sin \u03c8\u00b7 2 \u03c8 \u03c80+( )cos+[ ], u\u00b7\u00b72 2h2u\u00b7 2 \u03a92 2 1 2\u03bc\u03c72\u2013( )u2 2\u03bc\u03c72 \u03a91 2 u1 2\u03c8sin \u03a92 2 u2 2\u03c8cos+[ ]+ + + = e' \u03c8\u00b7\u00b7 \u03c8 \u03c80+( )cos \u03c8\u00b7 2 \u03c8 \u03c80+( )sin+[ ] g,\u2013 IO3\u03c8\u00b7\u00b7 2h1cu\u00b7 1 2h2su\u00b7 2\u2013 \u03a91c 2 1 \u03bc 3\u03c71 \u03c72+( )\u2013[ ]u1 \u03a92s 2 1 \u03bc \u03c72 \u03c71\u2013( )\u2013[ ]u2+ +{ } \u03c8sin+ + 2h1su\u00b7 1 2h2cu\u00b7 2\u2013 \u03a91s 2 1 \u03bc \u03c72 \u03c71\u2013( )+[ ]u1 \u03a92c 2 1 \u03c71 3\u03c72+( )\u2013[ ]u2\u2013+{ } \u03c8cos \u2013 \u03bc \u03c72 \u03c71\u2013( ) \u03a91c 2 u1 \u03a92s 2 u2+[ ] 3\u03c8sin \u03bc \u03c72 \u03c71\u2013( ) \u03a92c 2 u2 \u03a91s 2 u1\u2013[ ] 3\u03c8cos+ Mtor Mbr", + " ANALYSIS OF EQUATIONS OF MOTION As a result of the analysis of system (10), the following considerations can be expressed. If the torsion of the bent shaft is not taken into account, i.e., \u03bc = \u03b4tors/(2\u03b4bend) = 0, which is available since \u03bc 1, then, the deflected rotor will behave as the unbalanced one, which eccentricity e' is made of the vector of proper imbalance e with respect to the point O2 where the disk is fixed on the shaft and the vector of initial deflection of the shaft \u03b4 in this point e' = |e + \u03b4| (Fig. 2). Thus, the preliminary bending of the shaft depending on the modulus values and the mutual location of e and \u03b4 can lead to an increase and decrease in the eccentricity e' (Fig. 4), as well as in the acting centrifugal inertial forces and levels of rotor vibrations caused by these forces. Upon taking the torsion deformation at the deflected rotor oscillations \u03bc = \u03b4tors/(2\u03b4bend) \u2260 0 into account, the system of the equations of motion becomes parametric, and the parameter is the rotation angle of the rotor \u03c8 (Fig. 2). The given parametric form causes oscillations of twice the frequency of the working frequency. It was found that the system of equations of motion turns out to be similar to that of the rotor on the shaft having different principal moments of inertia of cross section [1]. The coefficients of parametric terms are proportional to the square of relationship between the initial deflection arrow and a half shaft span length \u03b42/l2. They are relatively small values. However, the parametric properties of the system can play an important role in resonance velocities (having double the value of the working one, namely, 100 Hz)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003340_cp.2014.0485-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003340_cp.2014.0485-Figure2-1.png", + "caption": "Figure 2. A schematic of the configuration of PC-PM motor.", + "texts": [ + " 1 shows the pole changing of motors that are suitable for variable speed operation. When the motor operates with six poles, it produces a high torque at low speeds. If the motor changes to two poles, it produces a high power at high speeds. In this paper, we explain the principle and basic characteristics of the motor by using a finite element method magnetic-field analysis, which has a PM magnetized by a pulse d-axis current of the armature winding. The results of our efforts show that the proposed motor reduces core loss by approximately 50% and triples the speed range. Fig. 2 shows the basic configuration of a pole-changing PM (PC-PM) motor that changes the number of magnetic poles by a factor of three. The rotor of the PC-PM motor has a salient core, and the PMs are embedded in an iron core. There are 2 two types of PMs having different magnetic properties. The four variable magnetized magnets are arranged adjacent to the constant magnetized magnet. Four PMs with low coercive force (variable magnetized magnet) and two PMs with high coercive force (constant magnetized magnet) can form six poles or two poles", + " When the daxis pulse current of the six-pole armature winding flows, only the variably magnetized magnets in the rotor are magnetized by the magnetic field of the pulse current. The two variably magnetized magnets are magnetized in the different polarity from the constant magnetized magnets and from a six-pole-field by PMs, as shown in Fig. 3(a). Finite element method (FEM) magnetic-field analysis was performed using FEM software (JMAG) to ascertain the basic pole-changing characteristics and verify the feasibility of the proposed PC-PM motor. The analytical PC-PM motor model is shown in Fig. 2. Table 1 shows the motor specifications of the analytical model, and Fig. 4 shows the winding connections for the pole changing of the stator. The variable magnetized magnet that we have previously experimentally produced is a samarium cobalt (SmCo) magnet [7]. Therefore, it seems that the cost of the variable magnetized magnet may be same or less than that of a conventional SmCo magnet because the variable magnet has the potential to reduce the volume of the cobalt component. We verified pole changing of the rotor using the FEM magnetic-field analysis discussed above" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001914_s11740-014-0582-7-Figure13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001914_s11740-014-0582-7-Figure13-1.png", + "caption": "Fig. 13 Result of first milling test", + "texts": [ + " Due to the feed-force saturation, the current depending force constant of the x-axis is stored in a lookup table. After the design of the controller with integrated compensation, the system was tested in an xz-axis milling process. A high-frequency spindle is installed in the hybrid kinematic machine tool to enable appropriate cutting speeds even with a 3 mm tool (see Fig. 5). The machine tool generates motion in the y-axis only. The spindle speed was adjusted to 18.000 min-1 and the feed per tooth was 0.03 mm. The depth of cut was set to 0.2 mm (Fig. 13). The tool path diameter of the milled circle was 10 mm. The result of this first milling experiment without cooling lubricant is shown in Fig. 13. Furthermore, the geometric deviation was measured with a coordinate measuring machine (Leitz PMM 866). The results are plotted in Fig. 14. To evaluate the influence of the hybrid kinematic on the process, also the measured motor position deviation is plotted. Especially at the turning points (see Fig. 14 points 1\u20134) of the single axes a significant geometric deviation can be seen. The measured position deviation at these points is between 14 and 17 lm. Because of the difference between the geometric and the measured deviation, it can be assumed that the stiffness of the machine tool and the end milling cutter influences the result of the milling process" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003587_icmsao.2011.5775511-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003587_icmsao.2011.5775511-Figure1-1.png", + "caption": "Fig. 1. Main structure of the Qua", + "texts": [], + "surrounding_texts": [ + "I. INTRODUCTION\nThe term Unmanned Air Vehicle (UAV) is recognised for its applications in the commer markets. The quad-rotor in progress is an vehicle that has six degrees of freedom concept is simple; the relative speeds of the movement of the vehicle producing angles pit Quad-rotors are evaluated in terms of their le well maneuvering abilities.\nII. QUADROTOR DYNAMIC\nalysis is applied the quad-rotor. D (Proportional the vehicle as it nal motions. A dvantages and e evaluation of ospheric noise,\ne of additional limitations of cussed means of\nD Controller,\nnow universally cial and military unmanned aerial [1]. The general four rotors allow ch, roll and yaw. vel of control as\nS\ndrotor.\nThe center of gravity is assumed structure. Conservation of angular vehicle to rotate to compensate for rates in the motors. The body axi frame through position vector ZYX in 3x3 rotational matrixes:\nwhere, , cos and The kinematic equations are: 11 1 where, , , The dynamic equations are:\nwhere input variables, The total thrust: The rolling moment: The pitching moment: The yawing moment: and, , , ,\nIII. CONTROL STR\nBy definition, a controller is a d affects the operational conditions system. The operational conditions\nto be at the center of the momentum will cause the the differences in rotation\ns is related to the inertial and Euler angles resulting\n(1)\n,\n(2)\n(3)\n(4)\nATEGIES evice which monitors and of a given dynamical are typically referred to as\n978-1-4577-0005-7/11/$26.00 \u00a92011 IEEE", + "output variables of the system which can be affected by adjusting certain input variables [2]. Here, we describe the conditions for the three main modes which are taking off, landing and flying.\nDuring hover mode, the four rotor flying vehicle is stable where the translational positions of (x,y,z) are constant and the pitch, roll and yaw are at zero value. However, when changes occur to the constant states of the hover mode, the quad-rotor is switched to another mode. When there is a deviation from the center in terms of left or right, the y-value is adjusted to maneuver it back to central position. The x-value is changed when there is a movement forwards or backwards. As for the vehicle climbing or descending, the z-value is checked. Lastly for changes in yaw, the angle is manipulated to allow it to return to initial center.\nThese changes that return the quad-rotor to its desired position create stability needed for successful flight to the desired final destination. As slight errors occur, the accumulation of errors can be detrimental to the system if not controlled effectively. Thus, if all solutions of the dynamical system that start out near equilibrium point and stays near it forever, then it is stable [3].\nIV. PROPORTIONAL DERIVATIVE CONTROLLER\nThe PD controller is chosen to promote predictive behaviour due to its simplicity in application, efficient tuning and broad range of coefficients that comply with necessary specifications of the system. PID stands for the three term controller where a term proportional to the error, another integral to an error and lastly the derivative of the error [4]. The control system is broken down into loops, each dependent on the loop before. The exterior loop, houses the translational controller that allows the inner loop to function by supplying the desired pitch and roll values that are dependent on the desired x and y positions. The interior loop, where the rotational controller produces the values necessary to generate the total thrust, rolling, pitching and yawing moments: 1 / 2 3 4\nwhere the proportional and derivative constants are calculated by converting equations to a second order system and solving for both rotational and translational motions. These time varying moments are then used as inputs for approximation of the Quadrotor\u2019s system dynamics. Lastly, using integrators, the values of measured acceleration, velocity and position are used as feedback for comparison of differences with desired references [5]. Thus, by specifying desired target location in the three dimensions, the Quadrotor is steered by the PD controller to minimize error producing movement from its initial coordinate to the desired space by getting closer with each sample period.\nIn this paper, the basic PD controller is simulated and analysis is performed for improvements based on overshoot, settling time, magnitude of oscillations, minimum and maximum values as well as errors. In the next section, using the optimised control system, the nonlinearities that affect an UAV is introduced to test the robustness of the system. These induced disturbances and parametric differences are extremely vital as to produce a controller capable of real life flights within aggressive conditions." + ] + }, + { + "image_filename": "designv11_84_0001210_iciafs.2014.7069541-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001210_iciafs.2014.7069541-Figure2-1.png", + "caption": "Fig. 2. Cross-section diagram of the CSM.", + "texts": [ + " This paper presents the utility of force transmission in the 2-DOF system by the proposed disturbance model. Bilateral control test is carried out to validate the utility. This paper is organized as follows. In section II, the structure of CSM is presented. In section III, bilateral control method is described. In addition, disturbance modeling procedure and that results are also described. In section IV, experimental set up and results are given and section V is conclusion. 978-1-4799-4598-6/14/$31.00 c\u20dd2014 IEEE Fig. 2 shows structure inside of the CSM. Fig. 3 shows overall structure of the CSM. Table I shows the specifications of stator and mover [4]. The mover part consists of a mover case and a mover bobbin with coils, and the stator part consists of bent pipe and magnets. Both edges of stator are fixed by two retainers. The stage moves along the guide rail that is bent as same as the pipe. The mover part is fixed on the stage. The optical linear encoder attached on the side of the stage reads position information" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002868_s12206-015-0143-9-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002868_s12206-015-0143-9-Figure1-1.png", + "caption": "Fig. 1. Design concept from a car.", + "texts": [], + "surrounding_texts": [ + "Since there is no cyclic control in the currently designed PAV, the term \u2018tilt-rotor\u2019 is inappropriate for use, and due to absence of wings, it is also inappropriate to use \u2018tilt-wing\u2019. The uses of four short propellers with high degree of twist lead to the name of quad tilt-prop PAV. If the wings and aerodynamic control surfaces are available, it is then possible to transit to the airplane mode; i.e., tilt angle \u03b3 = 90\u00b0 [11]. However, Fig. 4 shows the flight data that the inboard wing caused a serious uncontrollable state during the VTOL mode flight. It was barely controllable, but the ground effect under the wing caused an uncontrollable state near the ground. The inboard wing had to be modified to a circular support in the initial prototype due to its strong aerodynamic interference that is also prevalent in tilt-rotor [11]. Therefore, unlike a tilt-wing concept, the current prototype has 3 modes: i.e., the \u2018driving\u2019, \u2018VTOL\u2019, and \u2018forward flight\u2019 modes as il- lustrated in Fig. 3. During the forward flight mode, the nacelle tilt angle is limited under \u00b120\u00b0, where the dynamics is similar to a pure quad-rotor, and no additional aerodynamic control surfaces are needed. Fig. 5 shows the frame structure and applied mechanisms. It is mainly constructed with two longitudinal carbon pipes and four aluminum bulkhead frames for a light structure. Two torque tubes are installed on the longitudinal pipes by aluminum housings, where the front torque tube is installed with two nacelles. Each front nacelle is equipped with a propulsion motor, two wheels, a steering mechanism and a steering servo actuator. The rear torque tube is also installed with two nacelles, where each nacelle is equipped with a propulsion motor, two wheels, a driving motor with a reduction gear and a brake drum. The mechanism of tilt actuator is also included in the Fig. 5, where there are total of 8 high-torque servo actuator coupled in a pair that control tilt angle. A stopper pin is also added to release the power of servo actuators during the driving mode. As it was also shown in the Fig. 3 for the orientation of tiltable nacelles during each mode, during the driving mode, nacelles are fixed to 45\u00b0 facing each other to lower the overall height, and locked by the stopper pin to save power consumption. When the VTOL mode is initiated, two pairs of servo actuators deliver control forces to each torque tube without any gear to simplify the overall structure [8]. So far, controls are achieved by manual control during the driving mode, and the manual operator controls the steering angle of front wheels, driving motor and brake drums of rear wheels. During the VTOL mode, nacelles are fixed straight-up, and a pure quad rotor controls are applied. In the forward flight mode, forward velocity is controlled by acceleration created from tilted nacelles, where all nacelles are tilted with the same angle \u03b3. When these two modes are compared, the forward flight mode will minimize the discomfort of flight experience than the VTOL mode that has to tilt its whole body pitch angle to control forward velocity. Small body pitch angle may improve the discomfort during the acceleration and deceleration, however, if the passenger is not sitting at the exact position as the center of gravity, any pitching moment will result to a positive or negative gravity offset, and the passenger will have the feeling that he/she is constantly in a turbulent wind. Therefore, the VTOL mode is only used during initial take-off and at the end of landing, and the tilt angle \u03b3 will constantly be changing from the forward velocity feedback loop whenever it is off the ground. Fig. 6 shows onboard avionics in the PAV. A custom-made flight control computer was built to realize autonomous control using a 150 MHz 32-bit floating-point TMS320F28335 processor from Texas Instrument, Inc. A commercial GPSINS from Microbotics, Inc. was adopted to provide attitude, velocity and position data at 50 Hz rate. A commercial Bluetooth modem was used for wireless data communication that has maximum operating range of 300 ~ 1000 m depending on the line-of-sight condition. It is also equipped with a custommade data recorder that records all flight data in a micro-SD card." + ] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.6-1.png", + "caption": "Fig. 7.6 The two working modes of the leg i of the 3\u2013RPR PPM. a Working mode 1. b Working mode 2", + "texts": [ + "19) where xO Ai3 = x \u2212 di4 cos(\u03c6 + \u03b5i ), yO Ai3 = y \u2212 di4 sin(\u03c6 + \u03b5i ) are the coordinates of points O Ai3 expressed in the base frameF0. It can be finally deduced from (7.19) that, for i = 1, . . . , 3: qi2 = \u00b1 \u221a ( xO Ai3 \u2212 xO Ai1 )2 + ( yO Ai3 \u2212 yO Ai1 )2 . (7.20) Finally, the passive variables can be found by: qi1 = atan2 ( yO Ai3 \u2212 yO Ai1 , xO Ai3 \u2212 xO Ai1 ) , if qi2 \u2265 0 qi1 = atan2 ( yO Ai3 \u2212 yO Ai1 , xO Ai3 \u2212 xO Ai1 ) + \u03c0, if qi2 < 0 (7.21) and qi3 = \u03c6 + \u03b5i \u2212 qi1. (7.22) In (7.20), the sign \u201c\u00b1\u201d correspond to the different working modes of the robot (which are, in that particular case, equivalent\u2014Fig. 7.6). From a geometric point of view, solving these equations is equivalent to finding the distance between the points Ai1 and Ai3. The geometrical and analytical methodologies presented in Sects. 7.1.2.1 and 7.1.2.2 can be easily extended to any types of planar parallel robots andwill not be developed here. The reader is referred to Bonev (2002) for further investigations. TheOrthoglide is a parallel robot composed of three identical legs (Fig. 7.7) allowing three translational DOF of its end-effector (parameterized by the variables x , y and z that represent respectively the translation along x0, y0 and z0 of the base frameF0)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003042_amm.740.69-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003042_amm.740.69-Figure1-1.png", + "caption": "Fig. 1. Illustration of meshing line of spur gear transmission Fig. 2. Involue tooth profile with logarithmic modification", + "texts": [ + " In this paper, a logarithmical lead profile was applied on spur gears and the surface coordinate equation of logarithmic crowned tooth for manufacture was established. The Matlab programs were also developed to calculate the distribution of contact stress and von Mises stress field inside subsurface layer at every meshing position to reveal the influences the logarithmic modification has on the fatigue resistance of teeth surfaces during the gear transmission process. The mesh of spur gears teeth can be modeled as the Non-Hertzian Contact of two finite length circular cylinders with respective radii R2, R1 shown in Fig. 1, and R2, R1 equal to the radii of the gear and pinion teeth profiles curvatures at the contact point respectively. The composite curvature radius R\u2211 at a certain engaging position k can be written as : 2 1 1 1 = + tan + tan' ' b1 b b bR r a P K r a P K\u2211 \u2212 . (1) if 1 2 2 1 2 1 ( 1,1 ), single meshing state 2 2 [ , 1) [1 , ], double meshing state 2 2 2 2 Z Z K A B Z Z Z Z K B A B A \u03c0 \u03c0 \u03c0 \u03c0 \u03c0 \u03c0 \u2208 \u2212 \u2212 \u2208 \u2212 \u2212 \u222a \u2212 . (2) Where k b z K P = , ' 1tan tanaA= \u03b1 \u03b1\u2212 , ' 2tan tanaB = \u03b1 \u03b1\u2212 . (3) In these equations, kz is the z-coordinate of a certain contact point k on the actual line of contact, bP is the base pitch, 1br and 2br are the radius of base circle, 1 Z and 2 Z refer to the number of teeth, 1a \u03b1 and 2a \u03b1 stand for the pressure angle at outside radius, '\u03b1 is pressure angle at radius of pitch circle, the subscripts 1 and 2 represent the pinion and gear respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001600_ecc.2013.6669841-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001600_ecc.2013.6669841-Figure4-1.png", + "caption": "FIGURE 4. Schematic drawing of the VSA system integrated with a reaction wheel.", + "texts": [ + " On the other hand, the benefits of the fast dynamics of the reaction wheel can be exploited with small sampling times. Therefore, we anticipate that the usage of the NMPC for VSA robots with reaction wheels will be limited to low-DOF systems for the foreseeable future. IV. CASE STUDY A. MODEL OF THE CASE STUDY SYSTEM In order to show the merits of the reaction wheel augmentation of VSA robots, we conducted a case study using a single-DOF VSA robot with a reaction wheel attached to it. The schematics of this system is shown in Fig. 4. The task is to throw a ball attached to the distal end of the robot to the farthest distance. The link is actuated by two servo-motors via two NEEs and the disk is actuated by a brushless DC motor. The dynamics of the systems can be obtained using the modeling formalism presented in subsection III-A. The inertia matrix M\u0303 for this system is M\u0303 = [ M11 Jw1 Jw1 Jw1 ] (15) withM11 = ml1(L l c,1) 2 +mw1 (L w c,1) 2 +mbL2c,b+ J l 1+ J w 1 + Jb, where ml1, J l 1,L l c,1 are the mass, the inertia about center of mass, and the distance from the pivot to center of mass of the link" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003592_amr.199-200.1410-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003592_amr.199-200.1410-Figure1-1.png", + "caption": "Fig. 1 Finite element model of wind turbine blade bearing", + "texts": [ + "arge-size wind turbine blade bearings are used as connections in wind power generators. The structure of the bearing can be divided into two main parts; the rolling elements (balls or rollers) and the rings, which are fixed to the external structures such as hub and blade by means of bolted connections (Fig. 1). The supporting structures in this kind of large-size bearings are highly flexible. It is impossible to achieve a uniform distribution of load among the particular rolling elements in the slewing-mechanism bearing[1]. In order to choose the proper bearing and shape the load-carrying structure one must estimate the distribution of forces among the individual rolling elements. One way of analyzing the effect of the stiffness is using finite element models (FEMs) taking the geometry of the elements that are working during the loading into consideration", + " Smolnicki [3] also supported the importance of using the elasticity of the model during FEM analysis and for this purpose, he uses a superelement based element. Daidi\u00e9 et al. [4] used a similar simplification for the rolling elements, nonlinear traction springs between the centers of curvatures. This article describes a way of obtaining the force distribution in a four contact-points slewing bearing with two rows of ball bearings based on FEM analysis. The main feature is the introduction of the stiffness effect of the whole structure into the calculation. The model analyzed is illustrated in Fig. 1 and connects hub and blade using a slewing bearing. A connection body was fixed to the inner ring in order to avoid the influence of applied external load on the stress in bearing. Table 1 describes the parameters of the bearing. All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP, www.ttp.net. (ID: 132.174.255.3, University of Southern California, Los Angeles, United States of America-16/08/13,22:28:57) For the FEM-based large-size wind turbine blade bearing model, it is impossible to take into account the rolling element-race contact and the phenomena which occur in it because multiple contact pairs must be defined", + " The amount of all elements was 19398 with 36395 nodes. The rings were meshed with Solid45 hexahedral elements. The amount of all elements was 32508 with 40284 nodes. The connection body was also meshed with Solid45. Material properties were set for the model made of bearing steel described with Young modulus, E=210 000 MPa, and Poisson\u2019s ratio, \u03bd=0.3. The force(Fx,Fy,Fz) and torque(Mx,My) was applied on the center of connect body front surface and the main shaft hole was fully constrained(UX=0,UY=0,UZ=0), as showed in Fig.1 Fig.4 compares the load distribution of the rollers along the bear\u2019s circumference in combination with three different support structures and calculation producers. The blue line, for the rigidly supported bearing and the calculation by Hertz equation, shows the most uniform distribution of load along the up and down half contact region, but contact region is relatively narrow. The green line, for the rigidly supported bearing and the calculation by FEM, shows load distributed uniformly along the up and down half contact region, but the amount of rollers contacted is increased obviously" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001585_chicc.2015.7260190-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001585_chicc.2015.7260190-Figure1-1.png", + "caption": "Fig. 1: Configuration of dual controlled hypersonic missile", + "texts": [ + " In Section 3 the blended controller is designed via three portions, which are overload command allocation by fuzzy control approach, aerodynamic subsystem design using adaptive sliding mode controller (ASMC), and direct force subsystem design via two-dimension fuzzy controller. Simulation results are provided in Section 4. The research object of this paper is a hypersonic missile in wave-rider configuration. In view of the angle of attack constraint (usually 6 , 6 ) during the normal operation of scramjet, a dual control via aerodynamic and divert-control lateral jets is adapted to the missile in the cruise phase. As shown in Fig. 1, a number of micro impulse jets are installed near the center of mass. Assume all the jets are repeatable. Longitudinal short-period motion equations are established near a feature point in the cruise phase. Suppose that the mass and velocity are taken as constants, and the additional moment produced by elevator downwash are ignored. Then the motion equations can be described as follows 22 22 24 24 25 25( ) ( ) ( ) zq a a q a a a a (1) 34 34 35 35 36 36( ) ( ) ( )z Tq a a a a a a (2) where 22 24 25 34 35, , , ,a a a a a and 36a are dynamic coefficients, defined as 22 q z z Ma J , 24 z z Ma J , 25 z z z Ma J , 34 P Ya mV , 35 zYa mV , max 36 T a mV And q is pitch rate; is trajectory inclination angle; is angle of attack; z is elevator deflection" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003924_elektro.2014.6848891-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003924_elektro.2014.6848891-Figure10-1.png", + "caption": "Fig. 10. Cross section of induction machines.", + "texts": [ + " In the view of temperatures distribution the model must be most complex. Previous fact leads to made complex 3D model of induction machines. Boundary conditions have been estimated using analytical solution [3], [4], [5]. The equivalent thermal conductivity of construction elements is used the next relation n I I1=1 apr, (8) Where t5 is length of i material, ,1 is thermal conductivity of i material. V. SIMULATION RESULTS Fig. 9. and 10. shows the distribution of temperatures. Fig. 9. represent axial cross section of induction motor. Fig. 10. represent cross section of 1M. Temperature of stator winding, rotor shorting rings and rotor bars is important of thermal analysis. Highest temperature is placed at rotor bars. Maximal temperatures of rotor reach up to 160 0e. Temperature of stator winding is 135\u00b0e. VI. CONCLUSION The motor temperature depends on the type of the loading cycle and ambient temperature. The self-ventilated system is insufficient for the given loading cycle, because the temperature is over the class B, although under the class F" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002245_kem.577-578.45-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002245_kem.577-578.45-Figure3-1.png", + "caption": "Fig. 3: The asperity point load mechanism with an equivalent gear geometry. A cosine shaped axisymmetric asperity on the flat surface of the pinion is over-rolled by the follower. The aspect ratio of the asperity, \u03bbasp, is given as the height over the radius of the asperity.", + "texts": [ + " Crack growth was observed to occur in the rolling direction with a fairly shallow start angle to the surface, typically less than 30\u25e6, see Fig. 2. Asperity point load mechanism. Olsson [2] proposed the asperity point load mechanism which supplies tensile surface stresses for initiation [3] and propagation [4, 5] of surface initiated RCF damage. Asperities on the contact surface induce three-dimensional contact loads and act as local stress raisers. The mechanism is illustrated for a gear application in Fig. 3. Damage in the investigated gear application in Fig. 1 was modelled using an equivalent geometry as illustrated in Fig. 3. Isotropic linear elastic material behaviour was assumed for the case hardened gear steel, following SS2506. Surface treatments or running-in resulted in a constant residual surface stress in the rolling direction, \u03c3R. Asperity point loadmodel. The load transfer between the pinion and the follower was given by superposition of Hertzian cylindrical and spherical pressures and tractions. The magnitude of the asperity loads, both normal and tangential, was determined through FE simulations [4]. The tangential load transfers were determined with constant coefficients of friction for the two-dimensional cylindrical contact, \u00b5, and the three-dimensional spherical contact, \u00b5asp" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003667_amm.315.889-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003667_amm.315.889-Figure4-1.png", + "caption": "Figure 4. Contours of pressure for, (a) smooth liner bearing, and (b) wavy liner bearing.", + "texts": [ + " The maximum pressure of the smooth liner occurs approximately at the half of the bearing length while the wavy liner records almost three-quarters inside of the bearing length. As the rotation speed increases, both bearings produce insignificant maximum pressure. The generated average pressure of wavy liner bearing is higher than the smooth liner. Therefore, under a same condition, wavy liner bearing can carry more loads than the smooth liner while simultaneously influencing the bearing performance. Fig. 4 shows the contour of pressure of both smooth and rectangular wavy liner bearings. This contour pressure represents all images which were produced throughout the study of 0.108 in eccentricity ratio and 2500 rpm in shaft rotation. The figure clearly shows that the presence of wavy surface liner influences the pressure distribution as noted by the vicinity of high pressure that concentrates at the top of the wave. Also spotted from the figure, both bearings form maximum pressure areas at the almost half of each bearing in the direction of shaft rotation. the smooth liner bearing. On the other hand, the minimum pressure occurs at the top of both bearings where it is defined as the region of maximum fluid thickness. The produced pressure contours from Fig. 4 are in accordance with the prediction of formation of pressure regions that occurs on the bearing as illustrated by Gertzos et.al (2008) [3]. As seen previously the advantages of wavy liner bearing, further understanding on the pressure behavior of the wavy liner bearing is presented. The pressure profile of the bearing circumferentially is constructed. Fig. 5 shows the pressure distribution of the wavy liner bearing plotted circumferentially for various eccentricity ratios with shaft rotation speed of 200 and 5000 rpm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002342_kem.480-481.974-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002342_kem.480-481.974-Figure7-1.png", + "caption": "Fig. 7.", + "texts": [ + "6mm\u0445\u044413mm, outer ring land riding. Fig.4 Temperature distribution of bearing with shaft assembly 5 10 15 20 25 30 35 40 45 AXIAL FORCE(N) T E M P E R A T U R E (\u00b7 C ) Outer raceway temperature Temperature of C-type spacer Inner raceway temperature Shaft middle temperature Fig.6 Effect of axial load upon temperature rise 0 2 4 6 8 10 30 35 40 45 RADIAL FORCE (N) T E M P E R A T U R E ( \u00b7C ) Outer raceway temperature Temperature of C-type spacer Inner raceway temperature Shaft middle temperature Fig.7 Effect of radial load upon temperature rise The temperature of bearing with shaft varied with rotational speed as shown in Fig. 5 (the outer ring is stationary, axial load is 8N, radial load is 1N, the initialization temperature is 28\u2103).The temperatures of different parts of bearing with shaft increased when the rotational speed changes from 10000r/min to 30000r/min, however the temperature of inner raceway surface greatly changes (increasing by 16\u2103).The middle temperature of shaft increases relatively obviously" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001687_978-94-007-1643-8_30-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001687_978-94-007-1643-8_30-Figure1-1.png", + "caption": "Fig. 1 Geometry of the tippe-top", + "texts": [ + "et us consider a heavy solid body (a top) on a rough horizontal plane (Fig. 1). The top consists of two spherical segments with radii r1 and r2 (r1 > r2), the segments are connected by a rod, that goes through their centers O1 and O2 (this model is taken from [5]). Let the unit vector of the dynamical and geometrical axis of symmetry e = \u2212\u2212\u2212\u2192 O1O2/O1O2 make an angle of nutation \u03b8 with the up-going vertical \u03b3. The top touches the plane in point C1 by the first spherical segment if \u03b8 \u2208 [0,\u03c0 \u2212\u03b1), and by the second segment in point C2 if \u03b8 \u2208 (\u03c0 \u2212\u03b1,\u03c0 ] ; the top lies on two points if \u03b8 = \u03c0 \u2212\u03b1 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002937_amc.2014.6823323-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002937_amc.2014.6823323-Figure1-1.png", + "caption": "Fig. 1. The under-developing omnidirectional hybrid walker and wheelchair", + "texts": [ + " However, when an attendant-propelled wheelchair starts to ascend or descend a slope from a horizontal plane, the attendant and the wheelchair are not on the same plane (same slope or same horizontal plane). Consequently their respective traveling velocities are different, and this will result in a sudden change of the attendant\u2019s load. In other words, the load that the attendant exerts to the wheelchair will inevitably have a sudden change. Now we are developing a novel electric powered and attendant-propelled hybrid walker and wheelchair as shown in Fig. 1 ([9][10]). This robot can be used either for an elderly\u2019s walking support (Fig. 2(a)), a disabled\u2019s rehabilitation, or a caregiver\u2019s power assistance with admittance control (Fig. 2(b)). Moreover, as a hybrid powered walker and wheelchair, the robot should completely yield to the motion intention of the user (either an elderly/disabled person or an attendant), Therefore, the omnidirectional maneuverability is essential to this robot. In this study, the omnidirectional mobility of the robot is achieved by two active dual-wheel modules" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002309_icems.2011.6073725-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002309_icems.2011.6073725-Figure4-1.png", + "caption": "Fig. 4. Synthesization of refU .", + "texts": [ + " By (4) and (8), the reference of the stator voltage refU is calculated by \u23a9 \u23a8 \u23a7 +\u0394= +\u0394= sssss sssss TiRu TiRu /)( /)( \u03b2\u03b2\u03b2 \u03b1\u03b1\u03b1 \u03c8 \u03c8 (9) \u23aa\u23a9 \u23aa \u23a8 \u23a7 \u2220= += ref ssref U uuU \u03b3 \u03b2\u03b1 22 )()( (10) where Ts is the switching period, \u03b3 is the phase angle of refU . In this paper, six sectors are defined as following \u23aa \u23aa \u23aa \u23aa \u23a9 \u23aa\u23aa \u23aa \u23aa \u23a8 \u23a7 <<= <<= <<= <<= <<= <<= \u03c0\u03b8\u03c0 \u03c0\u03b8\u03c0 \u03c0\u03b8\u03c0 \u03c0\u03b8\u03c0 \u03c0\u03b8\u03c0 \u03c0\u03b8 23/5:SectorVI 3/53/4:SectorV 3/4:SectorIV 3/2:SectorIII 3/23/:SectorII 3/0:SectorI (11) which is shown in Fig. 2. refU is located in one sector determined by \u03b3 . Based on the theory of space vector, refU can be synthesized by two nearest active voltage space vectors ( kU , 1+kU ) and two zero voltage space vectors ( 0U , 7U ), as illustrated in Fig. 4. And the acting times of SVs are calculated by skkkkref TUTUTU /)( 11 +++= (12) \u23aa \u23aa \u23a9 \u23aa\u23aa \u23a8 \u23a7 \u2212\u2212= = \u2212= + + 10 1 /sin3 /)60sin(3 kks dcrefsk dcrefsk TTTT uUTT uUTT \u03d5 \u03d5o (13) where \u03d5 is the modulus after division of \u03b3 and o60 ; kT , 1+kT , 0T are the acting time of kU , 1+kU , zero voltage space vectors, respectively. The switch states are optimized to be center symmetric and illustrated in Fig. 5. When the inverter applies an active voltage space vector to the IM, one stator phase is connected in series to the dclink rail of the inverter, either to the positive or to the negative polarity, whereas the other two phases are connected in parallel to the opposite polarity" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003068_cacs.2014.7097182-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003068_cacs.2014.7097182-Figure5-1.png", + "caption": "Fig. 5. Cold gas attitude control system: (a) simplified model and (b) freebody diagram.", + "texts": [ + " To validate the design of the thruster, a test platform is built with a strain gauge to measure the force as shown in Fig. 4(a). The force measurement of the thruster is shown in Fig. 4(b). It is shown that as the solenoid valve is opened, the measured thrust of 1.5 N is obtained. By comparing the thrust of the theoretical value and actual thrust from the experimental measurement, the deviation may be due to pipe loss. 978-1-4799-4584-9/14/$31.00 \u00a92014 IEEE The basic features of a cold gas attitude control system is illustrated in Fig. 5. The world frame OXY Z is stationary at all time in O. By simple translation to the body centre, a body-fixed moving frame oxoyozo can be mapped in OXY Z. This moving frame relative to the world frame system is not shown in Fig. 5 for simplicity. We denote the body-fixed frame, oxpypzp, with its origin initially coincident with the origin of oxoyozo and then rotating relative to this body-fixed moving frame. From [12], the rotation matrix between these coordinate frames is given by \u23a1 \u23a3xp yp zp \u23a4 \u23a6 = pRo \u23a1 \u23a3xo yo zo \u23a4 \u23a6 (8) where pRo = \u23a1 \u23a3cos\u03b1 cos\u03b2 sin\u03b1 cos\u03b2 \u2212 sin\u03b2 \u2212 sin\u03b1 cos\u03b1 0 cos\u03b1 sin\u03b2 sin\u03b1 sin\u03b2 cos\u03b2 \u23a4 \u23a6 (9) and \u03b1 is the yaw angle, and \u03b2 is the pitch angle. Assuming that the position vector of the origin of the oxpypzp coordinate frame is r\u0304sys = xI\u0302 + yJ\u0302 + zK\u0302", + " Thus, the rotational kinetic energy of the system is Trot = 1 2 (Ipxx\u03b1\u0307 2 sin2 \u03b2 + Ipyy\u03b2\u0307 2 + Ipzz\u03b1\u0307 2 cos2 \u03b2). (16) We denote \u03c4\u0304 as the torque exerted on the system where \u03c4\u0304 = [\u03c4x \u03c4y \u03c4z] T , and F\u0304s is the actual controlled force generated by the thruster where F\u0304s = [0 F\u03b1 F\u03b2 ] T . This relation is given by \u03c4\u0304 = pR\u22121 o F\u0304sl. (17) Using (9) and (17), it yields \u03c4\u0304 = \u23a1 \u23a3F\u03b1l cos\u03b1 sin\u03b2 \u2212 F\u03b2l sin\u03b1 F\u03b1l sin\u03b1 sin\u03b2 + F\u03b2l cos\u03b1 F\u03b1l cos\u03b2 \u23a4 \u23a6 . (18) To compute the total force in the world frame coordinate, we introduce \u03b8 as illustrated in Fig. 5(b). \u03b8x and \u03b8y are the angles of the body centre with respect to the X and Y axes, respectively. They are given by \u03b8x = sin\u22121( x Lr ), (19) and \u03b8y = sin\u22121( y Lr ). (20) F\u0304r is denoted as a reaction force that tends to pull the body centre to O. From Fig. 5(b), F\u0304r is given by F\u0304r = \u2212 ( mg tan \u03b8xI\u0302 +mg tan \u03b8yJ\u0302 +mgK\u0302 ) . (21) The action force F\u0304t on the system in the world frame coordinate axes can be related to F\u0304s as follows F\u0304t = pR\u22121 o F\u0304s. (22) Thus, the total force F\u0304 exerted on the system in world frame coordinate is given by the summation of F\u0304r and F\u0304t. Using (21) and (22), it yields F\u0304 = \u23a1 \u23a3Fx Fy Fz \u23a4 \u23a6 = \u23a1 \u23a3F\u03b2 cos\u03b1 sin\u03b2 \u2212 F\u03b1 sin\u03b1\u2212mg tan \u03b8x F\u03b1 cos\u03b1+ F\u03b2 sin\u03b1 sin\u03b2 \u2212mg tan \u03b8y F\u03b1 cos\u03b2 \u2212mg \u23a4 \u23a6 . (23) Now, we derive the mathematical model of the cold gas attitude control system by using the Euler-Lagrange method [11] and [12]", + " 7, it follows that the control command is given by u = \u23a7\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23a8 \u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23a9 +1, as { \u03b8 + I 2\u03c4 \u03b8\u0307|\u03b8\u0307| \u2212 \u03b8T < 0 \u03b8 + I 2\u03c4 \u03b8\u0307|\u03b8\u0307| \u2212 \u03b8T = 0, \u2200\u03b8\u0307 < 0 \u22121, as { \u03b8 + I 2\u03c4 \u03b8\u0307|\u03b8\u0307| \u2212 \u03b8T > 0 \u03b8 + I 2\u03c4 \u03b8\u0307|\u03b8\u0307| \u2212 \u03b8T = 0, \u2200\u03b8\u0307 > 0 (42) Note that (42) gives the conditions on the control command so that the system trajectories converge to \u03b8T . So far, we have derived the control for one dimensional case only by assuming that \u03b8 is a general rotation angle. To design the controller for the cold gas dynamic system, by recalling the dynamics model in (29)-(33), we see that if the displacements of body center in x, y and z axes are very small, then (31)-(33) can be neglected. In Fig. 5(a), we observe that the pitch angle is small enough so that we assume \u03b2 \u2248 0. This assumption can be used to (29) and (30). From (29) it implies that sin2 \u03b2 \u2248 0, cos\u03b2 \u2248 1 and cos2 \u03b2 \u2248 1. Furthermore, we neglect [\u03b1\u0307\u03b2\u0307(Ipxx\u2212Ipzz) sin(2\u03b2)] higher-order term from (29), so that (29) can be simplified to \u03b1\u0308Ipzz = F\u03b1l. (43) By using the same assumptions, from (30) we discard [\u03b1\u03072(Ipzz\u2212 Ipxx) sin\u03b2 cos\u03b2] higher-order term in (30), so that it can be simplified to \u03b2\u0308Ipyy = F\u03b2l cos\u03b1. (44) We recall the system given in (34), by based on this system we can write (43) and (44) as Ipzz\u03b1\u0308 = u\u03b1\u03c4\u03b1, (45) Ipyy\u03b2\u0308 = u\u03b2\u03c4\u03b2 cos\u03b1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.14-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.14-1.png", + "caption": "Fig. 2.14. Two-link planar arm", + "texts": [ + " If the end-effector is a gripper, the origin of the end-effector frame is located at the centre of the gripper, the unit vector ae is chosen in the approach direction to the object, the unit vector se is chosen normal to ae in the sliding plane of the jaws, and the unit vector ne is chosen normal to the other two so that the frame (ne, se,ae) is right-handed. A first way to compute direct kinematics is offered by a geometric analysis of the structure of the given manipulator. Example 2.4 Consider the two-link planar arm in Fig. 2.14. On the basis of simple trigonometry, the choice of the joint variables, the base frame, and the end-effector frame leads to8 T b e(q) = \u23a1 \u23a2\u23a3n b e sb e ab e pb e 0 0 0 1 \u23a4 \u23a5\u23a6 = \u23a1 \u23a2\u23a3 0 s12 c12 a1c1 + a2c12 0 \u2212c12 s12 a1s1 + a2s12 1 0 0 0 0 0 0 1 \u23a4 \u23a5\u23a6 . (2.49) It is not difficult to infer that the effectiveness of a geometric approach to the direct kinematics problem is based first on a convenient choice of the relevant quantities and then on the ability and geometric intuition of the problem solver" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002011_2011-01-1691-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002011_2011-01-1691-Figure4-1.png", + "caption": "Figure 4. Selection of focused section", + "texts": [ + " | Volume 4 | Issue 21284 In the conventional application of TPA, contribution of transfer path ranks the measures. However, in the application to automotive body structure, it is difficult to extract the suitable parts for the response reduction, because VT path varies as vibration travels through the structure. In order to break through the difficulty, we propose the contribution analysis in multiple cutting sections between input and output by using FE model. The procedure is explained as follows. In this application, the contribution of the windshield and the front pillars is evaluated as shown in Figure 4 (a). First, contribution of this cross section is calculated, when the proposed techniques are the same as conventional TPA. Second, the section adjacent to the cross section in Figure 4 (a) is selected and is apart from it by one element. The automotive body structure is separated into two substructures at the boundary of this new section, and the contribution is calculated. This operation is repeated and the cut section rises to the roof as shown in Figure 4 (b) and 4 (c), and contributions of all cross sections in VT paths are derived. (a). First cross section (b). Medium cross section Contribution analysis in TPA is not only for few and limited paths, but also for all of DOF in the cross section as a body structure. However, the structure has many nodal points on the cross section, and this requires a fair amount of time to represent and analyze the contribution of each DOF of all the nodal points. This problem is solved by displaying contributions related to three DOFs (xyz or \u03b8x\u03b8y\u03b8z) of each nodal point on structural diagram where this contribution is described as influence degree" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002627_20110828-6-it-1002.03234-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002627_20110828-6-it-1002.03234-Figure9-1.png", + "caption": "Fig. 9. The setting of the numerical examples.", + "texts": [ + " Therefore, even if one of (a, b) is negative, the grasp can be stable when the other is positive large. The curve of e2 = 1 asymptotically converges to the lines a, b = 1. Therefore, the magnitudes of a and b have to be at least greater than 1 in the stable area of (a, b) of 1 < e2. This means that the stiffness effects A and B are necessary to be greater than the gravity effect M in order to realize the stable grasp with any COG position because a = A/M and b = B/M . Some numerical examples are carried out in this section. The setting of the examples is show in Fig. 9. The figure (a) is the overview of the grasp situation, where the hand has three fingers and the grasped object is a cylinder. The figure (b) shows the configurations of the reference frame \u03a3B and spring frame \u03a3Si . The mass of the object is m = 132 [g] and the distance of the COG is rg = 33.5 [mm]. The distance of \u03a3Si from \u03a3B is ri = 37.5 [mm]. In the latter, we check the area of the position of the COG with some parameter of the contact point \u03b1i, the initial displacement of the spring \u03b40i and the stiffness coefficients (kxi , kyi , kzi)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003428_amr.338.94-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003428_amr.338.94-Figure3-1.png", + "caption": "Fig. 3 Melting state and temperature distribution of a particle", + "texts": [ + " The temperature of external surface of the melting coat reaches xT over its melting point mT , while the temperature at the internal interface between melt and its solid core is mT . The temperature of center of un-melted core still keeps 0T , which is the initial temperature of powder bed. According to principle of conservation of energy, laser energy \u2212 E not only induces the melting and temperature rising of a metal particle, but also creates the interfaces of solid-liquid and liquid-gas. The volume of a spherical particle is: 3 3 4 rV \u03c0= (20) According to fig. 3, the volume ( 1sV ) of solid sphere in the core of a particle is: ( )31 3 4 xrVs \u2212= \u03c0 (21) Combining equation (20) and (21), the volume ( 2sV ) of the spherical shell with the thickness of x yields: ( ) ( )2233 12 33 3 4 3 4 3 4 xrxrxxrrVVV ss \u2212\u2212=\u2212\u2212=\u2212= \u03c0\u03c0\u03c0 (22) The area ( xrS \u2212 ) of a core sphere is: ( )24 xrS xr \u2212=\u2212 \u03c0 (23) The area ( rS ) of the interface of liquid-gas is equal to the surface area of the initial sphere: 24 rS r \u03c0= . (24) Equation (25) yields according to principle of conservation of energy: ( ) rsvlvxrsl T T ps T T ps SSdTCVLdTCVE mx m \u03b3\u03b3\u03b3\u03c1\u03c1 \u2212+++ += \u2212 \u2212 \u222b\u222b 0 111222 (25) Where 1\u03c1 , 2\u03c1 are the densities of liquid and solid metal respectively, 1pC , 2pC are the specific heat of liquid and solid metal, sl\u03b3 is the solid-liquid interface tension, lv\u03b3 is liquid-gas interface tension, sv\u03b3 is solid-gas interface tension" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002299_icssem.2011.6081289-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002299_icssem.2011.6081289-Figure2-1.png", + "caption": "Figure 2. 1120 model and boundary conditions", + "texts": [ + " The research results can provide theoretical basis for the engineering application of the HVD technology. II. COMPUTATIONAL MODEL It's known that the inner radius and outer radius of the friction disk are 130mm and 180mm, a number of the groove is 20, a depth of the groove is 0.5mm, a thickness of the oil film between the friction pair is 0.5mm. The oil film model built in 3-D software is shown in figure 1. At the same time, we build 1120 model according to the symmetry of the oil film, and boundary conditions are set in figure 2. The two side faces of the oil film model are defined as periodic boundary conditions. After all conditions are defined, the created models in Gambit software are exported to the software FLUENT. Then the viscous model, boundary conditions and material properties are defined in FLUENT. According to the principle of CFD, the model was solved with FVM [8-10]. No. 8 hydraulic transmission oil is used as working medium in the HVD. Properties of the oil are as follows: density = 820 kglm3, Cp = 2000 J/(kg\u00b7 \u00b7C), thermal conductivity = 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002480_2015-01-0680-Figure12-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002480_2015-01-0680-Figure12-1.png", + "caption": "Figure 12. Surface comparison between (a) a floating pin and pin bore after 30 minutes test (with no obvious surface damage, and (b) a fixed pin and piston bore after 5 minutes test (with severe damage, shown with red arrows)", + "texts": [ + " It was found that after 5 minutes of testing, there was severe damage on the pin bore surface and on the fixed pin on which some of the pin bore material was transferred. However, minimal scuffing occurred on the pin bore surface running against the floating pin for 30 minutes. Furthermore, there was no surface damage or material transfer on the floating pin. From the comparison, it is obvious that the floating pin has higher scuffing resistance than the fixed pin. The comparison of the scuffed surfaces is shown in Figure 12. The floating pins rotated in the piston bore when the tests began and the rotation motion became slower and slower until it halted. Two tests were run with fixed pins. For the fixed pins, after 5 minutes the tests were stopped due to noise, vibration and significant heat generation. A total of six tests were run with the floating pins. Table 1 shows the test condition for both fixed and floating pins. The pin rotation time from beginning of test to stop for the floating pins is summarized. From Table 1, it can be seen that the floating pin stop time ranges from 23 minutes to 39 minutes" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002039_cdc.2014.7039507-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002039_cdc.2014.7039507-Figure1-1.png", + "caption": "Fig. 1. Line-of-sight Measurements: The leader, Spacecraft 1, measures the LOS toward to distinct objects A,B to control its absolute attitude (the object A is selected as Spacecraft 3). To control the relative attitude between Spacecraft 1 and Spacecraft 2, they measure the LOS toward each other, and also toward the common object selected as Spacecraft 3.", + "texts": [ + " (iii) d dt\u03a821 = e21 \u00b7 e\u21262 + \u039321\u2016e21\u20162, where \u039321 is a positive constant that can be determined by ni [15]. (iv) \u2016 ddte21\u2016 = 1\u221a 2 (k\u03b121 + k\u03b221)\u2016e\u21262 \u2016+ (B\u2126d +B21)\u2016e21\u2016. Proof: See [15]. The position and velocity error vectors of Spacecraft 1 are given as ex1 = x1 \u2212 xd1, ev1 = v1 \u2212 vd1 = e\u0307x1 . (44) Similarly, the relative position and relative velocity error vectors are given by ex21 = x21 \u2212 xd21, ev21 = v21 \u2212 vd21 = e\u0307x21 , (45) Based on the error variables defined at the previous section, control systems are designed as follows. Proposition 3: Consider the formation of two spacecraft shown in Fig 1. For positive constants, k\u2126i , kxi , and kvi , i \u2208 {1, 2}, the control inputs are chosen as follows u1 = \u2212eb \u2212 k\u21261 e\u21261 + \u2126\u0302d1J1(e\u21261 + \u2126d1) + J1\u2126\u0307d1, (46) u2 = \u2212e21 \u2212 k\u21262 e\u21262 + \u2126\u0302d2J2(e\u21262 + \u2126d2) + J2\u2126\u0307d2, (47) f1 = \u2212kx1 ex1 \u2212 kv1ev1 +m1x\u0308 d 1, (48) f2 = \u2212kx2 ex21 \u2212 kv2ev21 +m2(x\u03081 + x\u0308d21). (49) Then, the zero equilibrium of tracking errors is exponentially stable. Proof: See [15]. The preceding results for two spacecraft are readily generalized for an arbitrary number of spacecraft. Here, we present the controller structure as follows, without stability proof that can be obtained by generalizing the proof of Proposition 3" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001121_jae-140037-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001121_jae-140037-Figure2-1.png", + "caption": "Fig. 2. Finite element model of the simulated induction motor.", + "texts": [ + " Under a certain level of static eccentricity, the magnitudes of these components in search coils vary with the air-gap length. Therefore, the position of search coil with the maximum magnitude of induced voltage harmonics is the position of minimum air gap. A 4-pole, 3-phase, 800-kW, squirrel-cage induction motor is simulated in this paper. The detailed information of the studied motor is listed in Table 2. In order to implement the multi-position magnetic field measurement approach, twelve search coils numbered from 1 to 12 are added in the finite element model as Fig. 2 shows. The 12 search coils are equally spaced in stator joke circumferentially. Here the number of search coils and their turns can be modified according to the actual conditions. Using Ansoft Maxwell 15 models the air-gap eccentricity of motor. The static eccentricity is obtained by moving the offset of the stator down and the dynamic eccentricity is obtained by moving the offset of the rotor upwards. Moving the offset of stator and rotor makes them and the offset of rotating of motor do not coincide and the mixed eccentricity is gained" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001288_ascc.2015.7244388-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001288_ascc.2015.7244388-Figure3-1.png", + "caption": "Fig. 3. The basic coordinate axes and forces acting on an aircraft", + "texts": [ + " Note that : Variable candidate; : Variable candidate movement amount; : Variable candidate movement direction The controller designed and MHC were applied to a fixed wing airplanemodel using Simulink MATLAB software. The fixed wing airplane system represented a complex 3rd order SISO process that usually requires a vast amount of analytical analysis and experimentation to achieve fine-tuned performance. This model uses airplane pitch angle only, represents a small part of dynamics system by changing the elevator deflection angle as shown in Figure 3[13][14][[15]. It is assumed that no effect on the speed of the airplane. The transfer function for FWA pitch angle with changes in the elevator deflection angle is[13]. The transfer function above demonstrates that it has a pole on origin in s-plane. This pole should guarantee zero study state error to step set-point change. However, the closed loop transfer function moves that pole away from the origin and brings it towards the zero close to it with gain increase. Consequently, oscillation and large overshooting result" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003143_amc.2014.6823298-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003143_amc.2014.6823298-Figure5-1.png", + "caption": "Fig. 5. Compression type mechanical gravity canceller (COC).", + "texts": [ + " 4, if the robots move in vertical motion, the estimated reaction force becomes the sum of reaction force and gravity force. Therefore, it is difficult to distinguish the reaction force of environment and the gravity force. III. THEORY OF COMPRESSION TYPE MECHANICA L GRAV ITY CANCE L LER (CGC) In this section, theory of compression type of mechanical gravity canceller (CGC) is explained. CGC is a simple way to compensate the gravity force using slider crank mechanisms and compression springs. The link model of CGC is shown in Fig. 5, where 0 denotes the rotation axis. The crank mechanism and the rotational axis are connected at point A, and the rod is connected with the link at point B. From Fig. 5, the tension of compression spring compensates the gravity torque by using the slider crank. The configuration of CGC is shown in Fig. 6, where 0', m, g, h, i, p and is respectively denote the tip of link, the weight of the link, the gravity acceleration, height between 0 and A, length between 0 and 0', length between 0 and B and the stretch of spring. When the posture of the link changes the rod slides, the interval between the spring stopper and the oscillating block slider decreases, and the spring is compressed", + " 6, the following relationship is introduced lssinip =pcosB. (17) Then, from Eq. (16) and Eq. (17), the torque of spring TL can be expressed in function of B TL = kph cos B. Besides, the gravity torque of link TO is given by TO = rngl cos B. Here, the spring constant k is defined as k = rngl. ph (18) (19) (20) In this case, the torque around rotation axis 0, Tgrav can be expressed as Tgrav TL -TO khp cos B - rngl cos B O. (21) Therefore, it is possible to compensate gravity force at any angle of link. In addition, from Fig. 5, the value of compression length is expressed as l\ufffd = p2 + h2 - 2phcosB. (22) The spring constant is defined by equaling the length of AB and the shrinkage of the spring. If the range of angle is defined as, (23) where B MIN and B MAX respectively denote minimum angle of motion and maximum angle of motion. Using Eq. (22) and maximum angle, the maximum spring displacement can be obtained as, (24) Moreover, using Eq. (22) and minimum angle, the initial spring displacement which is needed to get the characteristic of zero length spring can be expressed as, (25) When Eq" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003620_ropec.2014.7036354-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003620_ropec.2014.7036354-Figure4-1.png", + "caption": "Figure 4. Turbine representation.", + "texts": [ + " In other words, a causal half stroke put on the flow side of a bond means a flow imposed on the variable associated with the far end of this bond, while a causal half stroke put on the effort side of a bond means an effort imposed on the variables associated with the near end of this bond. Figure 2 shows the bicausal bonds representation The electric and aerodynamic connection of a turbine can be represented by the moment equations. As mentioned before in this article a rigid couple is considered, then the turbine mass, the gearbox and the generator masses are represented by only one moment of inertia J (Figure 4). III. SYSTEM DESCRIPTION. The tip-speed ratio A. is the relationship between the wind speed and blades angular speed in the turbine and it is given by equation (2), The aerodynamic conversion principle of WT is widely described in the literature [13-15]. The WT transforms wind kinetic energy. The power obtained by the turbine can be represented by equation (1). Torque Tt and the power of the turbine Pt are bounded by the angular speed of the turbine in equation (4). Neglecting mechanic losses by friction, meaning, friction coefficient Dt == O" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001757_cjme.2015.0119.050-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001757_cjme.2015.0119.050-Figure4-1.png", + "caption": "Fig. 4. Three cases incurred by the fourth point", + "texts": [ + " (9) means that P(4) is in the parallel area of L13,2, while the equation set in Eq. (10) means that P(4) is in the intersection of the 11 areas corresponding to the 11 equations. In a similar way, when the saddle line changes from L12,3 to L23,1, we also have an equation and an equation set, that is 23,1 23,1,M = (11) { }, 23,1 .M ij k > (12) Case 3. The fourth point P(4) is one of the three characteristic points; the saddle line changes. Apparently, P(4) will meet case 3 so long as the point dissatisfies cases 1 and 2. Fig. 4 shows the three different cases discussed and illustrate the relationship between the position and the effect of the fourth point. From sections 3 and 4, the way to find the multi-point saddle line can be concluded. For multiple positions, point P traces a planar discrete point set R(i) P (i=1, 2, 3, , n). Any three of the discrete points can be grouped together to obtain three distributing lines and three fitting errors. There are C3 n groups in total. If the other discrete points are in the parallel area of a fitting line, the fitting error is a valid error, otherwise, an invalid error" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002903_s1068798x11100248-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002903_s1068798x11100248-Figure3-1.png", + "caption": "Fig. 3. Combinations of the deviations of sinusoidal pin and bearing contact surfaces, with the same periods and with a phase shift of half a period.", + "texts": [ + " d m ax d r a d m in d m ax d m in d m ax d r a d m in d m ax d r a d m in d m ax d r a d m in x y O y y x x y x y x O O O O d r a Pin Bush Pin Pin Bush Bush Pin Bush Pin Bush (a) (b) (c) (d) (e) 970 RUSSIAN ENGINEERING RESEARCH Vol. 31 No. 10 2011 SANINSKII et al. their amplitudes. The maximum oil volume in the gap may be determined by means of AutoCAD software. Experiments indicate that the maximum oil vol ume in the gap is 212.8758 mm3. In that case, the period of the sinusoidal generatrices is the same, and the phase shift is half a period (Fig. 3). This combina tion ensures optimal liquid friction. We may identify the following types of frictional surface combinations: \u23afindividual cylindrical surfaces\u2014for example, a connecting slip bearing; \u23afindividual composite bearings, consisting of a series of coaxial bearings\u2014for example, the crank shaft bearings of a diesel engine, whose working aper tures are coaxial with a common axis [2]; \u23afcoaxial\u2013concentric multicomponent slip bear ings, consisting of a series of individual bearings, whose working apertures have different diameters and are coaxial with a common axis; norms are specified for the concentric structure or the deviation of each individual axis from the common axis [5]; \u23afcomposite slip bearings, whose surfaces consist of two, three, or more elements (bushes)\u2014for exam ple, the bearings with wear resistant plastic cores" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001388_047134608x.w4531.pub2-Figure13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001388_047134608x.w4531.pub2-Figure13-1.png", + "caption": "Figure 13. Loci showing small differences of rotating field vectors and large effect of anisotropy under the same peak flux density (1.2T) for clockwise and anticlockwise rotation (46).", + "texts": [ + " Rotational flux causes rotational losses in electrical steels at power frequencies and can be obtained from the following expansion of equation 15: Pr \u00bc 1 T Z T 0 hx dbx dt \u00fe hy dby dt dt (25) where hx and hy are instantaneous tangential surface field components in orthogonal directions. dbx=dt and dby=dt are the rates of change of corresponding flux density components. The components in equation 25 can be obtained experimentally. The rotational field variation needed to produce pure rotational flux density is complex in grainoriented material as shown in Figure 13. The B locus is circular whereas the instantaneous field variation is very complex due to the high anisotropy of the material and the difficulty of maintaining the flux density constant when the magnetization is along difficult directions in the plane of the sheet. POWERCORE strip is a consolidated sheet of iron-based amorphous material. The losses produced under rotational fields are generally higher than those produced under AC magnetization. Figure 14 shows the variation of rotational loss with flux density in typical electrical steels and amorphous magnetic material Metglas 2605-S2 in the form of a single ribbon and consolidated strip (23)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure10-1.png", + "caption": "Figure 10 OUTPUT GEAR AND CUTTER", + "texts": [], + "surrounding_texts": [ + "This example considers a spur bevel gear set for motion transmission between intersecting axes to introduce the developed process. The tooth profile is a standard involute tooth profile. The nominal gear pair data is presented in Table 2 whereas the nominal cutter data is presented in Table 3. Figures 9 and 10 show the gear elements in mesh with the hyperboloidal cutter elements. No geometric, rating, or manufacturing data are generated." + ] + }, + { + "image_filename": "designv11_84_0003789_amm.532.378-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003789_amm.532.378-Figure2-1.png", + "caption": "Fig. 2 A 3-R R PPM. Fig . 3 Wrenches and intersection points of wrenches.", + "texts": [ + "159, Pennsylvania State University, University Park, USA-25/05/15,18:46:48) ab \u039b cd = abc d \u2212 abd c = \u03b1e, (2) here the brackets denote the determinants of the matrices which contain the points as their columns and \u03b1 is a nonzero scalar and it dependents on the selection of points a, b, c and d. Two parallel lines intersect at infinity as shown in Fig. 1(c). 3.1 Forward statics. A wrench acts on a rigid body in 2-space may be express as the join of two distinct points on the wrench axis just like Eq. (1). It can be normalized in the form of = \uff04 = ( , ; ), (3) this representation separates the magnitude of the force f from the unit line segment\uff04. The total wrench which acts on the move platform of 3-RPR PPMs (Fig. 2) is consists of a linear combination of the three wrenches along prismatic joints and can be express as = , (4) where = \uff04 \uff04 \uff04 denotes the wrench Jacobian and = . 3.2 Inverse statics. The inverse statics problem of 3-RPR PPMs can be solved from a symbolic formula which without calculate the inverse of j is expressed in the form of = = \uff04 /\uff04 \uff04 \uff04 /\uff04 \uff04 \uff04 /\uff04 \uff04 . (5) where \uff04 , \uff04 and \uff04 are the normalized unitized homogeneous line coordinates of the lines through the intersection points labeled 23, 13 and 12, which are perpendicular to the plane of the page as shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003505_amm.687-691.7-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003505_amm.687-691.7-Figure2-1.png", + "caption": "Fig. 2 Relative position of the effective known points and the major flank face", + "texts": [ + " 1. Choose initial parameters\u2206b0, r0, and error control amount\u03b5h1, and let n=0; 2. Judge that whether|Sh1\u00b7r\u2206b1|\u2264\u03b5h1 and |Sh1\u00b7r1|\u2264\u03b5h1are valid, if so, then go on to step 4; 3. Letn=n+1, use Eq. (4) to calculate new \u2206bkand rk, then go on to step 2; 4. Let\u2206b*=\u2206bn and r*=rn, the distance from the known point to the major flank face is h h1 * * h1(\u2206 , , ) ( , )D = b r r\u03d5 \u03d5\u2211 \u2211 \u2211\u2212r r (5) Whether does the interference occur at the points of tooth surface through the following method to judge.As shown in Fig. 2, assuming the investigating point on the major flank face is P, the known point is Q. If point Q lies at the outside of the major flank face, then, this point is atno interference state, accordingly h1 1 0PQ 0 (21) ki is a positive design parameter. Step r: In the final step the external control v appears. Defining Vr \u00bc Vr 1 \u00fe 1 2 z2 r (22) gives V\u0307r \u00bc V\u0307r 1 \u00fe zrz\u0307r \u00bc Xr 1 j\u00bc1 \u00f0 k jz 2 j \u00de \u00fe zr\u00f0zr 1 \u00fe v y\u00f0r\u00des p a\u0307r 1\u00de (23) Now, the external control v is chosen as v \u00bc krzr zr 1 \u00fe y\u00f0r\u00des p \u00fe a\u0307r 1 (24) This obtains V\u0307r \u00bc Xr j\u00bc1 k jz 2 j ; ki > 0 (25) The design is complete. A schematic diagram of a continuous fermenter is shown in Fig. 1. The unstructured model of a continuous fermentation with a constant volume and well-mixed condition can be described by the following equations [15,18]: x\u03071 \u00bc Dx1 \u00fe m\u00f0x2; x3\u00dex1 (26) x\u03072 \u00bc D\u00f0S f x2\u00de 1 Yx1=x2 m\u00f0x2; x3\u00dex1 (27) x\u03073 \u00bc Dx3 \u00fe \u00bdam\u00f0x2; x3\u00de \u00fe b x1 (28) where states x1, x2 and x3 denote biomass concentration, substrate concentration, and product concentration, respectively, D is the dilution rate, Yx1=x2 is the cell-mass yield with respect to the limiting substrate, Sf is the feed substrate concentration and a and b are yield parameters for the product" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003468_robio.2011.6181550-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003468_robio.2011.6181550-Figure6-1.png", + "caption": "Fig. 6. y-phi coordinate", + "texts": [ + " f n indicates a tension generated by n-th motor. The Jacobian matrix of this tendon mechanism shown in (3) is J j = [ \u2212R1 R1 0 0 \u2212R1 R1 0 0 \u2212R2 R2 \u2212R2 R2 ]T (16) where, R 1 and R 2 are pulley radius of joint 1 and joint 2, respectively. Focusing on from first-row to fourth-row in (16), zeros appear and it contributes to reduction of interaction, which is required for the simple tension distribution. B. Definition of Workspace We have been deriving workspace force and workspace stiffness as shown in (14). Polar coordinate shown in Fig.6 is adopted. In this coordinate, a position is represented by rotation angle \u03c6 and distance y. A jacobian matrix transforming joint angle displacement \u03b4\u03b8 j = (\u03b4\u03b8 j1 , \u03b4\u03b8 j2 )T into \u03b4p = (\u03b4y, \u03b4\u03c6)T are expressed as follows: J = [ 0 \u2212L sin \u03b8 j2 2 1 1 2 ] (17) Here, L stands for a length of first and second link. Because human body structure has the same arm length, and we utilize biological extremity also for arm length to realize simple tension distribution. If the Cartesian polar is adopted, zero in (17) would not appear and the jacobian would become much complex", + "[ c 1 , \u00b7 \u00b7 \u00b7 , c 6 ] (18) d = [ d 1 , \u00b7 \u00b7 \u00b7 , d 6 ]T (19) H = R2 \u23a1 \u23a3 1 4 A2 1 4 A2 A2 A2 1 4 A2 1 4 A2 1 2 A 1 2 A 0 0 \u2212 1 2 A \u2212 1 2 A 1 1 0 0 1 1 \u23a4 \u23a6 (20) A = 1 L sin \u03b8 j2 2 By substituting (16), (18), (19), and (20) into (14), and expressing in the form of simultaneous equations, we obtain the relation of the workspace force F w = (F y , \u03c4 \u03c6 )T , the workspace stiffness S w = (S y , S y\u03c6 , S \u03c6 )T , and the tension f called basic equations. F ref\u2032 y = 1 2 f 1 \u2212 1 2 f 2 \u2212 f 3 + f 4 \u2212 1 2 f 5 + 1 2 f 6 \u03c4 ref \u2032 \u03c6 = f 1 \u2212 f 2 + f 5 \u2212 f 6 Sref\u2032 y = 1 4 c 1 f 1 + 1 4 c 2 f 2 + c 3 f 3 + c 4 f 4 + 1 4 c 5 f 5 + 1 4 c 6 f 6 Sref\u2032 y\u03c6 = 1 2 c 1 f 1 + 1 2 c 2 f 2 \u2212 1 2 c 5 f 5 \u2212 1 2 c 6 f 6 Sref\u2032 \u03c6 = c 1 f 1 + c 2 f 2 + c 5 f 5 + c 6 f 6 (21) Considering the workspace defined as Fig.6, F y and S y stand for y directional force and stiffness at the end of the arm, respectively. \u03c4 \u03c6 and S \u03c6 stand for torque and stiffness at the first joint of the arm, respectively. S y\u03c6 is interactive stiffness of y-direction and rotational direction. The commands such as F ref y are expressed with prime like F ref\u2032 y to make the basic equations as simple as possible. Commands related to rotational direction \u03c4 ref \u2032 \u03c6 , Sref\u2032 \u03c6 , and Sref\u2032 y\u03c6 are depending only on f 1 , f 2 , f 5 , and f 6 . On the other hand, y-directional commands F ref\u2032 y and Sref\u2032 y are depending on all of tension from f 1 to f 6 ", + " CONCLUSION The tension distribution algorithm for the tendon mechanisms utilizing musculoskeletal model of biological extremities is reported and its effectiveness is confirmed by numerical calculation. The features of the proposed method are utilizing musculoskeletal model and polar coordinate definition. The reason why we utilize musculoskeletal model is to reduce the interaction of each tensions when doing algebraic calculation for the tension distribution. And coordinate definition shown in Fig. 6 are adopted to remain the reduction of interaction even when considering tension distribution for workspace commands. As a result, we obtain tension distribution method, which is applicable to workspace commands. The proposed method can consider the tension limit and the workspace command priority and derive the tension command which achieves given workspace commands as much as possible in accordance with the priority. The capability to directly consider workspace command priority makes the proposed method more useful" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.25-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.25-1.png", + "caption": "Fig. 2.25. Stanford manipulator", + "texts": [ + " Notice that, as a consequence of the choice made for the coordinate frames, the block matrix R3 6 that can be extracted from T 3 6 coincides with the rotation matrix of Euler angles (2.18) previously derived, that is, \u03d14, \u03d15, \u03d16 constitute the set of ZYZ angles with respect to the reference frame O3\u2013x3y3z3. Moreover, the unit vectors of Frame 6 coincide with the unit vectors of a possible end-effector frame according to Fig. 2.13. The so-called Stanford manipulator is composed of a spherical arm and a spherical wrist (Fig. 2.25). Since Frame 3 of the spherical arm coincides with Frame 3 of the spherical wrist, the direct kinematics function can be obtained via simple composition of the transformation matrices (2.65), (2.67) of the previous examples, i.e., T 0 6 = T 0 3T 3 6 = \u23a1 \u23a2\u23a3n 0 s0 a0 p0 0 0 0 1 \u23a4 \u23a5\u23a6 . Carrying out the products yields p0 6 = \u23a1 \u23a3 c1s2d3 \u2212 s1d2 + ( c1(c2c4s5 + s2c5) \u2212 s1s4s5 ) d6 s1s2d3 + c1d2 + ( s1(c2c4s5 + s2c5) + c1s4s5 ) d6 c2d3 + (\u2212s2c4s5 + c2c5)d6 \u23a4 \u23a6 (2.68) for the end-effector position, and n0 6 = \u23a1 \u23a3 c1 ( c2(c4c5c6 \u2212 s4s6) \u2212 s2s5c6 ) \u2212 s1(s4c5c6 + c4s6) s1 ( c2(c4c5c6 \u2212 s4s6) \u2212 s2s5c6 ) + c1(s4c5c6 + c4s6) \u2212s2(c4c5c6 \u2212 s4s6) \u2212 c2s5c6 \u23a4 \u23a6 s0 6 = \u23a1 \u23a3 c1 ( \u2212c2(c4c5s6 + s4c6) + s2s5s6 ) \u2212 s1(\u2212s4c5s6 + c4c6) s1 ( \u2212c2(c4c5s6 + s4c6) + s2s5s6 ) + c1(\u2212s4c5s6 + c4c6) s2(c4c5s6 + s4c6) + c2s5s6 \u23a4 \u23a6 (2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001291_chicc.2015.7260301-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001291_chicc.2015.7260301-Figure3-1.png", + "caption": "Fig. 3. Current coordinate transformation", + "texts": [ + " For the kinematic control, the velocity controller can refer to the authors\u2019 former work [14]: ( cos sin ) ( cos sin ) ( sin cos ) ( sin cos ) x y d dc x y d dc c d z z c d k V V u e v eu k V V u e v ev w w k V r r k V c (14) The kinematic control algorithm can smooth the speed jump problem due to the initial tracking errors and posture change which has been discussed in detail in [14]. Here, the ocean current effect is considered in the kinematic design. If ocean current is considered in the control design, then the desired relative velocity should be changed. For the ocean current [ 0 0 0]T cE cx cy czu u w under inertial frame, through coordinate transformation, it can be transformed into the ocean current under body-fixed frame [ 0 0 0]T cb cu cv cwu v w (see Fig. 3). cos sin sin cos cu cx cy cv cx cy cw cz u u u v u u w w (15) where , ,cu cv cwu v w represents the surge, sway and heave current size respectively. Then, the desired relative velocity under ocean current can be given as , , ,d d cu d d cv d d cw d du u u v v v w w w r r (16) Then, the final kinematic control law can be proposed as: ( cos sin ) ( cos sin ) ( sin cos ) ( sin cos ) x y d dcr x y d dcr cr d z z cr d k V V u e v eu k V V u e v ev w w k V r r k V cr (17) After considering the current interference, the goal of dynamic controller design is to make relative current velocity r of UUV following the reference relative velocity cr under ocean current" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure6.4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure6.4-1.png", + "caption": "Fig. 6.4. Keystone distortion and its rectification", + "texts": [ + " Observer viewing comfort is also greatly increased when camera axes converge to within a few degrees of the eyes\u2019 natural convergence angle for the object being fixated (cf. the Bernardino et al. chapter in this book.) The most straightforward means of making the optical axes convergent is to rotate the cameras towards each other. An alternative method is to translate the sensor with respect to the lens\u2014which requires special mechanisms within the camera. It is possible to simulate sensor translation by using offset subsets of the sensor surfaces, but this can greatly limit the field of view. Fig. 6.4 shows the sensor planes (L and R) of two cameras converged on a planar rectilinear grid. It can be seen that the grid images projected onto the sensors are geometrically distorted such that there is both horizontal and vertical disparity. This is termed keystone distortion. It becomes greater as the angle of convergence increases. If these distorted images are presented to an observer in a stereoscopic 3-D display, the planar grid will appear curved. In the more general case of a scene, the entire scene space will appear curved" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001231_9781118984444.ch1-Figure1.1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001231_9781118984444.ch1-Figure1.1-1.png", + "caption": "Figure 1.1. Simple pendulum", + "texts": [ + " Such systems are called conservative systems. The natural frequencies of conservative systems may be obtained by equating the maximum kinetic energy (Tm) to the maximum total potential energy (Vm) associated with vibration. The meaning of these energy terms is very important. To illustrate the principle of conservation of energy, and the meaning of the energy terms let us study some simple vibratory systems. Consider the oscillatory motion of the simple pendulum consisting of a bob of mass m and a massless string of length L as shown in Figure 1.1. It would be at rest in a vertical configuration under gravity field. If it is given a small disturbance \u03b2m and then released, it will tend to vibrate about this equilibrium state. The restoring action of the gravity force will initiate a motion toward the equilibrium state but as the bob approaches the lowest point in its motion it has a velocity and therefore carries on swinging up on the other side until the gravity force causes it to come to a The Rayleigh\u2212Ritz Method for Structural Analysis, First Edition" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002494_amr.1028.105-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002494_amr.1028.105-Figure10-1.png", + "caption": "Fig. 10 Vibration mode of order 5", + "texts": [], + "surrounding_texts": [ + "The static structural analysis of the drive shaft is used for pre-stressed modal calculation in Model module, therefore the natural frequencies and corresponding vibration modes of the shaft can be got. Normally we do not have to find all the natural frequencies and mode shapes. Low order natural frequencies and vibration modes have bigger impact on the vibration of the stepper motor [4],so the first six natural frequencies and vibration modes obtained in ANSYS Workbench are concerned, as shown in Figure 6-11 and Table 1.Because the amplitude is relative value after treatment, it doesnot reflect the actual amplitude [3]. 1) Natural frequency analysis. Through access to relevant information, the stepper motor has a fixed resonance region. The resonance region of the two-four phase stepper motor is generally between 180 and 250PPS (step angle of 1.8 degrees). As the drive voltage of the stepper motor is higher and the load is lighter, the resonance region is upward. However, each order natural frequency of the drive shaft is more than 1745.2HZ, so to ensure the tube reaches the specified location accurately and avoid resonance between the stepper motor and drive shaft, the operating frequency of the stepper motor must be between 300PPS and 1700PPS. 2) Vibration mode analysis. Though the analysis of the first two modal shapes (shown in Figure 6, 7), the central part of the shaft has the greatest amplitude of the resonance, therefore probably becomes the weakest portion. So without affecting the transmission accuracy, belt pulleys are arranged on both sides of the shaft as far as possible." + ] + }, + { + "image_filename": "designv11_84_0002275_s1064230714040042-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002275_s1064230714040042-Figure6-1.png", + "caption": "Fig. 6. Trajectories of types A and B: (a) nondegenerate case; (b) degenerate case.", + "texts": [ + " Figure 5c shows the degenerate case when there is a degen erate segment with on the optimal trajectory (in Fig. 5c, this segment is associated with the points S1 and S2). Due to the optimality principle (any part of the optimal trajectory is also an optimal trajectory), the optimal trajectory can pass only through one of the points S1 or S2 and only once (Figs. 5b and 5c). Hence, only the control sequences listed in Proposition 2 are possible, and . If , then the initial position of the system is associated with two types of extremal trajectories (Fig. 6). Solutions of type A are associated with the trajectories OAD (the control sequence is ) or OAMD (the control sequence is ). Solutions of type B are associated with the trajectories OBD (the control sequence is ) or OBND (the control sequence is ). Any combination of degenerate and nondegenerate trajectories of types A and B is possible. Let be the time of motion on the trajectory of type A, \u03c4A be the time of motion on the segment OA, and in the degenerate case \u03c4M be the time of motion on the segment AM (Fig. 6). 2 2 1 1 1const ( ) 0. ( ) S S x x \u03c8 = = \u03c8 = > \u03c9 mu u= mu u= \u2212 + \u0393 mu mu\u2212 0u \u2261 1 1 2[ , ] [ , ]x S S\u2208 \u2282 \u2212\u03c0 \u03c0 0 0 1 2 1 2 1 2( , )x x R R R R\u2208 \u222a \u222a \u222a { },m mu u\u2212 { }, 0,m mu u\u2212 { },m mu u\u2212 { }, 0,m mu u\u2212 AT { }, 0,m mu u\u2212 { }, 0,m mu u\u2212 JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL Vol. 53 No. 4 2014 OPTIMAL ATTITUDE CONTROL OF TWO PIVOTALLY CONNECTED BODIES 619 In the nondegenerate case, the trajectory OAD on the segment OA with satisfies relations (2.8). The constants of integration d1 and d2 in (2", + "B A A B m m x x x T T u u \u03d5 \u03c0\u03c0 = \u2212 = \u03c0 = = \u2212 \u2212 \u03c9\u03c9 H + \u2212 \u2208\u0393 \u0393\u222a H H A BT T= H + \u2208\u0393 O A H= = B D= H \u2212 \u2208\u0393 O B H= = A D= 0 0 1 2 1 1( , )x x R R\u2208 \u222a 0 1 1 Dx x< \u2212 A BT T< 1 1R R \u222a H + \u2212 \u2208\u0393 \u0393\u222a 0 1 1 constHx x= = 0 1x 1 1 11 1 0 0 1 1 2 2 ( ) ( ) ( ) ( ), . 2 A A AA A m A A ud x d x d xdx x dx dx dx dx \u03d5 \u03d5 \u03d5 = \u03c9 = = JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL Vol. 53 No. 4 2014 OPTIMAL ATTITUDE CONTROL OF TWO PIVOTALLY CONNECTED BODIES 621 This implies By differentiating (A.3), we obtain (A.11) Similarly, we have (A.12) Let the initial position of O (Fig. 6a) be in the domain , that is, . The point E corre sponds to the initial position . Due to Lemma 2, we have and . The function is determined by (2.7); it is even and monotonically decreasing on the set (Fig. 7). Hence, , and, due to (A.11) and (A.12), (A.13) For the degenerate trajectories of types A and B, we differentiate (A.9) and (A.10) to obtain (A.14) Due to Lemma 3, for . Then, for (the initial position of O varies along the line HO), (A.13) and (A.14) imply ; that is, the trajectory of type A is optimal" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001312_icuas.2015.7152309-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001312_icuas.2015.7152309-Figure9-1.png", + "caption": "Figure 9: Vane approximation as flat plates normal to wind direction.", + "texts": [ + "01m2 at \u03b2c=20\u00b0, and a mass of 2kg, the device is estimated to have a terminal velocity of about 57ms-1; within range of the skydiver\u2019s terminal velocities. Yaw moment vs. Vane pitching The model was simulated under a nominal wind of 54ms-1 with \u03b2d= 0, \u03b2c= 0, only varying the vane pitch angle \u03b3. From figure 8, the fitted function, \ud835\udc36\ud835\udc40,\ud835\udc67\ud835\udc34 = \ud835\udc504\ud835\udefe + \ud835\udc505 (4) where c4 and c5 are constants, shows that the yaw moment varies linearly with the vane pitch angle \u03b3 for the entire range available (Equation 4). These were tested at the nominal collective angle of 20\u00b0. Figure 9 shows the vane as an approximated flat plate normal to the airflow direction, thus as the vane is pitched, the lift increases in a linear fashion, regardless of their curvature, due to their thin cross-section. Verification of stability With a \u03b2d of zero degrees, and \u03b2c at the nominal angle of 20\u00b0, it is shown that the device has a negative moment gradient when exposed to the nominal airspeed of 54ms-1 at various positive angles of attack (Figure 12). This suggests that the device shape is aerodynamically stable" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002384_2011-01-1424-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002384_2011-01-1424-Figure2-1.png", + "caption": "Fig. 2. Variator Module Test Rig", + "texts": [ + " It is from this background that the following work was undertaken to investigate the high temperature durability of the full-toroidal variator. At the same time a number of long term fatigue durability tests have been undertaken to investigate the applicability of an endurance limit on the disc and rollers. The fatigue endurance tests were undertaken at 1.78GPa, 1.91GPa & 2.03GPa The data presented in Figure 1 and the currently reported results were carried out on a power re-circulating four-square test rig, shown in Figure 2, which incorporates a 100mm (roller OD) full-toroidal variator test module. Hence real contact geometry and operating conditions could be used. The ratio (Rv) of the input disc speed to the output disc speed was set via the rig gearbox; testing being conducted at Rv= \u22120.998 such that the contact conditions on either side of each roller were very similar. The same nominal endload (axial clamping force) was used for the variator material durability tests, the roller conformity (ratio of roller crown radius and roller radius) being changed to obtain the required contact stress" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003814_iciea.2013.6566511-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003814_iciea.2013.6566511-Figure1-1.png", + "caption": "Fig. 1. n-DOF space manipulator model based on SOA method", + "texts": [ + "00 c\u00a92013 IEEE where 1H denotes the contribution from the base velocity to the system angular momentum, while 2H denotes the contribution from the joint angular velocity to the linear and angular momentum of the system. Substituting Eq. 4 into Eq. 1 leads to the relationship between end-effector velocity and joint angular velocity: ( )1 1 2 =e m b m m \u2212 \u2217= \u2212x J J H H q J q (5) where \u2217J denotes the generalized Jacobin matrix. B. Dynamic Modeling In order to improve the computational efficiency, the Spatial Operator Algebra (SOA) method is used to establish the dynamics equation. The manipulator composed of manipulator joints and links is shown in Fig. 1. Define the last joint as the 1st joint, and count from end-effector to the base, where the base can be expressed as the 1thn + joint. Symbols in the manipulator system are expressed as follows: I\u03a3 : Inertial coordinate system, which all the recursive computations are relative to it. k\u03a3 : Coordinate system of the thk link, which is defined at joint position. kJ : Joint k , n is the base, 0 is the connection between end-effector and outside. kC : The centroid of the thk link. ka : The vector from kJ to kC " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001440_ever.2015.7112927-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001440_ever.2015.7112927-Figure1-1.png", + "caption": "Fig. 1. Magnetostatic FEA: estimation of fluctuations in the average air-gap magnetic field introduced by the stator teeth, the rotor PM, respectively.", + "texts": [], + "surrounding_texts": [ + "Any mechanical piece that vibrates in the air produces noise. In electric machines, magnetic vibrations are mainly due to the excitation of the mechanical system by harmonics of air-gap Maxwell stress, also called electromagnetic forces. Our electromagnetic model calculates radial component of Maxwell stress on the average air-gap. These pressures are expressed by Eq.1 [11], where Br and B\u03b8 stand for the radial and tangential components of the air-gap flux density, respectively. \u03b1 is the spatial air-gap discretization, and t is the time discretization. \u03c3r(t, \u03b1) = 1 2\u00b50 [ B2 r (t, \u03b1)\u2212B2 \u03b8 (t, \u03b1) ] (1) The total air-gap flux density, B(t, \u03b1), is the product between the conjugate of the air-gap permeance \u039b(t, \u03b1) and the sum of magnetomotive forces developed by the stator winding fsmm(t, \u03b1), and PM rotor frmm(t, \u03b1) (exponents s and r mean stator and rotor respectively). Using complex notations, we have: B = Br + jB\u03b8 = \u039b\u2217 \u00d7 [Fr mm + Fs mm] (2) where B, \u039b and F are nt \u00d7 na matrix of complex flux density, complex air-gap permeance per area unit and complex magnetomotive forces respectively. na and nt are number of discretization in the space domain (following \u03b1) and in the time domain (following t) respectively. With this formulation, it is possible to apply all the mathematical operations to a single variable, grouping two data." + ] + }, + { + "image_filename": "designv11_84_0003646_amm.86.889-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003646_amm.86.889-Figure5-1.png", + "caption": "Figure 5. Tooth flank form error of gear (\u00b5m) Figure 6. Real contact patterns", + "texts": [ + " The long-cong-distance spiral bevel gears were manufactured based on CAD/CAM system. As mentioned above, the manufacturing processes were divided into three parts, namely, roughing, semi-finishing, and finishing in machining. The milling operation was carried out with a CNC machine in our laboratory as shown in figure 4. Tooth flank form error and tooth contact pattern The actual tooth surfaces are measured on the measurement device. The theoretically calculated tooth surface is used as a baseline for comparison. Fig. 5 compares the tooth topographies, obtained from the mathematical model and the data measured on the real manufactured gear. The tooth flank form errors are not more than \u00b10.20 mm of the gear. Moreover, the tooth contact patterns of the gears were observed and those positions were good, as Fig. 6 shows. From these results, the validity of the manufacturing method using CNC machining center was confirmed. Long-cone-distance gears are units of a small shaft angle and large power transmission. Due to its long cone distance, it is very difficult to or cannot be produced by the dedicated machine tool" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001438_rnc.3175-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001438_rnc.3175-Figure1-1.png", + "caption": "Figure 1. The ball and beam system.", + "texts": [ + "qk/ that assigns the dissipation structure R1.qk; pk/ D .1C 2/ gu.qk/g T u .qk/C gw.qk/g T w.qk/ that satisfies condition (36). Remark 7 Similar to Remark 4, for the systems that have already some dissipation R1; dk can be selected such that dTrk drkD 1 \u02db T QdTr Og 1 u .qk/ .1C 2\u02db2 T 2/ Ogu.qk/ Og T u C Ogw.qk/ Og T w.qk/ 2R1 Og Tu .qk/ Qdr 1 with QdT k Qdk D Im to obtain a less conservative controller. To analyse the performance of the proposed controller, we considered the ball and beam system illustrated in Figure 1, where x is the ball position, the angle of the bar, and L is the bar length. The dynamics of the ball and beam system can be described by the Euler\u2013Lagrange equations [24] Rq1 C g sin .q2/ q1 Pq 2 2 D 0; .LC q21/ Rq2 C 2q1 Pq1 Pq2 C gq1 cos.q2/ D u: (40) These equations are scaled in time and masses.L2 explicitly appears in equations because it is a factor of the moment of the inertia of the bar, for details, see [24]. These dynamics without dissipation can be expressed in the Hamiltonian system formalism as follows Copyright \u00a9 2014 John Wiley & Sons, Ltd", + " Hence, the Hamiltonian model of the system with external disturbance and uncertainty considered in the simulations is Pq Pp D .J Rd .q; p//r QHd .q; p/C 0 gw.q/ w C 0 gu.q/ u; where gTw.q/ D gTu .q/ D 0 1 ; Rd D 0; QHd .q; p/ D 1 2 pT QMd .q/ 1p C QPd .q/; QMd .q/ D Md .q/C Md .q/ and QPd .q/ D Pd .q/C Pd .q/, yielding Hd .q; p/ D 1 2 pT Md .q/ 1p C Pd .q/: The system parameters are taken as Kp D 1:0; L D 10 and g D 0:98ms 2 as in [24]. Simulations are run for T D 0:1. The disturbance input considered isw D 1N for t 2 \u01525; 7 and zero otherwise. The time response of the nominal system under the disturbance is shown in Figure 1. Figure 2 Copyright \u00a9 2014 John Wiley & Sons, Ltd. Int. J. Robust Nonlinear Control 2015; 25:1927\u20131940 DOI: 10.1002/rnc illustrates the closed-loop performances for three cases: attenuation without presence of uncertainty, attenuation in presence of uncertainty, and robust attenuation in presence of uncertainty. About 60% of uncertainty is considered in the masses of ball and bar (60% less than their nominal values). The robust control law to achieve D 0:04 is obtained as follows: Ogu and Ogw are taken as Ogu D I2; Ogw D I2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002457_amr.852.391-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002457_amr.852.391-Figure1-1.png", + "caption": "Figure 1. Quad-rotor model simplified schematic", + "texts": [ + " The tracking error state was used to design a slide mode surface, and a Lyapunov function which includes slide mode surface and unknown parameter was built. Further more, a robust adaptive control law was designed. At last, the designed control law was simulated, and the results justify the feasibility of the proposed control law. Quad-rotor is a multi-variable and strong coupling system which has nonlinear and uncertainties. it has an impact structure and the layout is novel. The four rotor are spread around trestle well-proportioned which are driven by four independent Motor(Figure 1) .The quad-rotor can take off and land vertical so that it has less requirements on the site. As a result, it can play an important role in the future. But because of its model and external environment uncertainties, the quad-rotor\u2019s control is also a difficult problem in academia. Recently, some scholars try to use linearization model and choose specific point to control it, but these kinds of methods have their own limitations and cannot be adapted to the complex and changeable situation. Usually, the classical linear controller design method based on linear model, such as LQ and PID controller design method, can not achieve a satisfactory control precision" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002955_amm.633-634.1111-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002955_amm.633-634.1111-Figure3-1.png", + "caption": "Figure 3", + "texts": [ + " Then further displacement of the puncheon ( )sc\u03b5\u03b5 > will correspond to the development of the main crack, and the condition ( )crsc \u03b5\u03b5 + will correspond to the finish of the destruction process, i.e. the fragmentation of the body. The ultimate stress 0\u03c3 can be determined using nondestructive methods, for example [7]. The durability of material is infinite without the achievement of this stress. The deformation of the rheological element before the start of the crack is determined by correlation: , e 0 sc E \u03c3 \u03b5 = (7) where eE is the coefficient of elasticity of the element. This coefficient can be found using the diagrams \u03c3-\u03b5 during the loading and the unloading. The model shown on the Fig. 3 can be used to verify the calculation of .eE In the model on the Fig. 3 the stress I 1\u03c3 appears in the element 2 when the first loading till the stress .\u03c3 In this case: ,v1 II \u03c3\u03c3\u03c3 \u2212= (8) where ( )I I f \u03b5\u03c3 =v and I\u03b5 is the rate of the relative deformation of the parallel elements during the first loading. The rigid constraint is formed due to the jamming of the element 1 during the unloading. It blocks an action of the viscous element and keeps the residual deformation :1 I\u03b5 . 1 v 1 1 1 EE II I \u03c3\u03c3\u03c3 \u03b5 \u2212 == (9) According to it, the stress-strain relation for whole body is determined by the element 2 during the unloading and at the beginning of the next loading as: ", + " If the maximum cyclic stress is constant, the total \u2211\u2206 i 1\u03c3 tends to the stress ,v I\u03c3 the stress i 1\u03c3 tends to ,v1 II \u03c3\u03c3 + and the stress 1\u03c3 tends to zero. Consequently, the residual deformation for each previous cycle will gradually decrease to zero. The total residual deformation will tend to the value . 1E \u03c3 The stress-strain relations are shown on the Fig. 4. It corresponds to the compression from zero stress to the constant stress \u03c3 with the constant deformation rate as it occurs in the model on the Fig. 3. From the equilibrium condition: ,2v1 \u03c3\u03c3\u03c3\u03c3 =+= (12) where ,1\u03c3 2\u03c3 and v\u03c3 is the stresses correspondingly taken by the element 1, the element 2 and the viscous element. Figure 4 The stress ( )\u03b5\u03c3 f=v does not depend on the value\u03b5 if the relative deformation rate is constant. In this case 21 \u03c3\u03c3\u03c3 ddd == . The absolute strains of the viscous element and the element 1 are equal ( v1 ll \u2206=\u2206 ). The total deformation is .21 lll \u2206+\u2206=\u2206 (13) All these deformations occur on the same length ,l therefore: ,21 \u03b5\u03b5\u03b5 += (14) , 2 2 1 1 \u03c3 \u03b5 \u03c3 \u03b5 \u03c3 \u03b5 d d d d d d += (15) , 111 21 EEE += (16) whereE is the equilibrium coefficient of elasticity ( \u03c3 \u03b5 d d E = if the relative deformation rate is constant)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003143_amc.2014.6823298-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003143_amc.2014.6823298-Figure6-1.png", + "caption": "Fig. 6. Configuration of COCo", + "texts": [ + " THEORY OF COMPRESSION TYPE MECHANICA L GRAV ITY CANCE L LER (CGC) In this section, theory of compression type of mechanical gravity canceller (CGC) is explained. CGC is a simple way to compensate the gravity force using slider crank mechanisms and compression springs. The link model of CGC is shown in Fig. 5, where 0 denotes the rotation axis. The crank mechanism and the rotational axis are connected at point A, and the rod is connected with the link at point B. From Fig. 5, the tension of compression spring compensates the gravity torque by using the slider crank. The configuration of CGC is shown in Fig. 6, where 0', m, g, h, i, p and is respectively denote the tip of link, the weight of the link, the gravity acceleration, height between 0 and A, length between 0 and 0', length between 0 and B and the stretch of spring. When the posture of the link changes the rod slides, the interval between the spring stopper and the oscillating block slider decreases, and the spring is compressed. The compression of spring depends on the range of motion. The torque of spring TL is given by TL = kh1s sin ip, (16) where, k denotes the spring constant. From the geometrical condition of Fig. 6, the following relationship is introduced lssinip =pcosB. (17) Then, from Eq. (16) and Eq. (17), the torque of spring TL can be expressed in function of B TL = kph cos B. Besides, the gravity torque of link TO is given by TO = rngl cos B. Here, the spring constant k is defined as k = rngl. ph (18) (19) (20) In this case, the torque around rotation axis 0, Tgrav can be expressed as Tgrav TL -TO khp cos B - rngl cos B O. (21) Therefore, it is possible to compensate gravity force at any angle of link" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001656_rdcape.2015.7281404-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001656_rdcape.2015.7281404-Figure1-1.png", + "caption": "Fig. 1. Two link robot manipulator", + "texts": [], + "surrounding_texts": [ + "\u2020 \u2020\n1\n( )( , ) ( , ) ( ) k i N k\ni i\nx q i u q i u Q\nx \u03b1\u03b7 \u03b1\n=\n\u2190 + \u0394 (7)\nwhere [0,1]\u03b7 \u2208 is the learning rate parameter.\nA. Lyapunov Stability Theory For analyzing stability of linear systems, we have several approaches like Routh\u2019s criteria, Root Locus. But these techniques can\u2019t be applied for stabilizing non linear (NL) systems. Inherently all physical systems are non linear. Lyapunov theory allows us to inspect stability of NL systems and has two main elements: Lyapunov indirect method and Lyapunov direct method. We propose to use lyapunov direct method as it can be applied directly to a non-linear system without linearization which is needed with indirect approach. Furthermore indirect method examines local stability while direct method could be used to achieve global stability [16]. Lyapunov\u2019s direct method has two steps: a) choose an appropriate positive definite function referred to as Lyapunov function, and b) Ensure that its first order time derivative along all system trajectories is negative definite. If the first order derivative of Lyapunov function is decreasing along the system trajectories it can be concluded that the system will finally settle down as the system energy is dissipating with the increasing time. For further details we refer the interested reader to [16].\nOur attempt herein is to design a controller that can not only handle continuous state action spaces that are encountered in the controlling of a robotic manipulator but also remains stable when handling disturbances and parameter variations. Robotic manipulators [14] are highly coupled, non linear and uncertain systems. Further a robotic manipulator has to handle varying payload mass when picking up and releasing objects. This presents a highly challenging and complex task for testing our proposed control approach. We test out approach on a 2 DOF robotic arm manipulator as the results obtained are applicable to the corresponding 6 DOF manipulator with less computations involved in the former. The fuzzy Q learning approach as detailed in the previous section is an excellent algorithm for extending Q learning to continuous state action space problems using fuzzy inference systems. However, it does not guarantee the stability of the control regime. Herein, we propose to formulate a controller that not only handled the uncertainties such as parameter variations and disturbances but is stable as well. This\nis sought to be achieved by incorporating a lyapunov based control element in the fuzzy Q learning approach. This combination of the two approaches is carried out in a gradual manner based on the experiential information obtained during the execution of the hybrid control approach. For each link of the robotic manipulator we have\ntwo state variables . 1 1 1( , )k k kx \u03b8 \u03b8= and .\n2 2 2( , )k k kx \u03b8 \u03b8= . The control law based on lyapunov theory that makes system to be locally asymptotically stable around the unstable equilibrium point for a very large attraction domain [11] for link 1 of the manipulator is:\n1 1 1 1 1 1 1\n1( ) [ cos ( , )] ( ) k k k k k lyap ku x k\u03b8 \u03b8 + \u03be \u03b8 \u03b8 \u03c3 \u03b8 \u2212= (8)\nwhere, 2 1 1 1( ) 1 cosk kk\u03c3 \u03b8 \u03b8= \u2212 2\n1 1 1 1 1( , ) ( [ ] cos )sink k k k k\u03be \u03b8 \u03b8 \u03b8 \u03b8 \u03b8= \u2212 +\n1 0k > is a strictly positive constant. Similar expression is obtained for the link 2 of the\nmanipulator for .\n2 2 2( , )k k kx \u03b8 \u03b8= [11].\nWe have taken the two-link robot model from [14]. A robot manipulator having n joints may be modeled as per:\n( ) ( , ) ( ) ( ) dM q q C q q q g q f q uu+ + + + = .. . . .\n(9)\nq corresponds to 1n \u00d7 vector of joint angles, u is 1n \u00d7 vector of torques applied at each joint, ( )M q\nspecifies n n\u00d7 inertial vector matrix, ( , )C q q . is n n\u00d7\nCentrifugal forces and Coriolis vector, ( )f q .\ncorresponds to the 1n \u00d7 frictional vector, ( )g q refers to 1n \u00d7 gravitational vector, while du refers to a 1n \u00d7 vector of external disturbance.\nEach link variable has been fuzzified into three partitions which lead to nine rules corresponding to one link and overall eighteen rules for both the links. We use Gaussian membership function to fuzzify each input variable.\n2( )\n22( ) e ; 1,2,3; 1,2,3, 4;\nl p j j\nj l j pp\nx x\nx l j \u03c3\u03bc\n\u2212 \u2212\n= = =\n(10)\ni.e, Gaussian membership function, pl is label (fuzzy) for the link j in joint n,", + "1 1 2 2( (1) (1), (2) (1), (1) (2), (2) (2))x x x x\u03b8 \u03b8 \u03b8 \u03b8= = = = . .\n. The centers and width corresponding to these fuzzy variables are ( ) ( ) ( )( 1)\nl p j j pjx n a n b n l= + \u2212 with\n1 1(1) (2) 2a a= = \u2212 , 2 2(1) (2) 10a a= = \u2212 , 1 1(1) (2) 2b b= = , 2 2(1) (2) 10b b= = , being center values and width values are 1 1(1) (2) 1.5\u03c3 \u03c3= = , 2 2(1) (2) 8\u03c3 \u03c3= = . The tracking error is calculated as:\n( ) ( ) ( ); 1, 2kk k de n n n n\u03b8 \u03b8= \u2212 = and cost function is\ndefined as ( ) ( ) ( ); 0k k k Tc n e n e n= + \u039b \u039b = \u039b > .\n[ (1) (2)]k k k d d d\u03b8 \u03b8 \u03b8= is the desired trajectory vector .\nSimulation model of standard two link manipulator is given below [14].\n2 2 1\n2 2\ncos cos cos \u03b1 \u03b2 \u03b7 \u03b8 \u03b2 \u03b7 \u03b8 \u03b8 \u03b2 \u03b7 \u03b8 \u03b2 \u03b8 + + + +\n2 1 2 2 2\n2 1 2\n(2 )sin sin \u03b7 \u03b8 \u03b8 \u03b8 \u03b8 \u03b7\u03b8 \u03b8 \u2212 + +\n1 1 1 1 2\n1 1 2\ncos cos( ) cos( ) e e e \u03b1 \u03b8 \u03b7 \u03b8 \u03b8 \u03b7 \u03b8 \u03b8 + + + + 1\n2\n\u03c4 \u03c4 =\nwhere, 2 1 2 1( )m m a\u03b1 = + , 2 2 2m a\u03b2 =\n2 1 2m a a\u03b7 = , 1\n1\nge a=\nManipulator parameters are: link lengths, 1 2 1a a m= = and link masses: 1 2 1m m kg= = . Desired trajectory 1 sin( )desired t\u03b8 = ,\n2 cos( )desired t\u03b8 = System state space is continuous and has four state variables, i.e. 1 2 1 2( , , , )kx \u03b8 \u03b8 \u03b8 \u03b8= .The control task is to\napply a sequence of torques at each joint of robot manipulator so as to minimize the error between desired trajectory and response of robot manipulator. The action set for link 1 or torque for link 1 are chosen from 1 [ 200 0 200]u \u2208 \u2212 + Nm and that for\nlink 2 is 2 [ 50 0 50]u \u2208 \u2212 + Nm. Computational\ncomplexity of both FQC and Hybrid-FQC Lyapunov approaches is roughly the same. The physical system has been simulated for 20s using 4th order Runge-Kutta method with fixed time step of 0.02s. Exploration level is decayed from 0.4\n\u21920.02 over the iterations. The discount factor is set to 0.9 and learning-rate parameter is set to 0.05. We evaluate the proposed approach for the following two cases: a) Control versus external disturbances: controller handles external disturbance i.e., torques having a uniform distribution in \u00b120% around the torque values applied at each joint, b) Control versus external disturbances and varying payload mass: controller handles, in addition to external disturbances as in a), payload mass variations meaning objects having different mass are picked up and released by robotic arm. The mass m2 is varied as: (a) 4t < s, m2 = 1 kg, (b) 4 7t< \u2264 s, m2 = 2.5 kg, (c) 7 9t< \u2264 s, m2 = 1 kg, (d)9 12t< \u2264 s, m2 = 4 kg, (e)12 15t< \u2264 s, m2 = 1 kg, (f) 15 18t< \u2264 s, m2 = 2 kg, and (g)18 20t< \u2264 s, m2 = 1 kg.\nFigures 2 and 3 depict a comparative performance of the proposed Hybrid FQC-Lyapunov controller against the baseline fuzzy Q learning controller. As is evident from the results that proposed controller outperforms the fuzzy Q learning controller with lower tracking errors for both the links.\nFigures 4 and 5 show the control torque requirement for both controllers. It is seen that the", + "proposed hybrid control scheme leads to lower controller torque requirement at both links leading to better controller. The proposed controller exhibits a stable performance under disturbances and parameter variations.\nNext, we compare the controllers when only external disturbances are present. Figures 6, 7, 8, and 9 show comparative performance of the controllers .in terms of, tracking errors, and torque requirements at both manipulator links. It is seen that the proposed control scheme leads to lower tracking errors with less control effort at both the links.\nFig. 7. Tracking errors for link 2: Disturbance only\nFig. 8. Control torque for link 1: Disturbance only\nThe work presented in this paper showcases a novel hybrid Lyapunov theory based adaptive fuzzy reinforcement learning controller with guaranteed stability. The controller uses experiential information based mixing of the Lyapunov theory based stable control with the fuzzy Q learning control. We test the controller and compare its performance against the fuzzy Q learning controller on a robotic arm manipulator. The controller outperforms fuzzy Q controller with lower tracking errors and control torques validating the superiority of the proposed control approach. In future we would be extending the approach to other manipulators such as selective compliance assembly robotic arm (SCARA) and combining it with robust control approaches such as Markov games [4].\n0 5 10 15 20\n-0.4\n-0.2\n0\n0.2\n0.4\nTracking errors for theta1\nTime [sec]\nEr ro\nr [ ra\nd]\nFQC (red) Hybrid FQC-Lyapunov (green)\n0 5 10 15 20 -500\n-300\n-100\n100\n300\n500 Control torque for link 1\nTime [sec]\nTo rq\nue [N\nm ]\nFQC (red) Hybrid FQC-Lyapunov (green)" + ] + }, + { + "image_filename": "designv11_84_0003007_s00542-010-1211-9-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003007_s00542-010-1211-9-Figure4-1.png", + "caption": "Fig. 4 Finite element model and pressure distribution of the FDBs", + "texts": [ + " In addition, the critical mass determined from the equations of motion for the rotating coordinates in Eq. (17) is the same as that determined from the equations of motion for the inertia coordinates in Eq. (14) because the displacement x in Eq. (16) is dependent upon u. Table 1 shows the major design specifications of the FDBs in a 2.5-inch HDD. The FDBs consist of two rotating grooved journal bearings, four plain journal bearings, two rotating grooved thrust bearings, and one plain thrust bearing. Figure 4 shows the finite element model and the pressure distribution of the coupled journal and thrust bearings. The fluid film was discretized using 5,640, four-node, isoparametric bilinear elements, and the Reynolds boundary condition was applied to guarantee the continuity of pressure and the pressure gradient. In this research, the disk-spindle system rotating at 5,400 rpm was assumed to have an equilibrium flying height of 7.85 lm, where the axial load generated by the FDBs was equal to the weight of a rotor" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002301_s0140525x13000617-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002301_s0140525x13000617-Figure8-1.png", + "caption": "Figure 8. The vestibular apparatus in the human brain. (a) Diagram showing the relationship of the vestibular apparatus to the external ear and skull. (b) Close-up of the vestibular organ showing the detectors for linear acceleration (the otolith organs \u2013 comprising the utricle and saccule) and the detectors for angular acceleration (the semicircular canals, one in each plane). (Taken from: http://www.nasa.gov/ audience/forstudents/9-12/features/F_Human_Vestibular_System_in_Space.html) Source: NASA.", + "texts": [ + " Detection of the vertical depends on integration of vestibular cues, together with those from the visual world and also from proprioceptive cues to head and body alignment (Angelaki et al. 1999; Merfeld et al. 1993; Merfeld & Zupan 2002). The vestibular apparatus is critical for the processing not only of gravity but also of movement-related inertial three- 534 BEHAVIORAL AND BRAIN SCIENCES (2013) 36:5 dimensional spatial cues. The relationship of the components of the vestibular apparatus is shown in Figure 8a, while the close-up in Figure 8b shows these core components: the otolith organs, of which there are two \u2013 the utricle and the saccule \u2013 and the semicircular canals. The otolith organs detect linear acceleration (including gravity, which is a form of acceleration) by means of specialized sensory epithelium. In the utricle the epithelial layer is oriented approximately horizontally and primarily detects earth-horizontal acceleration, and in the saccule it is oriented approximately vertically and primarily detects vertical acceleration/gravity" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.24-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.24-1.png", + "caption": "Fig. 2.24. Spherical wrist", + "texts": [ + "52) are for the single joints: A0 1(\u03d11) = \u23a1 \u23a2\u23a3 c1 0 s1 0 s1 0 \u2212c1 0 0 1 0 0 0 0 0 1 \u23a4 \u23a5\u23a6 Ai\u22121 i (\u03d1i) = \u23a1 \u23a2\u23a3 ci \u2212si 0 aici si ci 0 aisi 0 0 1 0 0 0 0 1 \u23a4 \u23a5\u23a6 i = 2, 3. Computation of the direct kinematics function as in (2.50) yields T 0 3(q) = A0 1A 1 2A 2 3 = \u23a1 \u23a2\u23a3 c1c23 \u2212c1s23 s1 c1(a2c2 + a3c23) s1c23 \u2212s1s23 \u2212c1 s1(a2c2 + a3c23) s23 c23 0 a2s2 + a3s23 0 0 0 1 \u23a4 \u23a5\u23a6 (2.66) where q = [\u03d11 \u03d12 \u03d13 ]T . Since z3 is aligned with z2, Frame 3 does not coincide with a possible end-effector frame as in Fig. 2.13, and a proper constant transformation would be needed. Consider a particular type of structure consisting just of the wrist of Fig. 2.24. Joint variables were numbered progressively starting from 4, since such a wrist is typically thought of as mounted on a three-DOF arm of a six-DOF manipulator. It is worth noticing that the wrist is spherical since all revolute axes intersect at a single point. Once z3, z4, z5 have been established, and x3 has been chosen, there is an indeterminacy on the directions of x4 and x5. With reference to the frames indicated in Fig. 2.24, the DH parameters are specified in Table 2.5. The homogeneous transformation matrices defined in (2.52) are for the single joints: A3 4(\u03d14) = \u23a1 \u23a2\u23a3 c4 0 \u2212s4 0 s4 0 c4 0 0 \u22121 0 0 0 0 0 1 \u23a4 \u23a5\u23a6 A4 5(\u03d15) = \u23a1 \u23a2\u23a3 c5 0 s5 0 s5 0 \u2212c5 0 0 1 0 0 0 0 0 1 \u23a4 \u23a5\u23a6 A5 6(\u03d16) = \u23a1 \u23a2\u23a3 c6 \u2212s6 0 0 s6 c6 0 0 0 0 1 d6 0 0 0 1 \u23a4 \u23a5\u23a6 . Computation of the direct kinematics function as in (2.50) yields T 3 6(q) = A3 4A 4 5A 5 6 = \u23a1 \u23a2\u23a3 c4c5c6 \u2212 s4s6 \u2212c4c5s6 \u2212 s4c6 c4s5 c4s5d6 s4c5c6 + c4s6 \u2212s4c5s6 + c4c6 s4s5 s4s5d6 \u2212s5c6 s5s6 c5 c5d6 0 0 0 1 \u23a4 \u23a5\u23a6 (2", + " A comparison of the vector p0 6 in (2.68) with the vector p0 3 in (2.65) relative to the sole spherical arm reveals the presence of additional contributions due to the choice of the origin of the end-effector frame at a distance d6 from the origin of Frame 3 along the direction of a0 6. In other words, if it were d6 = 0, the position vector would be the same. This feature is of fundamental importance for the solution of the inverse kinematics for this manipulator, as will be seen later. A comparison between Fig. 2.23 and Fig. 2.24 reveals that the direct kinematics function cannot be obtained by multiplying the transformation matrices T 0 3 and T 3 6, since Frame 3 of the anthropomorphic arm cannot coincide with Frame 3 of the spherical wrist. Direct kinematics of the entire structure can be obtained in two ways. One consists of interposing a constant transformation matrix between T 0 3 and T 3 6 which allows the alignment of the two frames. The other refers to the Denavit\u2013Hartenberg operating procedure with the frame assignment for the entire structure illustrated in Fig", + "33: shoulder\u2013right/elbow\u2013up, shoulder\u2013left/elbow\u2013 up, shoulder\u2013right/elbow\u2013down, shoulder\u2013left/elbow\u2013down; obviously, the forearm orientation is different for the two pairs of solutions. Notice finally how it is possible to find the solutions only if at least pWx = 0 or pWy = 0. In the case pWx = pWy = 0, an infinity of solutions is obtained, since it is possible to determine the joint variables \u03d12 and \u03d13 independently of the value of \u03d11; in the following, it will be seen that the arm in such configuration is kinematically singular (see Problem 2.18). Consider the spherical wrist shown in Fig. 2.24, whose direct kinematics was given in (2.67). It is desired to find the joint variables \u03d14, \u03d15, \u03d16 corresponding to a given end-effector orientation R3 6. As previously pointed out, these angles constitute a set of Euler angles ZYZ with respect to Frame 3. Hence, having computed the rotation matrix R3 6 = \u23a1 \u23a3 n3 x s3x a3 x n3 y s3y a3 y n3 z s3z a3 z \u23a4 \u23a6 , from its expression in terms of the joint variables in (2.67), it is possible to compute the solutions directly as in (2.19), (2.20), i.e., \u03d14 = Atan2(a3 y, a 3 x) \u03d15 = Atan2 (\u221a (a3 x)2 + (a3 y)2, a 3 z ) (2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003225_msf.697-698.213-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003225_msf.697-698.213-Figure2-1.png", + "caption": "Fig. 2 Middle cutter tooth", + "texts": [ + " ( ) ( ) ( ) ( ) ( )2 m sF cos - sin tan \u03b2 - \u03b3 + F cos \u03b2 - \u03b3 R a = cos - \u03b3 -\u00b5sin - \u03b3 cos\u03b2 \u03a6 \u03a6 \u2022 \u03a6 \u03a6 (5) So component force perpendicular to cutting speed direction can be got. ( ) ( ) n sin \u03b2 - \u03b3 F = R a cos\u03b2 (6) Then decomposing the results above into X Y Z directions, one can calculate forces in all directions of the inside blade. x1 nF = F cos\u03b8\u2022 (7) y1 nF = F sin\u03b8\u2022 (8) z1 cF = F (9) The outside blade is similar to that one we mentioned above. Basically, Fx, Fy, Fz calculation method is the same as the above one. So there is no need to go into details. The Calculation of Cutting Force of Middle Cutter Tooth. As shown in Fig 2, The cutting edge of middle cutter tooth is a line of arc cutting edge, so it belongs to no free cutting process. The natural scraping directions of arbitrary arc section on blade are its normal directions, but actually along the cutter axis direction should be right [2-3]. Little arc in ds blade can be seen as free cutting process of straight blade whose cutting force can be calculated from edge cutting method inside. Section of Shear \uff1a c 1 s A f sina A = = ds sin\u03a6 sin\u03a6 \u2032 \u2032 \u2032 \u2032 (10) In the Eq.10, ds is cutting width, zzf sin\u03b1 is cutting depth, \u03b3 \u2032 is rake of middle cutter tooth" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003424_iecon.2011.6119317-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003424_iecon.2011.6119317-Figure4-1.png", + "caption": "Fig. 4. Camera\u2019s Field of view", + "texts": [ + " In a TOF camera unit [16], a modulated light pulse is transmitted by the illumination source and the target distance is measured from the time taken by the pulse to reflect from the target and back to the receiving unit. The coordinates of the obstacle with respect to the PMD are obtained as a 48 by 64 matrix, each element corresponding to a pixel. The specifications of the camera are provided in Table I. The contour of the obstacle can be extracted from its range information by data processing and filtering. The camera also provides the 3D orientation (X, Y, Z) of the obstacles with respect to the centre of the camera [17]. As described in the Fig. 4, (X, Y, Z) are the three dimensional coordinates in which X represents horizontal plane, Y represents vertical plane, Z represents the plane along the optical axis. The origin (0, 0, 0) is at intersection of the optical axis with the front face of the camera. Consider an obstacle moving from a point towards the camera. The initial position is sensed by the camera in the image frame 1 at a distance D1 and its 3D coordinates (X1, Y1, Z1) . After time t, the camera senses the obstacle position in the image frame at a distance D2 with its 3D coordinates (X2, Y2, Z2)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001528_icphm.2015.7245019-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001528_icphm.2015.7245019-Figure2-1.png", + "caption": "Figure 2. Seeded bearing defects used in this study", + "texts": [ + " 1 shows a machinery fault simulator developed to conduct research on incipient bearing fault diagnosis and a cylindrical roller bearing (i.e., FAG NJ206-E-TVP2) used for experiments in this study. To continuously capture AE signals, a general-purpose AE sensor (type R3a from Physical Acoustics Corporation) was used and 78 one-second AE signals sampled at 1 MHz were obtained for each bearing condition: a defect-free bearing (DFB), a bearing with a crack on its outer raceway (BCOR), a bearing with a crack on its inner raceway (BCIR), and a bearing with a crack on its roller (BCR). Fig. 2 illustrates the various seeded bearing defects used in this study, and Table I presents a detailed description of the bearing failures. As mentioned in Section I, time-frequency analysis is needed to understand non-stationary AE signals. Among various time-frequency analysis tools, such as wavelet transform [12], [28]\u2013[31], empirical mode composition [32]\u2013 [36], and Hilbert-Huang transform [37]\u2013[39], the comprehensive fault diagnosis method uses DWPT for its ability to split an input AE signal into low- and high-frequency sub-band components" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000869_s00170-021-07749-1-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000869_s00170-021-07749-1-Figure8-1.png", + "caption": "Fig. 8 (a) A schematic diagram showing the sampling point of the test specimen and its sketch with dimensions; (b) the tensile stressstrain curves for each specimen, (c) produced test specimens before the test", + "texts": [ + " Higher values of nanohardness in the area of inner bending arc in comparison to outer bending arc are the consequence of high compressive stresses and higher plastic deformation in this area. Higher plastic deformation is also proven by measurement of the Inconel layer of original vs bent pipe. The layer thickness was reduced maximally by about 25% in the outer arc but enlarged in the inner arc by maximally about 55% after bending. The test consists in deforming the test specimen by a tensile load into the fracture, in order to determine the mechanical properties of the material of which the test specimen was made. For the tensile test, a total of six test specimens (shown in Fig. 8c) were produced according to the ISO 6892-1 standard, respecting Annex E, and heads have been flattened for gripping in the testing machine [39], which were taken from individual zones of Pipe 1 according to Fig. 8a. It was one specimen from the cladding zone of Inconel 625, then three specimens from the heat-affected zone between the materials and one specimen from the zone of the base material of the 16Mo3 pipe. Another test specimen contained all the mentioned zones of the pipe (Inconel cladding, HAZ and the base pipe of 16Mo3) and was hereinafter referred to as Pipe. These bodies were manufactured using a RobocutAlpha-0iD wire EDMmachine, all of which were 1.5 mm thick, and only the last body containing all areas was 6.1 mm thick. From the obtained tensile test results shown in Fig. 8b and in Table 3, it can be seen that the values of the yield strength are 460 MPa and the values of the tensile strength are 594 MPa of the tested sample from the zone of the base material of the 16Mo3 pipe that reaches higher values than in the material sheet [40]. In contrast, the determined value of elongation is approximately one-third compared to the value stated in the material sheet. The difference between the values obtained by the tensile test and the values given by the material sheet can be caused, for example, by different heat treatment of the material before the tensile test" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002060_j.automatica.2014.10.063-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002060_j.automatica.2014.10.063-Figure1-1.png", + "caption": "Fig. 1. Forces acting on the aircraft.", + "texts": [ + " The change of coordinates z = \u03d5\u0304(x\u0304) , z1 = x\u03041 + \u221e s=1 (\u22121)sx\u0304s4 s! Ls\u22121 g\u03042 (g\u030421)(x\u0304) = x\u03041 \u2212 x\u03044(\u22122x\u03044)\u2212 x\u030424 = x\u03041 + x\u030424 z3 = x\u03043 + \u221e s=1 (\u22121)sx\u0304s4 s! Ls\u22121 g\u03042 (g\u030423)(x\u0304) = x\u03043 \u2212 x\u03041x\u03044 \u2212 x\u030434 zj = x\u0304j elsewhere rectifies the vector fields and composition z = \u03d5\u0303 \u25e6 \u03d5(x) z = \u03d5\u0303 \u25e6 \u03d5(x) , z1 = x1 \u2212 x5 + x24 z2 = x2 + 2x4x5 z3 = x3 \u2212 x1x4 + x4x5 z4 = x4 z5 = x5 linearizes the original system. Example 3.2. The dynamics of the VTOL aircraft is modeled as (Serrani, Isidori, Byrnes, & Marconi, 2000) (Fig. 1). \u03be\u03081 = \u2212 sin(\u03b8) T M + cos(\u03b8) 2 sin\u03b1 M F \u03be\u03082 = \u2212 cos(\u03b8) T M + sin(\u03b8) 2 sin\u03b1 M F \u2212 g \u03b8\u0308 = 2l J cos\u03b1F (3.1) where M , J , 2l and g denote the mass, moment of inertia, distance between wingtips and gravitational acceleration; T denotes the thrust, F the torque, (\u03be1, \u03be2) the center mass, and \u03b8 the rolling angle. Let x1 = \u03be1, x2 = \u03be\u03071, x3 = \u03b8, x4 = \u03b8\u0307 , x5 = \u03be2, x6 = \u03be\u03072. The system rewrites in the form \u03a3 : x\u0307 = f (x)+ g1(x)u1 + g2(x)u2, x = (x1, . . . , x6) \u2208 R6 with f (x) = (x2, g tan x3, x4, 0, x6, 0)\u22a4, g1(x) = (0, \u03b7(x3), 0, 1, 0, 0)\u22a4 and g2(x) = (0, tan x3, 0, 0, 0, 1)\u22a4, where \u03b7(x3) = J tan\u03b1 Ml cos2 x3\u2212sin2 x3 cos x3 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.3-1.png", + "caption": "Fig. 7.3 The planar five-bar mechanism (the gray pairs denote the actuated joints)", + "texts": [ + " The solutions of the IGM are at the intersections of those two configuration loci (Fig. 7.2c). It must be mentioned that, in usual cases, the obtained configuration loci Cij are defined by algebraic equations. Therefore, for having an idea of the maximal number of intersection points, usual methods (such as the B\u00e9zout bounds B\u00e9zout 1764) can be used. In the next sectionswe present the IGM of somePKM that are solved by analytical methods and/or by geometrical approaches. The planar five-bar mechanism (Fig. 7.3) is a 2 DOF parallel robot able to achieve two translations in the plane (O, x0, y0) (see AppendixA) and which is composed of two legs: \u2022 A leg composed of 3 R joints whose axes are parallel, directed along z0 and located at points A11, A12 and A13, the joint located at point A11 being actuated, and \u2022 A leg composed of 2 R joints whose axes are parallel, directed along z0 and located at points A21 and A22, the joint located at point A21 being actuated. All other joints are passive. Thus, the vector of actuated coordinates is qT a = [q11 q21]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002480_2015-01-0680-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002480_2015-01-0680-Figure7-1.png", + "caption": "Figure 7. Schematic of load position detection system", + "texts": [ + " This is accomplished by a metal target which is attached to the rotation wheel. When it is close to the proximity switch, the proximity switch will be triggered to a high voltage output. When the metal moves away from the proximity switch, the proximity switch will return to low voltage. The duration of high voltage will depend on the width of the metal target. From the output signal (low voltage triggered to high voltage), the rotation angle of the connecting rod at which the normal load is applied is detected, as seen in Figure 7. With a flip-flop, the frequency of the modified position signal is decreased by half of the original position signal. The modified position signal is combined through an AND logic calculation with the original signal. This generates the load signal satisfying the frequency requirement (Fig. 8). The experimental position and loading signal is shown in Figure 9. Because the PZT (Piezoelectric Transducer) actuator needs 100 ms to react from its original position to maximum displacement, the width of the high level loading signal which triggers the actuator should be at least 100 ms" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001979_amm.284-287.461-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001979_amm.284-287.461-Figure2-1.png", + "caption": "Fig. 2 Typical rotor configuration and coordinates", + "texts": [ + "[15] developed a new dynamic model for the gear set that the pressure angle and the contact ratio as time-varying variables. The configuration of a geared rotor-bearing system is shown in Fig. 1. Two uniform flexible shafts are of length 1L and 2L , and the gear pair is mounted on the shafts. An external torque M exerts upon the driving gear. The contacting mesh force is represented by the gear mesh stiffness mk and damping mc along the pressure line. Four bearings are modeled as flexible elements with damping and stiffness denoted as b jc and b jk . A single shaft system with a rigid disk is shown in Fig. 2, in All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 142.103.160.110, University of British Columbia, Kelowna, Canada-14/07/15,06:35:22) which the fixed reference frame, ZYX , is used to describe the system motion. Five degrees of freedom , , , , WV are considered at each nodal point of the shaft, where V and W are lateral displacements along Y and Z directions, respectively, and are rotational displacements, is the torsional displacement, and the axial translational vibration is neglected" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001136_978-3-319-19788-3_1-Figure1.14-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001136_978-3-319-19788-3_1-Figure1.14-1.png", + "caption": "Fig. 1.14 Examples of robots with 2 DOF able to position a device with constant orientation in a plane: Robot PacDrive Delta 2 (courtesy of Schneider Electrics)", + "texts": [ + " Methods to compute the mobility of mechanisms are given in Appendix A. Many PKM have been designed in order to be able to move their platform in a plane. We call them the Planar Parallel Manipulators (PPM). We can classify them into three main groups: 1. robots with 2 DOF able to position a point in a plane (Fig. 1.13), 2. robots with 2 DOF able to position a device with constant orientation in a plane (two translational DOF in the plane and one constrained (constant) platform orientation around the axis normal to the plane\u2014Fig. 1.14), 3. robots with 3 DOF able to position a device in a plane (two translational DOF in the plane and one rotationalDOF around the axis normal to the plane\u2014Fig.1.15). There obviously exist other types of possible mobilities (1T1R), but they are not common. Most of the robots of this category are planar, i.e. all their elements are constrained to move in parallel planes. However, in order to increase the stiffness of robots with 1Throughout this book, when we mention the number and types of DOF of the PKM, we refer to the number and types of DOF of its mobile platform" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003039_icmech.2013.6518554-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003039_icmech.2013.6518554-Figure6-1.png", + "caption": "Fig. 6. The optimization process of optical mouse placement: (a) the second optical mouse at somewhere in the second quadrant and (b) the second optical mouse at the left end of the major principal axis.", + "texts": [ + " In the case of cm , any regular triangular array of three optical mice is found to be optimal, for which their geometrical center is at the origin( ). On the other hand, in both cases of cm and cm , three optical mice which are on the major principal axis( axis for cm ; axis for cm ) with two optical mice at one end and one optical mouse at the other end is found to be optimal, for which their geometrical centers do not coincide with the origin( & for cm ; & for cm ). For the case of cm , Fig. 6 shows an intermediate snapshot during the optimization process of optical mouse placement, where the first and third optical mice are at the right and left ends of the major principal axis( axis), respectively, and the second optical mouse is at somewhere on the elliptical path belonging to the second quadrant. If the second optical mouse is moved downward to the left along the elliptical path, then the condition of p pp , given by (32), can be satisfied so that the value of the performance index increases. This is shown in Fig. 6(a). By repeating the same pattern of positional changes, the second optical mouse can reach the left end of the major principal axis( axis). As shown in Fig. 6(b), no further positional change of the second optical mouse satisfying (33) is possible, which implies that the current optical mouse placement is optimal. VI. CONCLUSION In this paper, we presented the optimal optical mouse placement of optical mice for the velocity estimation of a mobile robot. It was assumed that there can be some restriction on the installation of optical mice at the bottom of a mobile robot. The main contributions of this paper can be summarized as 1) the velocity kinematics of a mobile robot with an array of optical mice, 2) the error characteristics of the mobile robot velocity estimation, and the performance index for the optimal optical mouse placement, 3) the global and local optimization strategies in terms of the positions of optical mice and their positional changes, and 4) the optimal placements of three optical mice within a given elliptical region" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.30-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.30-1.png", + "caption": "Fig. 2.30. Workspace of a two-link planar arm", + "texts": [ + " Along the segments ab, bc, cd, ae, ef , fd a loss of mobility occurs due to a joint limit; a loss of mobility occurs also along the segment ad because the arm and forearm are aligned.12 Further, a change of the arm posture occurs at points a and d: for q2 > 0 the elbow-down posture is obtained, while for q2 < 0 the arm is in the elbow-up posture. In the plane of the arm, start drawing the arm in configuration A corresponding to q1m and q2 = 0 (a); then, the segment ab describing motion from q2 = 0 to q2M generates the arc AB; the subsequent arcs BC, CD, DA, AE, EF , FD are generated in a similar way (Fig. 2.30). The external contour of the area CDAEFHC delimits the requested workspace. Further, the area BCDAB is relative to elbowdown postures while the area DAEFD is relative to elbow-up postures; hence, the points in the area BADHB are reachable by the end-effector with both postures. kinematic singularity of the arm. In a real manipulator, for a given set of joint variables, the actual values of the operational space variables deviate from those computed via direct kinematics. The direct kinematics equation has indeed a dependence from the DH parameters which is not explicit in (2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003102_gt2014-26275-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003102_gt2014-26275-Figure6-1.png", + "caption": "Figure 6. Rotor Eccentricity Inside the AMB", + "texts": [ + " Active magnetic bearings are a highly suitable apparatus for exciting the rotor in the system identification process. Careful attention is required to maintain concentricity of the rotor and stator as part of the AMB installation process on the machine, though some deviation cannot be avoided. As the machine rotates, the location of the center of the shaft will change due to rotor vibration. Since AMB forces are a function of rotor position, this affects the circularity of the force profile. The rotor eccentricity is shown in Fig. 6. The effect of rotor eccentricity in the magnetic bearings on identified parameters is investigated by introducing an elliptical force profile into the simulation. In this paper, the amplitude of sine sweep force in the x and y directions is 540N and 460N, respectively. Also, 500N is used to calculate the Frequency Response Function (FRF). The waterfall spectrum diagram of the predicted bearing vibration data at node 10 under forward and backward elliptical force excitation is shown in Fig. 7. It can be seen that the forward elliptical sine sweep excitation will excite rotor backward response, especially near the natural frequency" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001126_2015-01-2515-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001126_2015-01-2515-Figure10-1.png", + "caption": "Figure 10. Horizontal inject injector in purge position", + "texts": [ + " Another challenge of feeding different diameters of fastener with the same hardware is pushing the fastener the proper distance into the feed nose. To make sure that the proper diameter fastener gets pushed the proper distance, the tube end fittings double as a hard stop for the pusher\u2019s up position. When using the tube dedicated to a certain diameter, the pusher\u2019s forward stroke is set by that tube\u2019s end fitting, ensuring that the proper hard stop is used for the proper fastener. This can be seen by looking closely at figure 8. Horizontal Inject Version The parallel gripper injector is implemented in vertical axis riveting machines. In Figure 10 you can see that the designer chose a rotary version of the parallel gripper to make the system more compact. You can see where a replaceable urethane catcher's mitt is incorporated into the track. This feature is easily replaced. Multiple grips and diameters can be fed down the same injector. Due to the synchronized motion the fastener is guided to the center of the fingers. By reversing the air as shown in Figure 10 the guide chutes are opened wide to allow any fasteners or FOD to fall free of the machine and be captured by a reject removal component. This new style of rivet injector is in production use on a variety of fastening machines used by major aircraft manufacturers. We have implemented the parallel gripper in both vertical axis and horizontal axis riveting applications. It is equally effective in both orientations. We have implemented the parallel gripper rivet injector on headed rivets, threaded bolts, ribbed swage bolts and unheaded (slug) rivets" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001979_amm.284-287.461-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001979_amm.284-287.461-Figure1-1.png", + "caption": "Fig. 1 The configuration of a geared rotor-bearing system", + "texts": [ + " Shiau et al.[12] analyzed the lateral response due to torsional excitation of geared rotor, and Lee et al.[13] developed the coupling of lateral and torsional vibration for the geared rotor-bearing system. The effects of the residual shaft bow and viscoelastic supports were investigated by Kang et al. [14] Kim et al.[15] developed a new dynamic model for the gear set that the pressure angle and the contact ratio as time-varying variables. The configuration of a geared rotor-bearing system is shown in Fig. 1. Two uniform flexible shafts are of length 1L and 2L , and the gear pair is mounted on the shafts. An external torque M exerts upon the driving gear. The contacting mesh force is represented by the gear mesh stiffness mk and damping mc along the pressure line. Four bearings are modeled as flexible elements with damping and stiffness denoted as b jc and b jk . A single shaft system with a rigid disk is shown in Fig. 2, in All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www", + "110, University of British Columbia, Kelowna, Canada-14/07/15,06:35:22) which the fixed reference frame, ZYX , is used to describe the system motion. Five degrees of freedom , , , , WV are considered at each nodal point of the shaft, where V and W are lateral displacements along Y and Z directions, respectively, and are rotational displacements, is the torsional displacement, and the axial translational vibration is neglected. Fig. 3 shows the configuration and generalized coordinates for the pair of gears [15], which are mounted on the shafts as shown in Fig.1. The teeth are regarded as flexible cantilever beams and are deformed by both the bending and shear. The mass, transverse mass moment of inertia, polar mass moment of inertia of the driving gear and driven gear are, respectively, 1dm , 1dDI , 1dpI , and 2dm , 2dDI , 2dpI . The motion of the gear set can be defined with ten generalized coordinates, which are the same as the node coordinates of two shafts, where the gears are mounted. The dashed and solid circles represent the gear pair before and after motion, respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure9.15-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure9.15-1.png", + "caption": "Fig. 9.15 Graphic representation of the type C LCM, a intervertebral space of the apical region, pedicle of vertebral arch. b Apical vertebra [23]. Reprinted with permission from SpringerNature publishers", + "texts": [ + " For these parameters, all the changes that could appear in the lumbar region due to scoliosis are taken into account and three different types of modifiers could be applied: \u2022 A (slight). \u2022 B (moderate). \u2022 C (severe). Taking the center of the sacrum as reference, a straight line is drawn up to the highest point of the column\u2019s image (from an anterior\u2013posterior point of view). \u2022 Type A lumbar column\u2019s modifier (Fig. 9.13.) \u2022 Type B lumbar column\u2019s modifier (Fig. 9.14.) 236 J. A. Beltr\u00e1n-Fern\u00e1ndez et al. Fig. 9.13 Graphic representation of the type A LCM [22]. Reprinted with permission from SpringerNature publishers \u2022 Type C lumbar column\u2019s modifier (Fig. 9.15.) This parameter is defined by the hump\u2019s extension. The values are expressed in \u201c\u2212, N, +\u201d [21]: \u2212 kyphosis between T5 and T12 minor to 10\u00b0. N kyphosis between T5 and T12 between 10\u00b0 and 40\u00b0. + kyphosis between T5 and T12 mayor to 40\u00b0. 9 Design of an Auxiliary Mechanical System for the Diagnosis \u2026 237 Fig. 9.14 Graphic representation of the type B LCM [23]. Reprinted with permission from SpringerNature publishers With the support of the Germ\u00e1n D\u00edaz Lombardo Hospital, the research team could have access to the clinical history of the patients and their archives, including radiographies, which were used to classify those patients who suffer from deformities of interest for the project" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003912_detc2013-12420-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003912_detc2013-12420-Figure1-1.png", + "caption": "FIGURE 1. INITIAL CONFIGURATION FOR A THREE-DOF NEWTON\u2019S CRADLE.", + "texts": [ + " It is also assumed that the impact event occurs over a very short time period in which the position and orientation of the system remains constant, established in [6]. This type of discrete approach is used here. The first key issue which arises when applying the discrete approach to Newton\u2019s cradle is that the simultaneous, multiple point impact problem is indeterminate with respect to the impact forces. This means that the rigid body model does not include enough equations to solve for the unknown forces. The indeterminacy is problematic because the impact forces dictate the post-impact behavior of the system. For example, consider the Newton\u2019s cradle in Fig. 1 with three balls labeled A, B, and C, hanging from massless strings of length L which are separated by a distance W . Note that there are eight unknown impact forces, f1 through f8 in Fig. 1, but only three rotational generalized coordinates, q1, q2, and q3. The consideration of frictional impact introduces the tangential forces, f2, f4, f6, and f8 which are not considered in other works that assume frictionless impacts [7\u201310]. The three generalized coordinates imply three 1 Copyright c\u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/asmep/77589/ on 05/02/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use degrees-of-freedom (DOFs) and three equations of motion, M q\u03081 q\u03082 q\u03083 \ufe38 \ufe37\ufe37 \ufe38 q\u0308 + g(q) = \u03b3(q) = JT f1 ", + " Works in robotic grasping eliminate infeasible force solutions based on the situationspecific contact geometry of a grasped object to solve the inde- terminate equations [13]. In contrast to these methods, this work strictly adheres to the assumptions of rigid body modeling in conjunction with the notion that the configuration of the system does not change in the short time span of the collision. These assumptions imply that the Jacobian in (1) is constant during the collision, which enforces a kinematic relationship between the impact points, corresponding to those shown in Fig. 1. However, when considering rigid body systems, there exists a dual relationship between the forces and component velocities at the impact points, \u03d1 = v1 ... v8 = J q\u03071 q\u03072 q\u03073 = J q\u0307 (2) which also enforces a constraint between velocities at the impact points that must satisfy rigid body modeling assumptions [14]. The second key issue in this work concerns addressing the elasticity present between the spheres. Capturing the elasticity between spheres allows prediction of energy dissipation in the simultaneous impact", + " The following section will present the application of the two key elements of the proposed discrete approach to treat indeterminate collisions. Analysis of a three-ball Newton\u2019s cradle with three-DOF, and the results from the proposed approach will be discussed. Then, a six-DOF Newton\u2019s cradle is discussed and compared to theoretical and experimental results found in other works. A similar comparison of results will follow. The paper will end with a summary of the major conclusions drawn from the analysis of Newton\u2019s cradle and effectiveness of the proposed framework. The three-DOF Newton\u2019s cradle presented in Fig. 1 is considered as a benchmark case to demonstrate the application of the proposed approach. In this study, Ball A is released from rest at an arbitrary height and impacts Ball B with initial velocity q\u0307\u22121 , or v\u22121 in terms of its normal component velocity. Balls B and C are initially at rest and in contact, q\u0307\u22122 = q\u0307\u22123 = 0. The balls hang from massless strings of length L and each string is separated by distance W , where W = 2R. Newton\u2019s third law allows for the definition of relations among the impact forces, such that the total number of forces considered in Fig. 1 are reduced. f3 = f1 f4 = f2 f7 = f5 f8 = f6 (4) These expressions are composed into a matrix z and used to reduce F as, F = z F\u2217 , F\u2217 = (zT z)\u22121zT F (5) A definite integration of the equations of motion (1) over a very short time \u03b5 , which is infinitesimally small for the impact event, \u222b t+\u03b5 t (M q\u0308+g (q)) dt = \u222b t+\u03b5 t JT (q) z F\u2217 dt (6) yields the impulses in p\u2217 as, M ( q\u0307(t + \u03b5)\u2212 q\u0307(t) ) = JT z p\u2217 = JT z p1 p2 p5 p6 \u2261 JT z pn1 pt1 pn2 pt2 (7) A velocity constraint consistent with rigid body assumptions is used, as in [14], for Ball B to define a relationship between the component velocities of the two impact points", + " An event-driven scheme stops the simulation when an impact is detected. Multiple impact events can occur in the collisions simulated. The proposed approach is used to treat the impact events and determine the post-impact velocities of the system. These velocities serve as the initial conditions when the simulation is restarted. This technique is followed herein each time an impact event is detected in a simulation. Case I: Three-DOF with e\u2217 = 1 and \u00b51/\u00b52 = 1 The first case considered for the three-DOF system in Fig. 1a examines balls with equal, unit mass mA = mB = mC = 1 kg, e\u2217 = 1 and \u00b51 = \u00b52 = 0.8. The motion obtained from simulating this case is shown in Fig. 4. This figure only shows the results of the first collision, which triggers a single impact event. This impact event is indeterminate with respect to the impact forces. Ball A is released from rest, swings down, and impacts Balls B and C which are at rest and in contact. Ball A has pre-impact velocity v\u22121 = 1.304 m/s. All of the momentum of Ball A is transferred to Ball C, such that Balls A and B are motionless and in contact after the collision", + " However, the solution demonstrates that the effect of differing friction properties between pairs of balls in the chain affects the motion obtained. It is possible to adjust the friction coefficients to match results obtained in experiment. This is not explored further in this work. Case III: Six-DOF with e\u2217 = 1 and \u00b51/\u00b52 = 1 A six-DOF Newton\u2019s cradle is now considered as shown in Fig. 10 using the proposed approach. This system is composed of three balls as presented for the three-DOF case in Fig. 1 but has six-DOFs. The three additional DOFs are included in the model to allow each ball to rotate freely with respect to the massless string it hangs from. Similar to the three-DOF cases, Ball A is released from rest at an arbitrary height and impacts Balls B and C, which are initially at rest and in contact. The first case examined for this model considers balls with 6 Copyright c\u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/asmep/77589/ on 05/02/2017 Terms of Use: http://www" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002236_icma.2011.5985628-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002236_icma.2011.5985628-Figure1-1.png", + "caption": "Fig. 1. Model of environment-contacting task with free-joint structure", + "texts": [ + " In this paper, first, we show a model of kinematics of this operation, and some conditions to judge whether each operation is able to achieved or has danger not to achieve the operation. With this model, we develop a new method to analyze these conditions with linear programming method. With this method, we can easily judge conditions whether an environment-contact task is able to achieved or not. We show some examples of analysis of this manipulation possibility. II. MODEL OF POSITION- CONTROLLED OBJECT WITH CONTACT WITH ENVIRONMENT Like as shown in Fig. 1, we introduce a free-joint structure at wrist part of a manipulator and analyze its geometric and kinematic models. When position-controlled manipulators cooperatively ma nipulate an object, there exists rigid closed loop structure and it may cause excessive contact force. Osumi introduced a method to avoid excessive contact force by introduction of passive joint into the closed loop structure[7], [8]. By regarding contact between an object and environment as pseudo link structure, this research aims to avoid excessive contact force with same discussion. A free joint mechanism will be added at wrist. With this free joint mechanism and pseudo passive joint in contact, a manipulator and environment manipulate an object cooperatively. We consider a model as shown in Fig. 1. We set ef as a d.o.f. of the free motion space by the contact between the object and the environment and fJ as a d.o.f. of the free motion space by the mechanism. Conditions to avoid excessive contact force but to support an object rigidly in closed loop structure as \"environment - object - free joint - manipulator (=environment)\" are as follows. First, it is enough to avoid excessive contact force that rank of free motion space by free joint and pseudo joint is full (3 in 2-D case and 6 in 3-D)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003595_physreve.84.011701-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003595_physreve.84.011701-Figure1-1.png", + "caption": "FIG. 1. (Color online) Structure of the smectic layers in the fiber, showing the coordinate system (R,\u03c6,z). The external radius of the fiber is Rf and the radius of its inner defect core is Rc. The width L of each layer depends on the dimensions L1,L2, and L3 of a single mesogen as well as the molecular orientation: \u03b8 is the angle between the director n and the radial direction; \u03b1 is the angle between p and the z axis.", + "texts": [ + " Finally, we comment on extensions of this approach to the case where the molecular-dipole orientation \u03b1 is not constant as well as possible applications of our results. 011701-11539-3755/2011/84(1)/011701(8) \u00a92011 American Physical Society PE\u0301REZ-ORTIZ, GUZMA\u0301N, AND REYES PHYSICAL REVIEW E 84, 011701 (2011) Our model follows that presented by Bailey and coworkers for the case of homogeneous orientation [11]; in order to facilitate comparisons with that case, we adopt most of the notation introduced in that model. We consider the simple fiber structure shown in Fig. 1: a cylindrical fiber with external radius Rf is composed by coaxial smectic layers. Near the center of the fiber, we assume that there is a defect core of size Rc Rf ; the results of the analysis are independent of Rc and the energy associated with this core [11]. We take into account five contributions to the Helmholtz free-energy density of a bent-core smectic fiber: elastic distortions of the director, layer compression, divergence of polarization, electrostatic dipole interaction, and surface tension. To describe these contributions, we use a molecular basis formed by the director n, the molecular-dipole vector p, and m = n \u00d7 p, and introduce cylindrical coordinates (R,\u03c6,z). We assume that n is confined to planes normal to the fiber axis, with a radially dependent tilt \u03b8 (R); we also assume a constant molecular-dipole orientation \u03b1 (see Fig. 1). The molecular basis is parametrized as: n = cos \u03b8R\u0302 + sin \u03b8 \u03d5\u0302, (1) p = sin \u03b8 sin \u03b1R\u0302 \u2212 cos \u03b8 sin \u03b1\u03d5\u0302 + cos \u03b1z\u0302, (2) m = sin \u03b8 cos \u03b1R\u0302 \u2212 cos \u03b8 cos \u03b1\u03d5\u0302 \u2212 sin \u03b1z\u0302. (3) In this parametrization, the director is always perpendicular to the fiber\u2019s axis but with the possibility of tilting away from the radial direction. The tilt angle \u03b8 = 0 corresponds to a director that is along the radial direction (that is, normal to the smectic layers and the surface of the fiber), while \u03b8 = \u03c0/2 corresponds to the director perpendicular to the radial direction" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003153_0309-524x.35.4.381-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003153_0309-524x.35.4.381-Figure3-1.png", + "caption": "Figure 3: Simplified sketch of the pitch system, showing its main components and the four interfaces.", + "texts": [ + " A blade bearing is designed based on the loads at the interfaces between the blade and the bearing and the interfaces between the bearing and the hub, as well as the required life time of the bearing. The life time is dependent on the friction of the bearing, which is dependent on the loads at the interfaces. If the actual friction moment during operation does not correspond to the calculated friction moment, this may have consequences for the blade bearing life time and for the pitch drive (e.g. overload or fatigue). For the pitch system four different interfaces have been determined, as shown in figure 3 [11]: 1. The interface between the blade & the pitch system (bearing) 2. The interface between the hub & the pitch system (bearing) 3. The interface between the hub & the pitch system (transmission & drive) 4. The interface between the controller & the pitch system (drive) An interface has been defined similar to the definition given in [12] as a defined boundary of the specific system that is either a physical mount to another wind turbine subcomponent or a path of exchange such as control signals, hydraulic fluid, or lubricant" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002387_ijit.2013.053298-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002387_ijit.2013.053298-Figure6-1.png", + "caption": "Figure 6 The twin rotor multi-input multi-output system (TRMS)", + "texts": [ + " The whole design process of the ACO used in this study is given as follows: t = 0, give the value of \u03c1, r0, mant and initialise pheromone trials \u03c4i,j(0) = \u03c40, (ip = 1,\u2026,Np) and (j = 1, 2,\u2026,J) For t = 1 (first iteration) For t = 2,\u2026,itermax For the first and kant = 1, For the first and kant = 1, Choose the parameters vector p ant i kP\u23a1 \u23a4\u23a3 \u23a6 using equation (31) Choose the parameters vector p ant i kP\u23a1 \u23a4\u23a3 \u23a6 using equation (31) Leave traces on the graph using (33) if ( )P p ant i kF \u03c1< is verified Leave traces on the graph if: ( )P ( 1)p ant i bestkF F t< \u2212 and ( )P p ant i kF \u03c1< End For each ant kant = 2,\u2026,mant Set ( )P ( 1)p ant i bestkF F t= \u2212 if ( )P ( 1)p ant i bestkF F t> \u2212 End Choose the parameters vector p ant i kP\u23a1 \u23a4\u23a3 \u23a6 using equation (31) For each ant kant = 2,\u2026,mant Leave traces on the graph if (32) is verified Choose the parameters vector p ant i kP\u23a1 \u23a4\u23a3 \u23a6 using equation (31) Evaluate the obtained cost function ( )P p ant i kF according to (34) Leave traces on the graph according to (33) if (32) is verified End Evaluate the obtained cost function ( )P p ant i kF according to (34) Set ( )( ) P p ant i best kF t F= End Increment 0int 0 0 0 1: rr r r iter \u2212 = + Increment 0int 0 0 0 1: rr r r iter \u2212 = + End End In this section, we present two illustrative examples to show the effectiveness of the proposed decentralised control scheme. Where, it is applied respectively to a laboratory helicopter called the TRMS and to interconnected inverted pendulums on carts; these systems have a strong interconnection between their subsystems and they can be considered as good examples to test our control method. The TRMS is a laboratory set-up developed by Feedback Instruments Ltd. (1998) for control experiments. As shown in Figure 6, this system is characterised by a complex highly non-linear dynamics, where some states are inaccessible for measurements, and hence this set-up can be considered as a challenging engineering problem. From the control point of view, TRMS exemplifies a high order non-linear system with significant cross-couplings. Moreover, the control problem of this system has been considered as a challenging research topic (Allouani et al., 2011, 2012), it has gained a lot of attention because the dynamics of the TRMS and a helicopter are similar in certain aspects" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003816_amr.291-294.2115-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003816_amr.291-294.2115-Figure2-1.png", + "caption": "Fig. 2 Architecture and coordinate system", + "texts": [ + " Light weight linkages are used for better performance with high speed. However, lightweight linkages are more likely to vibrate, and the trajectory accuracy of the system motion is sacrificed. All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 152.14.136.77, NCSU North Carolina State University, Raleigh, USA-10/04/15,03:33:54) deformed linkages are shown in Fig. 2. Using Lagrange\u2019s method, the dynamic equations of motion for the 3-DOF system is given as [2]: fgM K M M F\u03c1\u03b7 \u03b7 \u03c1 \u03b2\u03b2+ = \u2212 \u2212 + (1) where M is the modal mass matrix of the system, K is the structural modal stiffness matrix, Ffg is the modal force from the effect of the rigid body motion on the elastic vibration and couple each other, and M M\u03c1\u03c1 \u03b2\u03b2\u2212 \u2212 reflects the effect of elastic vibration of the linkages due to the rigid body motion. Classic Multi-Mode Input Shaper. The classic multi-mode input shaper is used when the vibration of the system is increased due to of the effects of secondary or higher mode of system" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003505_amm.687-691.7-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003505_amm.687-691.7-Figure1-1.png", + "caption": "Fig. 1 Slicing cutter", + "texts": [ + " Tool and workpiece rotate Synchronous at high speed, and workpiece feed along its axis at the same time.Then the forcedmeshing movement of the cutter and workpiece isperformed, and a small groove is formed on workpiece. Inthis way, many grooves are formed by forcing meshing ofthe cutter and workpiece on a series of positions along thetooth trace. The tooth surface is formed by these grooves [1] . Based on the above machining principle, the error-free slicing cutter is specially designed for gear slicing in Ref. [2], the slice cutter shown in Fig. 1(a) consists offive components. They are rake face, top flank face, major flank face, top edge, and major edge. The interference happens between machining tooth surface and major flank face. For building the model of interference amount, the model of major flank face is built firstly. The model of the major flank face is designed for further research. According to the Ref. [3], the coordinate system of slice cutter shown Fig. 1(b) is built, and the model of the major flank face is 2 2 h 2 2 0 2cos( ) sin( ) ( )r r R r R \u2206b\u03d1 \u03b7 \u03d1 \u03b7= + + + + \u2212 \u2212 +r i j k (1) Where,\u03d1 is the unfold angle measured on the projection of major edge on plane x2O y2, which can be expressed as y y tan \u03b1 \u03b1\u03d1 = \u2212 (2) Where, \u03b1y is the pressure angle measured on the projection of major edge on plane x2Oy2, which can be expressed as b y arccos( ) r \u03b1 r = (3) Where, rb is base circle radius of the generating gear. In the Fig. 1(b), r means the distance from the point on major edge to axis z2; \u2206b means the distance between the plane x2Oy2 and front end surface after cutter grinding; R means the spherical radius of rake face; \u03b1z' means the relief angle measured on the top edge; \u03b3z' means the rake angle measured on the top edge; ra means the addendum circle radius of slice cutter. Model of interference amount For building the model of the interference amount, the model of tooth surface is builtfirstly, and the model of surface is given in the Ref" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001498_b978-1-4557-3116-9.00001-9-Figure1.3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001498_b978-1-4557-3116-9.00001-9-Figure1.3-1.png", + "caption": "FIGURE 1.3", + "texts": [ + "2C. In the next step, a fiber-free cloth was used to polish the surface that mechanically breaks the top part of the CNTs and exposes the tip of the CNTs as shown in Figure 1.2D. Finally, the sample surface was rinsed in deionized water, and an insulated copper wire (0.5 mm in radius) was attached to the corner of the substrate by applying a drop of conductive silver epoxy followed by insulating epoxy. The copper wire NEAs assembly was left to cure in air at room temperature for several hours. Figure 1.3 shows the structure of the final CNT NEAs. The CNT NEAs have been applied for detection of toxic metal ions. With Hg-coated CNT NEA, linear relationship between the Pb signals and the Pb concentration in the solutions ranging from 1 to 100 ppb (g/L) has been obtained [55]. Because Hg is highly toxic and not suitable for field-deployable use, the relatively benign bismuth, Bi(III), has been evaluated as a Hg substitute [56,57]. Bi-based electrodes performed as well as Hg-based electrode for Cr(VI) quantitation after the accumulation of Cr (VI)-diethylenetriammine pentaacetic acid (DTPA) chelate [56]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003051_amm.394.245-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003051_amm.394.245-Figure3-1.png", + "caption": "Fig. 3 Sketch map of on-machine measuring simulation", + "texts": [ + " Firstly, the solid model of workpiece and grinding wheel is created according to the method listed in literature[7-8] and then the gear is machined by the Boolean operation between the solid model of workpiece and grinding wheel. Secondly, the solid model of probe is created according to the dimensions of probe. In order to decrease the measuring simulation time, the on-machine measuring mechanism is replaced by probe solid model. Thirdly, As for the simulation of on-machine measuring, the movement of gear solid model and probe solid model is controlled by software according to the on-machine measuring principle and method. As shown in Fig. 3, the contacting status between the solid model of gear and probe is determined by the Intersect Boolean operation. When the probe solid model approach the tooth profile of gear solid model along the direction of normal vector of theoretical tooth profile, it represents there is no contact if volume of solid generated by Intersect Boolean operation equals to zero. Otherwise, it represents they are contacting and the probe is triggered if volume of solid generated by Intersect Boolean operation is not equal to zero" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003430_amr.383-390.6762-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003430_amr.383-390.6762-Figure2-1.png", + "caption": "Figure 2. Loads on the rivet and stress acting on the element", + "texts": [ + " Deformation analysis for the riveting press process The riveting process is similar to the metal flow problem due to large plastic deformation of rivet and sheet material around the rivet. It is impossible to calculate the deformation directly. As a result, the deformation analysis in this paper is based on a series of assumptions. a) The materials of the rivet and sheets are isotropic. b) Assume that the punches are ideally rigid bodies and they are unable to deform during the riveting process. c) The displacements of rivet and sheets are independent of coordinate z, and shearing stresses Theoretical relation between deformations and contact loads. Figure 2. shows loads on the rivet during the riveting press process. The rivet, top sheet and bottom sheet are considered as ideal cylinders. Thus, the stresses and displacements can be similarly described as follows [5]: 222 22 22 22 1)( rab ppba ab bpap baba r \u2212 \u2212 \u2212 \u2212 \u2212 =\u03c3 (1) 222 22 22 22 1)( rab ppba ab bpap baba t \u2212 \u2212 + \u2212 \u2212 =\u03c3 (2) ( ) r Erab ppba E r ab bpap E u z ba ba \u03c3 \u03bd\u03bd \u03bd \u2212 \u2212 \u2212+ + \u2212 \u2212\u2212 = 11 1 22 22 22 22 (3) Where r\u03c3 are radial stresses (MPa); t\u03c3 are circumferential stresses (MPa); z\u03c3 are axial stresses (MPa); u is displacement (mm); r is the current radius (mm); a is the inner radius of the cylinder (mm); b is the outer radius of the cylinder (mm); ap , bp are uniform pressures on its inner surface and outer surface (MPa); E is the elastic modulus (MPa); \u03bd is Poisson\u2019s ratio" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002011_2011-01-1691-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002011_2011-01-1691-Figure3-1.png", + "caption": "Figure 3. Focused sections and adjacent sections in VT paths at front-side body", + "texts": [ + " In this method, the stiffness of spring should be specified properly because inappropriate stiffness coefficient causes low accuracy. It takes notice that each element adjoining the focused cutting section of body structure to be investigated plays the part of a transfer path. Calculation of contribution by element requires accurate and numerous transmitted forces in the contribution analysis. Transmitted force is obtained by the use of element force of which calculation is used the vibration response xc, xa of the FE model for the elements between the cut section and the adjacent section shown in Figure 3 that zooms in the red circle in Figure 2. The cut section \u201cc\u201d is expressed by dotdashed lines and the adjacent section \u201ca\u201d is expressed by dashed lines. Transmitted force vectors Rc, Ra in each DOF are derived by the following equation of motion for elements. (2) where m is mass matrix, k is stiffness matrix and x is response vector of the assembly. Based on the reciprocity theorem, transfer functions defined from boundary points to an output point are derived at once by frequency response results when analysis gives a unit impulse input to a response point [7]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001077_978-1-4419-7979-7_6-Figure6.26-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001077_978-1-4419-7979-7_6-Figure6.26-1.png", + "caption": "Fig. 6.26 (a and b) Graphical construction [2] of the exciting current if based on given timedependent flux function and the associated F\u2013iF characteristic; (a) induced voltage e, flux F, and exciting current iF, (b) corresponding (F\u2013iF) loop", + "texts": [ + ", double-valued) magnetic properties of the core mean that the waveform of the exciting current differs from the sinusoidal (impressed) waveform of the flux or voltage. The waveform of the exciting current as a function of time can be found graphically from the magnetic B\u2013H characteristic, as illustrated below. B and H are related to F and iF, respectively, by known geometric constants: from the continuity of flux condition f\u00bcBcAc and from Ampere\u2019s law follows iF \u00bc \u00f0Hc\u2018c\u00de=N. The AC (B\u2013H) loop has been drawn in Fig. 6.26b in terms of f and iF, and minor loops as well as the trajectory through the origin have been neglected. Cosine and sine waves of induced voltage e(t) and flux F(t), respectively, are shown in Fig. 6.26a. At any given time t0 the flux F0 can be identified in Fig. 6.26a which corresponds to F0 of the (F\u2013iF) loop of Fig.6.26b, which corresponds to the value of iF 0. The iF 0 will be then plotted in Fig. 6.26a which represents one point of the iF 0(t) function. This procedure can be repeated for many more points. Notice that because the B\u2013H loop is multi-valued (e.g., double-valued), it is necessary to associate the rising-flux values of the time function with the rising-flux portion of the (B\u2013H) loop; the same applies to the falling portions. Also, the amplitude of the flux-time function must have the same amplitude value as the maximum value of the (F\u2013iF) loop. Notice that the graphically constructed exciting current (graphical construction can be replaced by numerical computer solution) is nonsinusoidal: it is about triangular for a sinusoidal flux" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure8-1.png", + "caption": "Figure 8 COMMANDS FOR 6-AXIS HOBBING MACHINE", + "texts": [ + " The hobbing process includes the specification of the following parameters: A-axis: hob swivel\u00f1 B-axis: hob rotation\u00f1 C-axis: workpiece/gear rotation\u00f1 X-axis: radial in-feed\u00f1 Y-axis: hob shift (axial)\u00f1 Z-axis: gear shift (axial).\u00f1 Determination of the command signals for the 6-axis gear hobbing machine are obtained by establishing a fixed frame of reference (X , Y , Z ). The command signals are determinedf f f by assuming that each element of the machine is a rigid body. Six coordinates are used to specify the joint commands necessary for the fabrication of gears using the gear hobbing machine depicted in Figure 7. These six coordinates are indicated in Figure 8. The six parameters necessary to specify these command signals are listed in Table 1. The position and orientation of the cutter relative to the gear blank is accomplished by a combination of absolute displacements of both the gear blank and the cutter. This combination of absolute displacements is analyzed as two mechanisms operating together. 4 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use This example considers a spur bevel gear set for motion transmission between intersecting axes to introduce the developed process" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002968_gt2011-45903-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002968_gt2011-45903-Figure1-1.png", + "caption": "Fig. 1. Hyperstatic shaft-line.", + "texts": [ + " From the analysis of the current literature it results that a complete analysis of the dynamic effect of rigid coupling misalignment on a real shaft line, i.e. a hyperstatic rotor with several bearings and couplings, is lacking. In this paper, the authors propose a complete and original method to simulate the behaviour of real shaft line, supported by several oil-film bearings, with rigid coupling misalignment. Nonlinear effects are highlighted and the spectral components of system response are analyzed, in order to give pertinent diagnostic information. Let\u2019s consider a hyperstatic shaft-line, like that represented in Fig. 1, which is composed by three different rotors, connected by two rigid couplings. In particular, the first rotor is the HP-IP (high pressure \u2013 intermediate pressure) turbine, the second is the LP (low pressure) turbine and the last one is the generator. The model proposed is anyway applicable for other types of machines, with different number of couplings or rotor bearings. The shaft-line is modelled in a standard way by means of a finite beam model and rigid disks, considering only the lateral vibrations, and 4 degrees of freedom (d", + " (6), the fully assembled system of equations is nonlinear, because many terms of it depends on the angular position t : [ ] ( ) ( ) ( )C u bt M x C G x K x F W F F (16) where the over-bars indicate that the corresponding vectors are padded with zeros on the foundation d.o.f.s, and the matrices are those of the fully assembled system. The nonlinear system of equations in Eq. (16) is integrated in the time domain using the Newmark\u2019s implicit method, in which all the quantities depending on are evaluated for each integration step. The model of the machine, used to show the results obtained with the described method, is relative to a steam turbo-generator unit of about 320 MVA, already sketched in Fig. 1. Node number is equal to 175 and bearing #1, #2 and #7 are of tilting pad type, while the others of 2-lobes type. The system response is calculated at the operating speed of 3000 rpm, considering also the presence of an unbalance on about the mid of the LP turbine (0.3 kgm with phase 0\u00b0). Different combinations of radial and angular misalignment conditions of the rigid coupling between HP-IP and LP turbines have been analysed. An example is shown in Figs. 4-7, where only the shaft orbits, in nodes corresponding to the vibration measuring planes close to the bearings #1-#4 of the actual machine, are shown" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure10-1.png", + "caption": "Figure 10 Rear suspension model(see online version for colours)", + "texts": [ + " These types of manoeuvres can be performed using the simulation model, which supports virtual suspension testing and optimisation. In the virtual model, any value can be measured at any time step. This project was initially limited to study the final phase: the nose-up exit. Generally, this phase corresponds to the acceleration of the bike out of corners. The other phases during cornering were also briefly studied. The rear suspension contains a shock absorber that rests on a fulcrum placed on a connecting triangle, which is linked to a set of connecting rods (pull rods) and the swing arm. Figure 10 shows this layout. In recent years, SRT has used several different linkage geometries for the rear suspension system of their bikes. All of these versions are based on the same underlying concept, but the dimensions and the properties of the spring and the damper are slightly different. The Fedem model in this study was based on the geometries of the Ilmor X3 2007 model. The rear suspension was modelled, as shown in Figure 10, by connecting a spring (1) and a damper (2) between the parts (placed below the swing arm) and the cross beam (4), which was attached to the frame. It was important that the spring had the correct preload and spring stiffness and that the damper coefficient was correct in both compression and extension. The primary spring (1) and the damper (2) were represented by \u00d6hlins\u2019 TTX36 monoshock unit, which is shown in Figure 11. The free joint (3) between the link (triangle) and the cross beam at the frame had the properties of the top out spring and the end stops, which limited the maximum movement of the suspension" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure23-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure23-1.png", + "caption": "Figure 23 Crank shaft rotations (see online version for colours)", + "texts": [ + " The steering systems were not activated in these tests. The original bike was able to ride steadily and did not drift very far away from the initial straight line. The bike with wheels with 50% of the gyroscopic effect from the wheels of the original bike was more unstable: larger roll angles were obtained than those for the original bike, and the bike also drifted more obviously away from the initial line. The pivot axle, which connects the swing-arm to the main frame, is movable in the vertical direction. Figure 23 illustrates the two different settings of the pivot axle position that were tested in Fedem. The axle is moveable on real bikes primarily to alter the effect of the chain forces. The chain forces have both positive and negative effects on the suspension performance and hence the behaviour of the entire bike. The positive effect is that the chain tension stabilises the rear suspension by compressing the spring and the damper during the acceleration of the bike. The chain pull tends to lift the swing-arm, whereas the driving force that is transmitted to the ground pushes the wheel axle forward" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003975_amr.411.222-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003975_amr.411.222-Figure2-1.png", + "caption": "Fig. 2 The processing of rack tool", + "texts": [ + " The shape of spline tooth profile is enveloped from the rack tooth profile into the relative movement. Correct tooth profile form the necessary and sufficient condition is to keep two pitch curves of the rack and the gear blank pure rolling inrelative motion. In the simulation process of MATLABsoftware ,in order to examine the envelope characteristics of cutting tool conveniently. assuming a fixed gear blanks,pitch curve of rack tool around reference circle of gear blank to do pure rolling, as shown in Fig. 2b [2]. Establish a new m-files in MATLAB,Eq.5 and the known parameters incorporated into the new m-file.The theoretical profile can be obtained according to Eq.5, this equation is a transcendental equation. The maximum profile angle with the hob and the minimum profile angle are is divided into n equal portions. Because it is a complex curve, so the number of copies n is not too small, otherwise it will not reach the precision.Then make other curve and two curves are symmetrical,their symmetry axis is half of the normal tooth thickness, the normal tooth can be obtained which is connected by straight lines the top of the two curves" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001210_iciafs.2014.7069541-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001210_iciafs.2014.7069541-Figure1-1.png", + "caption": "Fig. 1. Overall view of the 2-DOF system.", + "texts": [ + " At least, 5-DOF, yawing, pitching, rolling, grasping and linear motion of forceps, is required to perform laparoscopic surgery. In this paper, especially yawing and pitching motion system is treated. The half-circle-shaped tubular linear motor, \u201ccircular shaft motor (CSM)\u201d that enables the motion along circumference of circle [3] is utilized to realize these two motions. The CSM can translate contact force thanks to its high back-drivability. Each of yawing and pitching motion can be realized by arranging CSM flatly and vertically to the ground. The 2-DOF system shown in Fig. 1 with vivid force transmission is developed. This paper especially focuses on pitching motion influenced by gravity force. Gravity and friction force deteriorates force transmission characteristics. In addition, gravity force in the direction of motion and friction characteristics are changed by the position and the velocity of mover. Then, these two disturbances are modeled as a function of position and velocity. Gravity and friction compensation by modeling disturbance can improve force transmission" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002148_amm.539.3-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002148_amm.539.3-Figure2-1.png", + "caption": "Figure 2 Bearing misalignment of flexible rotor system", + "texts": [ + " Dynamic models of LP rotor with bearing misalignment According to the structure and mechanical characteristics of high and LP rotor system of aero-engine, and the classification of misalignment problems, the aero-engine rotor system bearing misalignment problems are divided into two categories to be studied. The dynamic model of high and LP rotor with misalignment excitation are established, then differential equations of motion are acquired, finally the connection between misalignment structure characteristic parameters and vibration response mechanical characteristic parameters is established. Based on that, the mechanism of the dynamic response characteristics of the rotor system caused by misalignment is studied. The dynamic model of rotor system with multi-supports is shown as Fig.2. The whole rotor system consists of two flexible rotors which are connected by the connection structure with angular rigidity of kc. The rotor system is supported by three bearings, all of them has different support rigidity. When the third bearing has an eccentricity of \u03b4 compared to the first two bearings, deformation occurs on the rotor system and the deformation mainly appears on the section with weak rigidity. The angle \u03b8 between two rotor axis tangent lines on connection structure is used to describe the relative deformation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002398_cisda.2014.7035629-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002398_cisda.2014.7035629-Figure1-1.png", + "caption": "Fig. 1. Model of the missile using aero-fins and RCS", + "texts": [ + " and xn(t) is i nM THEN ( ) ( ) ( ) ( ) ( ) i i i t t t t t = + = x A x B u y C x , 1, 2,...,i r= where i jM (j = 1,2,...,n) denote input fuzzy sets, r is number of fuzzy rules, [ ]2( ) ), ( ), , ( ) T 1 nt x (t x t x t=x denotes the state vector, y(t) and u(t) are output and input vectors; (Ai, Bi, Ci) are coefficient matrices of linearization system model with respect to fuzzy rule ith. For more details, it is possible to reference to [10]. The nonlinear longitudinal model of missiles using aerodynamic and reaction-jet control systems is introduced in [9] and illustrated in Fig. 1. Mathematically, the nonlinear longitudinal system are represented by the following forms: ( ) 1 1 sin( ) cos( ) Z y y y IM C qS P C qS mV P n \u03b1 \u03b4\u03b1 \u03c9 \u03b1 \u03b1 \u03b4 \u03b1 = \u2212 + \u2212 + (1) } 1 1 1 1 1 21 cos( ) cos( ) Z Z y F Z Z Z y F y IM GD qSLC qS x m I V C qS x P x n \u03c9\u03b1 \u03b1 \u03b4 \u03b4 \u03c9 \u03b1 \u03b1 \u03c9 \u03b1 \u03b4 \u03a6= + + + (2) 1 W ( )ZV \u03c9 \u03b1= \u2212 (3) where: - \u03b1 is angle of attack (AOA); - 1Z\u03c9 is pitch rate; - V and W are velocity and normal acceleration, respectively; - yC\u03b1 and yC\u03b4 are normal force coefficient derivatives with AOA and aero-fin angle, respectively; - S is reference area; - P and IMP are forces of jet engine propulsion and impulsive reaction jet, respectively; - q is velocity pressure; - n is number of impulses; - m is mass; - 1 1 Z Zm\u03c9 is pitching moment coefficient derivative with pitch rate; - 1ZI is moment of inertia; - L\u03a6 is the length of missile; - y\u03b4 is aero-fin angle; - , ,F F GDx x x\u03b1 \u03b4 are center of pressure of missile, center of pressure on aero fins, and center of gravity of lateral impulsive thruster" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003421_ilt-01-2011-0007-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003421_ilt-01-2011-0007-Figure2-1.png", + "caption": "Figure 2 The stresses", + "texts": [ + " Continuity equation: 1 r \u203a \u203ar \u00f0rvr\u00de \u00fe 1 r \u203avu \u203au \u00fe \u203avz \u203az \u00bc 0 \u00f023\u00de Projection of the motion equation on the r axis: 2 \u203ap \u203ar \u00fe \u203ah \u203ar \u00f0A1\u00derr \u00fe 1 r \u203ah \u203au \u00f0A1\u00deur \u00fe \u203ah \u203az \u00f0A1\u00dezr \u00fe h \u203a\u00f0A1\u00derr \u203ar \u00fe 1 r \u203a\u00f0A1\u00deur \u203au \u00fe \u203a\u00f0A1\u00dezr \u203az \u00fe \u00f0A1\u00derr 2 \u00f0A1\u00deuu r \u00fe 0:85K \u203ay1 \u203ar A2 1 rr \u00fe 1 r \u203ay1 \u203au A2 1 ur \u00fe \u203ay1 \u203az A2 1 zr \u00fe y1 \u203a A2 1 rr \u203ar \u00fe 1 r \u203a A2 1 ur \u203au \u00fe \u203a A2 1 zr \u203az \u00fe A2 1 rr 2 A2 1 uu r \" #) 2 1 2 K \u203ay1 \u203ar \u00f0A2\u00derr \u00fe 1 r \u203ay1 \u203au \u00f0A2\u00deur \u00fe \u203ay1 \u203az \u00f0A2\u00dezr \u00fey1 \u203a\u00f0A2\u00derr \u203ar \u00fe 1 r \u203a\u00f0A2\u00deur \u203au \u00fe \u203a\u00f0A2\u00dezr \u203az \u00fe \u00f0A2\u00derr 2 \u00f0A2\u00deuu r \u00bc 0 \u00f024\u00de Projection of the motion equation on the u axis: 2 1 r \u203ap \u203au \u00fe \u203ah \u203ar \u00f0A1\u00deru \u00fe 1 r \u203ah \u203au \u00f0A1\u00deuu \u00fe \u203ah \u203az \u00f0A1\u00dezu \u00fe h \u203a\u00f0A1\u00deru \u203ar \u00fe 1 r \u203a\u00f0A1\u00deuu \u203au \u00fe \u203a\u00f0A1\u00dezu \u203az \u00fe \u00f0A1\u00deru \u00fe \u00f0A1\u00deur r \u00fe 0:85K \u203ay1 \u203ar A2 1 ru \u00fe 1 r \u203ay1 \u203au A2 1 uu \u00fe \u203ay1 \u203az A2 1 zu \u00fe y1 \u203a A2 1 ru \u203ar \u00fe 1 r \u203a A2 1 uu \u203au \u00fe \u203a A2 1 zu \u203az \u00fe A2 1 ru \u00fe A2 1 ur r \" #) 2 1 2 K \u203ay1 \u203ar \u00f0A2\u00deru \u00fe 1 r \u203ay1 \u203au \u00f0A2\u00deuu \u00fe \u203ay1 \u203az \u00f0A2\u00dezu \u00fey1 \u203a\u00f0A2\u00deru \u203ar \u00fe 1 r \u203a\u00f0A2\u00deuu \u203au \u00fe \u203a\u00f0A2\u00dezu \u203az \u00fe \u00f0A2\u00deru \u00fe \u00f0A2\u00deur r \u00bc 0 \u00f025\u00de Projection of the motion equation on the z axis: 2 \u203ap \u203az \u00fe \u203ah \u203ar \u00f0A1\u00derz \u00fe 1 r \u203ah \u203au \u00f0A1\u00deuz \u00fe \u203ah \u203az \u00f0A1\u00dezz \u00fe h \u203a\u00f0A1\u00derz \u203ar \u00fe 1 r \u203a\u00f0A1\u00deuz \u203au \u00fe \u203a\u00f0A1\u00dezz \u203az \u00fe \u00f0A1\u00derz r \u00fe 0:85K \u203ay1 \u203ar A2 1 rz \u00fe 1 r \u203ay1 \u203au A2 1 uz \u00fe \u203ay1 \u203az A2 1 zz \u00fe y1 \u203a A2 1 rz \u203ar \u00fe 1 r \u203a A2 1 uz \u203au \u00fe \u203a A2 1 zz \u203az \u00fe A2 1 rz r \" #) 2 1 2 K \u203ay1 \u203ar \u00f0A2\u00derz \u00fe 1 r \u203ay1 \u203au \u00f0A2\u00deuz \u00fe \u203ay1 \u203az \u00f0A2\u00dezz \u00fey1 \u203a\u00f0A2\u00derz \u203ar \u00fe 1 r \u203a\u00f0A2\u00deuz \u203au \u00fe \u203a\u00f0A2\u00dezz \u203az \u00fe \u00f0A2\u00derz r \u00bc 0 \u00f026\u00de where K is the normalization coefficient. For _g! 0\u00f0y1=h\u00de! \u00f0y10=ho\u00de \u00bc 1 and K \u00bc \u00f0y10=ho\u00de\u00f0Vs=R\u00de. The Weissenberg number is We \u00bc l\u00f0Vs=R\u00de. Taking We \u00bc 2.5 and the time constant l \u00bc 8.04 s we have Three-dimensional lubrication theory in viscoelastic short-bearing S\u0327ule Celasun Volume 65 \u00b7 Number 6 \u00b7 2013 \u00b7 357\u2013368 K \u00bc 2.5/8.04 \u00bc 0.32. It is worthwhile to notice that for K \u00bc 0 we go back to the generalized Newtonian fluid. The normal and shear stresses are as below (Figure 2): trr \u00bc 2p\u00fe h\u00f0A1\u00derr \u00feK 0:85y1 A2 1 rr 2 1 2 y1\u00f0A2\u00derr tuu \u00bc 2p\u00fe h\u00f0A1\u00deuu \u00feK 0:85y1\u00f0A2 1\u00deuu 2 1 2 y1\u00f0A2\u00deuu tzz \u00bc 2p\u00fe h\u00f0A1\u00dezz \u00feK 0:85y1 A2 1 zz 2 1 2 y1\u00f0A2\u00dezz tru \u00bc h\u00f0A1\u00deru \u00feK 0:85y1 A2 1 ru 2 1 2 y1\u00f0A2\u00deru trz \u00bc h\u00f0A1\u00derz \u00feK 0:85y1 A2 1 rz 2 1 2 y1\u00f0A2\u00derz tzu \u00bc h\u00f0A1\u00dezu \u00feK 0:85y1 A2 1 zu 2 1 2 y1\u00f0A2\u00dezu \u00f027\u00de The torque M and the load components: Wx \u00bc Z l=2 2l=2 Z 2p o pR cos u du dz Wy \u00bc Z l=2 2l=2 Z 2p o pR sin u du dz W* x \u00bc Z 1 21 Z 2p o p* cos u d u dz W* y \u00bc Z 1 21 Z 2p o p* sin u du dz W* \u00bc ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi W *2 x \u00feW *2 y q M \u00bc Z l=2 2l=2 Z 2p O truR 2d u dz\u00fe Z R\u00feho R Z 2p o tzuR 2drd u M* \u00bc Z 1 21 Z 2p O t*rududz\u00fe R l=2 Z 1\u00fe\u00f0ho=R\u00de 1 Z 2p o t*zudr *du W \u00bc W*Vsho l 2 M \u00bc M*VshoR l 2 We shall apply the FEM method with using of in threedimensional geometry the 15-node quadratic prism element for the velocities (Figure 3), and a nine-node linear prism element for the pressures, because for stability the pressure field must be interpolated with a polynomial one order lower than the velocity terms (Reddy and Gartling, 2001)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002344_detc2013-12837-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002344_detc2013-12837-Figure1-1.png", + "caption": "Figure 1. SCHEMATIC ILLUSTRATION OF GENERATION PROCESS FOR A CURVILINEAR GEAR", + "texts": [ + " Litvin [2, 3] developed a simplified method for determination of principal and normal curvatures for gear tooth surfaces in threeparameter form. The principal directions and curvatures of the generated surfaces can be represented in terms of those of their corresponding generating tool surfaces by utilizing the methodology developed by Litvin [2, 3]. Several researchers have applied Litvin\u2019s methodology to determine the contact ellipses of mating gear surfaces [6-8]. Liu [9] proposed a practical manufacturing method by using a fly-cutter mounted on a rotating disk Fig.1(a) and also examined the characteristics of curvilinear gears. Dai et. al. applied a CNC machine to manufacture curvilinear gear set with complementary male and female fly cutters [10]. Their manufactured curvilinear gear set possesses line contact and is sensitive to axial misalignments. Tseng and Tsay [11, 12] developed the mathematical model of curvilinear gears generated by an imaginary straight-edged male cutter. The contact characteristics of their curvilinear gear set comprising both pinion and gear generated by the same straight-edged male fly cutters reveals point contact and localized bearing contact", + " Finally, the effects of the design parameters on the contact ellipses and Hertzian contact stress are also presented in numerical examples. MATHEMATICAL MODEL OF THE MODIFIED CURVILINEAR GEAR SET The modified curvilinear gear set is composed of a modified pinion and modified gear with curvilinear teeth. Both pinion and gear of the modified curvilinear gear set are generated by a male fly cutter with a circular-arc normal section. The proposed gear set is expected to exhibit point contact, and parabolic TE under ideal meshing condition. Figure 1(b) depicts the schematic formations of the rack cutter (generating) surface P . Herein, rack cutter surface P generates the involute pinion surface 1 , while surface G generates the modified circular-arc gear surface 2 . In the tooth contact analysis (TCA), the pinion surface generated by the left-side of rack cutter P is meshing with the gear surface generated by the right-side of rack cutter G . Equations of Rack Cutter P As shown in Fig.2, the normal section of the rack cutter surface P used for pinion surface generation is a circulararc", + "org/about-asme/terms-of-use Principal Directions and Curvatures of Rack Cutter Surface P The position vector and unit normal vector of the rack cutter surface P are expressed in Eqs.(1) and (2), respectively. According to Rodrigues\u2019 equation, at the point of contact, the relative velocity rV is collinear with rn (the velocity of the tip of the surface unit normal) along the principal directions. Thus the principal curvatures of a regular surface can be obtained from the Rodrigues\u2019 equation: rrIII nV , , (11) where III , represents the two principal curvatures of the surfaces at the instantaneous contact point. As shown in Fig.1, P and P are the surface parameters of the generating surface P . In the following, Eqs.(1), (2) and (11) are used to attain the principal directions and curvatures of the rack cutter surface P . The Rodrigues\u2019 equation for rack cutter P is derived expressed as follows: rz rz ry ry rx rx III V n V n V n , , (12) Thus . }coscos] 2 )cos(cos{[sinsin coscossinsin }sinsin] 2 )cos(cos{[cossin sincoscossin cos cos dt d E S R dt d R dt d dt d dt d E S R dt d R dt d dt d dt d R dt d P PPP P nPP P PPP P PP P PP P PPP P nPP P PPP P PP P PP P PP P P (13) For nontrivial solution of Eq" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002044_s11768-013-2012-3-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002044_s11768-013-2012-3-Figure3-1.png", + "caption": "Fig. 3 Correction using (a) deviated pursuit and (b) proportional navigation.", + "texts": [ + " There exists an error between Ti+1(t) and Ti(t) when the path of the pursuer P0 is curved and the distance traveled by the pursuer Pi+1 is smaller than the distance traveled by the pursuer Pi. Thus, the deviated pursuit and the proportional navigation can be used for the path correction of the pursuer. The correction is implemented by choosing the appropriate values of the deviation angles or the navigation constant. For the deviated pursuit and proportional navigation, these are illustrated in Fig. 3. The velocity pursuit requires the pursuer Pi+1 to move on the line of sight with the pursuer Pi, and this path is different from Ti. Under the deviated pursuit, the deviation angles \u03b1i+1 and \u03b2i+1 are applied to conduct a lag pursuit in order to match Ti+1 and Ti. Under the proportional navigation, the deviation angles J\u03c3i,i+1 and J\u03b3i,i+1 are also applied to conduct a lag pursuit. In the sequel, the aim is to briefly study the influence of sensor noise on the path of the following pursuer. Sensor noise can affect all measured quantities, such as the position or linear velocity of the lead pursuer" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002044_s11768-013-2012-3-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002044_s11768-013-2012-3-Figure2-1.png", + "caption": "Fig. 2 The pursuit situation between two successive pursuers. In this paper, the model of two successive pursuers is presented in the polar coordinates. Based on [16], the differential equations for the range, the pitch angle of Li,i+1 and the yaw angle of Li,i+1 are\u23a7\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23a8 \u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23a9", + "texts": [ + " The mathematical formulation of pursuer convoy is on the basis of the kinematics equations and the proposed guidance laws. The ith pursuer is denoted by Pi for i = 0, 1, . . . , N \u2212 1. The pursuer Pi has the following kinematics equations:\u23a7\u23aa\u23a8 \u23aa\u23a9 x\u0307i = vi cos \u03b8i cos \u03c6i, y\u0307i = vi cos \u03b8i sin\u03c6i, z\u0307i = vi sin \u03b8i, (1) where (xi, yi, zi) are the coordinates of the pursuer Pi in the reference coordinate system, \u03b8i and \u03c6i are the flight path and heading angles, and vi is the linear velocity of the pursuer Pi. The pursuit situation of two successive pursuers is illustrated in Fig. 2. The LOS between the pursuer Pi and the pursuer Pi+1 is denoted by Li,i+1. \u03c3i,i+1 is the pitch angle of Li,i+1, and \u03b3i,i+1 is the yaw angle of Li,i+1. r\u0307i,i+1 = vi cos \u03b8i cos \u03c3i,i+1 cos(\u03c6i \u2212 \u03b3i,i+1) + vi sin \u03b8i sin\u03c3i,i+1 \u2212 vi+1 cos \u03b8i+1 cos \u03c3i,i+1 cos(\u03c6i+1 \u2212 \u03b3i,i+1) \u2212 vi+1 sin \u03b8i+1 sin \u03c3i,i+1, ri,i+1\u03c3\u0307i,i+1 = \u2212vi cos \u03b8i sin \u03c3i,i+1 cos(\u03c6i \u2212 \u03b3i,i+1) +vi sin \u03b8i cos \u03c3i,i+1 +vi+1 cos \u03b8i+1 sin\u03c3i,i+1 cos(\u03c6i+1 \u2212\u03b3i,i+1) \u2212 vi+1 sin \u03b8i+1 cos \u03c3i,i+1, ri,i+1\u03b3\u0307i,i+1 cos \u03c3i,i+1 =vi cos \u03b8i sin(\u03c6i\u2212\u03b3i,i+1) \u2212 vi+1 cos \u03b8i+1 sin(\u03c6i+1\u2212\u03b3i,i+1)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001180_s00604-015-1579-4-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001180_s00604-015-1579-4-Figure3-1.png", + "caption": "Fig. 3 Scheme of the positive effect of thiol (1) and both SDS and thiol (2,3) in the fluorescence of the Fe3O4@PFR@ AuNPs nanospheres", + "texts": [ + " With the aim of assessing the potential effect of surfactants on the fluorescent behavior of this system, the influence of SDS, CTAB and Triton X-100, selected as representative anionic, cationic and non-ionic surfactants, was studied below and above their critical micellar concentrations (c.m.c.). The luminescence of the system was slightly modified in the presence of CTAB and Triton X-100 but, as Fig. 2 shows, it increased about 5-times in the presence of SDS when its concentration was above its c.m.c. This luminescence increase can be ascribed to the negative charge of the SDS micelles, which contributes to inhibit the FRET process by increasing the distance between the negatively charged Fe3O4@PFR nanospheres and the AuNPs surface. A scheme of the process is shown in Fig. 3. Figure 4 shows the emission spectra obtained for the Fe3O4@PFR@AuNPs nanospheres in the presence of SDS and different thiol compounds. As can be seen, the presence of SDS causes a slight increase in the luminescence of the nanospheres, as a result of the electrostatic repulsion, but the luminescence notably increases in the presence of thiols, showing all of them a similar behavior. Variables affecting the synthesis of the core-shell nanospheres and those affecting the chemical system were optimized using the univariate method" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003721_detc2014-34297-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003721_detc2014-34297-Figure9-1.png", + "caption": "FIGURE 9. Parallel platform where the mobile platform generates a Scho\u0308nflies subalgebra, xu\u03021 , with respect to the fixed platform.", + "texts": [ + " , n, no matter how many legs are associated to the same subspace , Vi, the directions of the unit vectors, u\u03021, associated to a Scho\u0308nflies subalgebra, xu\u03021 and to a subspace S1,u\u03021 , must coincide, except when this subspace generates a subalgebra t\u22a5u\u03021 or t\u22a5\u2217u\u03021 , in which case the directions of the unit vectors xu\u03021 , must coincide with the unit vectors associated to the subspaces S2,u\u03022 . With this three conditions, Step 3 of the synthesis process has been completed. Examples Two examples of the synthesis of parallel platforms where the velocity states of the mobile platform with respect to the fixed platform, V m/ f generates the Scho\u0308nflies subalgebra, xu\u03021 are shown here. 1. Consider the parallel platform shown in Figure 9. Leg 1 generates a Scho\u0308nflies subalgebra in the direction of u\u03021, corresponding to the case 9, Table 2, where the first and last kinematic pair generate the subalgebra cu\u03021 , while the kinematic pairs from two to four, generate the subalgebra t\u22a5\u2217u\u03022 . Leg 2 generates the Scho\u0308nflies subalgebra in the direction of u\u03021, corresponding to the case 2, Table 2, where the first two kinematic pairs together with the last kinematic pair, generate the subalgebra gu\u03021 , while the kinematic pairs three and four generate the subalgebra t\u2217u\u03023 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001878_acs.2499-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001878_acs.2499-Figure2-1.png", + "caption": "Figure 2. Elements that define an elliptical orbit in the plane.", + "texts": [ + " According to the adopted simplifications, this article examines the basic maneuver constituting orbital transfer with zero energy (or accelerating to parabolic velocity). Spacecraft motion is described by means of the orbital elements. Orbital elements are six quantities determining the shape and size of the orbit of a celestial body, its position, and the position of the celestial body in space. Orbital elements describe the law of motion of a celestial body at any given time. The shape and dimensions of the orbit are determined by the semi-axis of the orbit, and by the eccentricity \" of the orbit (Figure 2). For the elliptical orbit, the eccentricity values are enclosed within the 0 6 \" < 1. For values of \" from 0 to 1, the orbit shape is increasingly elongated. The orbit orientation in space is defined with respect to a plane, taken as basic. For the planets, comets, and other elements of the solar system, such plane is the plane of the ecliptic. Figure 2 shows the orbit, ellipse, where one focus F1 is the origin of the coordinates, the center of gravity. Major axis of the ellipse \u02db connects the apses: \u02db is the aphelion, is the perihelion, A is length of the semi-major axis, B is the length of the semi-minor axis, and C is the distance from the focus to the center of the ellipse. Parameters A, B , and C are related by the following relations through the parameter p and the eccentricity \": A D p 1 \"2 ; B D p p 1 \"2 ; C D p\" 1 \"2 (1) where p is the focal parameter of a conic section and \" is the eccentricity of the conic section" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001201_20140313-3-in-3024.00080-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001201_20140313-3-in-3024.00080-Figure1-1.png", + "caption": "Figure 1 Launch vehicle in Pitch plane", + "texts": [ + " In section 3 lag-lead compensator design based on multi-constraint satisfaction algorithm and traditional approach is detailed. Simulation results are presented in section 4. Also, the closed loop behaviour of slosh dynamics with the variation in actuator command amplitudes is discussed and the necessity for having required minimum slosh phase margin is brought out. The concluding remarks are given in Section 5. 2.PROBLEM FORMULATION The attitude control of launch vehicle during atmospheric flight is considered here (fig.1). During the atmospheric phase, the attitude control has to ensure stability and command tracking, while maintaining the structural integrity of the vehicle. Also it needs to ensure disturbance rejection with respect to winds. To achieve these control objectives, the following specifications (Jiann-Woei et al. 2011) have to be met. i) Tracking error < 1deg ii) Low frequency gain margin > 6dB iii) High frequency gain margin > 6dB (1) iv) Rigid Body Phase margin > 30 deg v) Slosh Phase margin > 30 deg The rigid body dynamics is modelled by third order system with state vector [ z]x \u03b8 \u03b8 \u2032= & & " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001136_978-3-319-19788-3_1-Figure1.19-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001136_978-3-319-19788-3_1-Figure1.19-1.png", + "caption": "Fig. 1.19 Examples of spatial robots with 6 DOF. a A Hexapod (courtesy of Sym\u00e9trie). b The Hexa (Pierrot et al. 1990)", + "texts": [ + " Some of them have been designed with an additional wrist which compensates for the undesirable rotations and have found some industrial applications, especially for milling (Fig. 1.18) \u2022 robots with three translational DOF and one rotational DOF around one given axis (also called Sch\u00f6nflies motion generators): they are usually used for pickand-place operations, most often at high-speed. The most functional robot of this type is probably the Adept Quattro (Fig. 1.4) \u2022 robots with six DOF: such as the Hexapod (also known as the Gough-Stewart platform\u2014Fig. 1.19a) and the Hexa (Pierrot et al. 1990) (Fig. 1.19b). 1.3.3 Redundant PKM Redundancy occurs when the number of active joints, na , is greater than the number ndof of independent variables required to define the platform configuration. Redundancy in PKM is usually used in order to avoid their singularities which are considered as one of the main drawbacks of such robots (see Sect. 7.5). Redundancy in parallel manipulators can be divided into two main groups: 1. Kinematic redundancy: in such a case, na = Ndof > ndof , 2. Actuation redundancy: in such a case, na > Ndof " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.31-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.31-1.png", + "caption": "Fig. 2.31. Admissible postures for a two-link planar arm", + "texts": [ + " Finally, the angle \u03d13 is found from (2.90) as \u03d13 = \u03c6\u2212 \u03d11 \u2212 \u03d12. An alternative geometric solution technique is presented below. As above, the orientation angle is given as in (2.90) and the coordinates of the origin of Frame 2 are computed as in (2.91), (2.92). The application of the cosine theorem to the triangle formed by links a1, a2 and the segment connecting points W and O gives p2 Wx + p2 Wy = a2 1 + a2 2 \u2212 2a1a2 cos (\u03c0 \u2212 \u03d12); the two admissible configurations of the triangle are shown in Fig. 2.31. Observing that cos (\u03c0 \u2212 \u03d12) = \u2212cos\u03d12 leads to c2 = p2 Wx + p2 Wy \u2212 a2 1 \u2212 a2 2 2a1a2 . For the existence of the triangle, it must be \u221a p2 Wx + p2 Wy \u2264 a1 + a2. This condition is not satisfied when the given point is outside the arm reachable workspace. Then, under the assumption of admissible solutions, it is \u03d12 = \u00b1cos\u22121(c2); the elbow-up posture is obtained for \u03d12 \u2208 (\u2212\u03c0, 0) while the elbow-down posture is obtained for \u03d12 \u2208 (0, \u03c0). To find \u03d11 consider the angles \u03b1 and \u03b2 in Fig. 2.31. Notice that the determination of \u03b1 depends on the sign of pWx and pWy; then, it is necessary to compute \u03b1 as \u03b1 = Atan2(pWy, pWx). To compute \u03b2, applying again the cosine theorem yields c\u03b2 \u221a p2 Wx + p2 Wy = a1 + a2c2 and resorting to the expression of c2 given above leads to \u03b2 = cos\u22121 \u239b \u239dp2 Wx + p2 Wy + a2 1 \u2212 a2 2 2a1 \u221a p2 Wx + p2 Wy \u239e \u23a0 with \u03b2 \u2208 (0, \u03c0) so as to preserve the existence of triangles. Then, it is \u03d11 = \u03b1\u00b1 \u03b2, where the positive sign holds for \u03d12 < 0 and the negative sign for \u03d12 > 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003526_kem.572.355-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003526_kem.572.355-Figure2-1.png", + "caption": "Figure 2 Generation of internal gear", + "texts": [ + " Thus, equation of meshing may be represented in coordinate system cS as follows [1] : i cy i c i c i cx i c i c n yY n xX \u2212 = \u2212 (4) Symbols i cX and i cY represent the coordinates of a point on the instantaneous axis of gear rotation I-I in coordinate system cS ; i cx and i cy and are the coordinates of the instantaneous contact point on the rack cutter surface; i cxn and i cyn , are the direction cosines of the rack cutter surface unit normal i cn . Angle 1\u03c6 is the rolling parameter and the symbol 1pr denotes the radius of the gear pitch circle. Recalling that Eq. (4) represent the equation of meshing between the generated tooth surface and the rack cutter, it can be rewritten as follows [1]: )/()()( 11 i cxp i cy i n i cx i nj nrnxnyl \u2212=\u03c6 (5) By simultaneously considering Eqs. (3) and (5), the mathematical model of the generated gear can now be obtained. Figure 2 illustrates the relationship between driving external spur gear and driven internal gear. The coordinate system ),( hhh YXS is the reference coordinate system, the coordinate system ),( 222 YXS denotes the internal gear blank coordinate system, and the coordinate system ),( 111 YXS represents pinion gear coordinate system. On the basis of gear theory, driving gear rotates through an angle 1\u03c6 while the gear blank rotates through an angle 2\u03c6 . Based on the above idea, the coordinate transformation matrix from 1S to 2S can be represented as [1] : [ ] \u2212\u2212\u2212 \u2212\u2212\u2212\u2212 = 100 sin)()cos()sin( cos)()sin()cos( 2121212 2121212 21 \u03c6\u03c6\u03c6\u03c6\u03c6 \u03c6\u03c6\u03c6\u03c6\u03c6 pp pp rr rr M (6) According to the theory of gearing [1], the mathematical model of the generated internal gear tooth surface is a combination of the meshing equation and the locus of the pinion surfaces" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003330_20130918-4-jp-3022.00029-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003330_20130918-4-jp-3022.00029-Figure1-1.png", + "caption": "Fig. 1. Coordinate systems", + "texts": [ + " A fully numerical time domain simulation was carried on and the results showed the benefits of the cooperative control, when compared to the noncooperative one. In that paper, the cooperative controller was designed using LQG-LTR control theory applied to the multivariable system model involving the states of both vessels. The DP System is only concerned about the horizontal motions of the vessel, that is, surge, sway and yaw. The motions of the vessels are expressed in two separate coordinate systems (Fig. 1): one is the inertial system fixed to the Earth, OXYZ (also known as global reference system); and the other, A or B , are the vessel-fixed non-inertial reference frames (also known as local reference system). The origin for this system is the intersection of the mid-ship section with the ship\u2019s longitudinal plane of symmetry. The axes for this system coincide with the principal axes of inertia of the vessel with respect to the origin. The motions along the local axes are called surge, sway e yaw, respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003172_sii.2014.7028011-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003172_sii.2014.7028011-Figure2-1.png", + "caption": "Fig. 2. Distribution of DOF for MH-2.", + "texts": [ + " The subscript numbers 1 to 4 coincide with motor number for the 4-DOF redundant arm and 5 to 8 coincident with motor number for the wrist which is driven by 4 motors. The influence of gravity can be disregarded at the wrist, because the wrist weight is very light. )(+)(+)(=)( \u03b8g\u03b8d\u03b8f\u03b8M (1) [ ]T)()()()()( \u03b8\u03b8\u03b8\u03b8\u03b8M 8321 \u039c\u039c\u039c\u039c = (2) [ ]T)()()()()( \u03b8\u03b8\u03b8\u03b8\u03b8f 8321 ffff = (3) [ ]T)()()()(=)( \u03b8\u03b8\u03b8\u03b8\u03b8d 8321 dddd (4) [ ]T)()(=)( 000041 \u03b8\u03b8\u03b8g gg (5) A. Interference of joint The distribution of the joint is illustrated in Fig. 2. Since MH-2 uses looped wire at each joint except wrist joints, the whole length of the wire does not change. But interference occurs at joints 3\u03b8 and 4\u03b8 by the joint 2\u03b8 .This relationship is described in equation (6)~(9), where )=( 41nnk is the reduction ratio between the motor and pulley at the joint, and )=( 41nn\u03b8 is the angle of each joint. 111 \u03b8kf =)(\u03b8 (6) 222 \u03b8kf =)(\u03b8 (7) 22333 \u03b8k\u03b8kf +=)(\u03b8 (8) 22444 \u03b8k\u03b8kf +=)(\u03b8 (9) B. Inverse kinematic of 3-DOF parallel wire mechanism for a wrist 3-DOF parallel wire mechanism is employed for the wrist" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.23-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.23-1.png", + "caption": "Fig. 2.23. Anthropomorphic arm", + "texts": [ + "50) yields T 0 3(q) = A0 1A 1 2A 2 3 = \u23a1 \u23a2\u23a3 c1c2 \u2212s1 c1s2 c1s2d3 \u2212 s1d2 s1c2 c1 s1s2 s1s2d3 + c1d2 \u2212s2 0 c2 c2d3 0 0 0 1 \u23a4 \u23a5\u23a6 (2.65) where q = [\u03d11 \u03d12 d3 ]T . Notice that the third joint does not obviously influence the rotation matrix. Further, the orientation of the unit vector y0 3 is uniquely determined by the first joint, since the revolute axis of the second joint z1 is parallel to axis y3. Different from the previous structures, in this case Frame 3 can represent an end-effector frame of unit vectors (ne, se,ae), i.e., T 3 e = I4. Consider the anthropomorphic arm in Fig. 2.23. Notice how this arm corresponds to a two-link planar arm with an additional rotation about an axis of the plane. In this respect, the parallelogram arm could be used in lieu of the two-link planar arm, as found in some industrial robots with an anthropomorphic structure. The link frames have been illustrated in the figure. As for the previous structure, the origin of Frame 0 was chosen at the intersection of z0 with z1 (d1 = 0); further, z1 and z2 are parallel and the choice of axes x1 and x2 was made as for the two-link planar arm", + " A comparison of the vector p0 6 in (2.68) with the vector p0 3 in (2.65) relative to the sole spherical arm reveals the presence of additional contributions due to the choice of the origin of the end-effector frame at a distance d6 from the origin of Frame 3 along the direction of a0 6. In other words, if it were d6 = 0, the position vector would be the same. This feature is of fundamental importance for the solution of the inverse kinematics for this manipulator, as will be seen later. A comparison between Fig. 2.23 and Fig. 2.24 reveals that the direct kinematics function cannot be obtained by multiplying the transformation matrices T 0 3 and T 3 6, since Frame 3 of the anthropomorphic arm cannot coincide with Frame 3 of the spherical wrist. Direct kinematics of the entire structure can be obtained in two ways. One consists of interposing a constant transformation matrix between T 0 3 and T 3 6 which allows the alignment of the two frames. The other refers to the Denavit\u2013Hartenberg operating procedure with the frame assignment for the entire structure illustrated in Fig", + "70) and n0 6 = \u23a1 \u23a3 c1 ( c23(c4c5c6 \u2212 s4s6) \u2212 s23s5c6 ) + s1(s4c5c6 + c4s6) s1 ( c23(c4c5c6 \u2212 s4s6) \u2212 s23s5c6 ) \u2212 c1(s4c5c6 + c4s6) s23(c4c5c6 \u2212 s4s6) + c23s5c6 \u23a4 \u23a6 s0 6 = \u23a1 \u23a3 c1 ( \u2212c23(c4c5s6 + s4c6) + s23s5s6 ) + s1(\u2212s4c5s6 + c4c6) s1 ( \u2212c23(c4c5s6 + s4c6) + s23s5s6 ) \u2212 c1(\u2212s4c5s6 + c4c6) \u2212s23(c4c5s6 + s4c6) \u2212 c23s5s6 \u23a4 \u23a6 (2.71) a0 6 = \u23a1 \u23a3 c1(c23c4s5 + s23c5) + s1s4s5 s1(c23c4s5 + s23c5) \u2212 c1s4s5 s23c4s5 \u2212 c23c5 \u23a4 \u23a6 . By setting d6 = 0, the position of the wrist axes intersection is obtained. In that case, the vector p0 in (2.70) corresponds to the vector p0 3 for the sole anthropomorphic arm in (2.66), because d4 gives the length of the forearm (a3) and axis x3 in Fig. 2.26 is rotated by \u03c0/2 with respect to axis x3 in Fig. 2.23. Consider the DLR manipulator, whose development is at the basis of the realization of the robot in Fig. 1.30; it is characterized by seven DOFs and as such it is inherently redundant. This manipulator has two possible configurations for the outer three joints (wrist). With reference to a spherical wrist similar to that introduced in Sect. 2.9.5, the resulting kinematic structure is illustrated in Fig. 2.27, where the frames attached to the links are evidenced. As in the case of the spherical arm, notice that the origin of Frame 0 has been chosen so as to zero d1", + " Hence, it is \u03d11 = 2Atan2 ( \u2212pWx \u00b1 \u221a p2 Wx + p2 Wy \u2212 d2 2, d2 + pWy ) . Once \u03d11 is known, squaring and summing the first two components of (2.94) yields d3 = \u221a (pWxc1 + pWys1)2 + p2 Wz, where only the solution with d3 \u2265 0 has been considered. Note that the same value of d3 corresponds to both solutions for \u03d11. Finally, if d3 = 0, from the first two components of (2.94) it is pWxc1 + pWys1 \u2212pWz = d3s2 \u2212d3c2 , from which \u03d12 = Atan2(pWxc1 + pWys1, pWz). Notice that, if d3 = 0, then \u03d12 cannot be uniquely determined. Consider the anthropomorphic arm shown in Fig. 2.23. It is desired to find the joint variables \u03d11, \u03d12, \u03d13 corresponding to a given end-effector position pW . Notice that the direct kinematics for pW is expressed by (2.66) which can be obtained from (2.70) by setting d6 = 0, d4 = a3 and replacing \u03d13 with the angle \u03d13 +\u03c0/2 because of the misalignment of the Frames 3 for the structures in Fig. 2.23 and in Fig. 2.26, respectively. Hence, it follows pWx = c1(a2c2 + a3c23) (2.95) pWy = s1(a2c2 + a3c23) (2.96) pWz = a2s2 + a3s23. (2.97) Proceeding as in the case of the two-link planar arm, it is worth squaring and summing (2.95)\u2013(2.97) yielding p2 Wx + p2 Wy + p2 Wz = a2 2 + a2 3 + 2a2a3c3 from which c3 = p2 Wx + p2 Wy + p2 Wz \u2212 a2 2 \u2212 a2 3 2a2a3 (2.98) where the admissibility of the solution obviously requires that \u22121 \u2264 c3 \u2264 1, or equivalently |a2\u2212a3| \u2264 \u221a p2 Wx + p2 Wy + p2 Wz \u2264 a2 +a3, otherwise the wrist point is outside the reachable workspace of the manipulator" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003561_j.protis.2012.12.001-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003561_j.protis.2012.12.001-Figure1-1.png", + "caption": "Figure 1. Timeline of events in \u2018doing\u2019 the book.", + "texts": [ + " A full list of contents with a brief description (say 10 lines) per chapter, although the more information you can include the more the reviewers will have to go on. 4. A very brief CV (0.25 page for each coauthor) detailing professional backgrounds. Undaunted and now armed with actual instructions, I wrote a book proposal including details such as a target cover price of under D50 with 20 pages of color illustrations. Also included were the number of copies held in libraries in the USA of previous, somewhat similar, titles of various ages. These holdings ranged from about 200 to 500. Here begins the timeline (Fig. 1). The proposal went out to all the publishers I could find, namely all those listed in Table 2. In the following couple of weeks I had polite refusals from most. Many responses noted that the cover price was unrealistic and repeated Arthur Finder\u2019s opinion that such a book was for a small market and commercially impossible. A couple stated they would take the proposal under consideration. Some publishers never responded and some took months to acknowledge receipt of the proposal. As it turns out, I had made several mistakes already" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002673_holm.2014.7031067-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002673_holm.2014.7031067-Figure7-1.png", + "caption": "Fig 7. The spring returning structure schematic diagram", + "texts": [ + " Because there are two contact pairs which is caused by the moving contact leading way, the difficulty of contact resistance stability in the process of manufacture is increased. From armature returning mode, the AEMR can be classified into four types: spring returning, permanent magnet returning, permanent magnet magnetic circuit returning and polarized magnetic circuit returning. Clapper type relay generally uses spring to make armature return, it belongs to the structure of spring returning. Balance armature type AEMR can use both spring and permanent magnet to accomplish the return process. Spring returning type relay (showed in Fig. 7) is used widely. To meet the requirement of vibration resistance, the spring can be replaced by permanent magnet (showed in Fig. 8). In Fig. 8, the permanent magnetic circuit is an independent open magnetic circuit, which has a low magnetic efficiency. Compared with the spring returning mode, the advantage of permanent magnet returning is that it has a strong retentivity force when the armature is on the steady release state. And the disadvantage is that in a small initial stage the force is very low when armature is releasing" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003721_detc2014-34297-Figure12-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003721_detc2014-34297-Figure12-1.png", + "caption": "FIGURE 12. Parallel platform where the mobile platform generates a subspace, xu\u03021 \u2295 ru\u03022 , with respect to the fixed platform.", + "texts": [ + " Leg 1 generates a subspace xu\u03021 \u2295 hu\u03022,p, corresponding to case 2, Table 3. Leg 2 generates the subspace xu\u03021 \u2295 ru\u03022 , corresponding to case 1, Table 3. Leg 3 generates the subspace gu\u03021 \u2295Sgu\u03022 , corresponding to case 3, Table 3. Moreover, the conditions for forming a platform were also used, that is, the directions associated to xu\u03021 , in legs 1 and 2, and gu\u03021 , in leg 3, are parallel, while the directions associated to hu\u03022,p, ru\u03022 and Sgu\u03022 , in legs 1, 2 and 3, respectively, are also parallel. 2. Consider the parallel platform showed in Figure 12. Leg 1 generates a subspace xu\u03021 \u2295 hu\u03022,p, corresponding to case 2, Table 3. Leg 2 generates the subspace xu\u03021 \u2295 ru\u03022 , corresponding to case 1, Table 3. Leg 3 generates the subspace Sgu\u03021 \u2295gu\u03022 , corresponding to case 4, Table 3. Moreover, the conditions for forming a platform were also used, that is, the directions associated to xu\u03021 , in legs 1 and 2, and Sgu\u03021 , in leg 3, are parallel, while the directions associated to hu\u03022,p, ru\u03022 and gu\u03022 , in legs 1, 2 and 3, respectively, are also parallel" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003206_acc.2014.6858697-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003206_acc.2014.6858697-Figure1-1.png", + "caption": "Figure 1. The inverted pendulum subjected to high-frequency excitation at the base.", + "texts": [ + " (34) Remark 6: When the continuous-time system (1) is given in the form of a forced nonlinear oscillator with the system function being ,t t t t \u0393 x f x g , (35) the proposed discrete-time function (10) yields the following discrete-time function: , 0 0 0 0 1 , , 1 1 1 , T k T k k k T D kT d k k T D d k T D T k T D T T e kT d T e kT d T e d T e kT d T \u0393 x f x f x f x \u0393 x \u0393 x f x g f x g (36) which is identical to one proposed in [11]. IV. EXAMPLE: AN INVERTED PENDULUM SUBJECTED TO HIGH-FREQUENCY EXCITATIONS Open-loop stabilization of an unstable equilibrium state of an inverted pendulum was shown to be possible in [15], where a high frequency excitation is used in the vertical axis with no feedback. This excited pendulum, shown in Fig. 1, is a nonlinear non-autonomous system, which consists of a point mass m attached to the top end of a massless rod of length l . The bottom end is periodically excited along the vertical axis with the amplitude and angular frequency of excitation being ea and \u03c9, respectively. Its equation of motion is given by [15] 22 2 cos sin 0ead c d g t ml dt l ldt (37) where c is a viscous damping coefficient of the pivot at bottom. The system (37) can be rewritten in a vector form as 2 cos sineac g t ml l l , (38) for which the proposed discrete-time model is obtained as 1 1 1 , , k T k k k kT k T d T \u0393 x , (39) where 2, cos sin k k e k k t ac g t ml l l \u0393 x (40) Peano-Baker series , is given by (25) with Jacobian matrix being 2 0 1 cos cos k e k D ag c t l l ml \u0393 x " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001936_12.2052216-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001936_12.2052216-Figure1-1.png", + "caption": "Figure 1: Schematic view of the NIR-CGM sensor. (a): Sensor housing with integrated disposable fluidic chip, electronic board and vials for collecting dialysate and perfusate. (b) Disposable polymer chip showing the sample and reference cells. (c) Cross section of the detection area of the disposable polymer chip with concept of the inand out-coupling of the optical signals.", + "texts": [ + " The light emitted from the multi emitter LED is split up into two partial beams at a beam splitter, integrated in the polymer chip and deflected sideways by 90\u00b0. After passing through the reference and measuring channels, the light beams are deflected downwards again by 90\u00b0, onto the photodiodes D1 and D2. When the cover is installed, the system is opaque, so that the measurement is not disturbed by ambient light. A schematic of the fluidic concept for the disposable part as a polymer chip is shown in Figure 1. For each wavelength the optical transmission in the measuring and reference cell is measured and a difference signal is generated via electronic processing. The wavelengths have been selected in a way that a significant change of absorption appears in the measuring channel if glucose is present, resulting in a change of the difference signal, correlated to the glucose level in the dialysate. The fluidic chip is a low cost component, consisting of a medical grade polycarbonate, fabricated by established mass production techniques, like hot embossing and laser welding" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003794_amr.343-344.28-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003794_amr.343-344.28-Figure1-1.png", + "caption": "Fig. 1 The equivalent model of single DOF F ig. 2 Wear coefficient 2k vs. slip angle", + "texts": [ + " pxc , pyc are stiffness per unit length of tire, sx is start sliding point x\u03c3 , y\u03c3 are the slip rate in the longitudinal and lateral direction respectively. With neglect of the slippage of tire in longitudinal direction and assumption that the coefficients of static friction in longitudinal and lateral direction are the same as \u00b5 , we can get expression as follows considering the impact of rolling resistance fF on friction circle and the start sliding point sx can be obtained. Modeling of single-degree-of-freedom on slippage region Figure 1 shows the equivalent model of single-degree-of-freedom model on slippage region based on the reference [5]. The dynamic model of slippage in lateral direction of tire tread can be obtained according to Newton Euler Equations. ( ) zhr FVKxxCxM \u00b5=++ . (6) WhereK , C , M , zhF are the stiffness, the lateral damping factor, the mass, the applied vertical load of tire tread on slippage region respectively, xVVr \u2212\u22c5= \u03b1sin , \u03b1sin\u22c5=VVb , ( )rV\u00b5 is the lateral friction coefficient of contact patch which can be gained as an expression with LuGre friction model [6] proposed by Canudas de Wit Carlos" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003583_detc2013-12991-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003583_detc2013-12991-Figure3-1.png", + "caption": "Figure 3. VARIABLE TOPOLOGIES OF THE 3(rT)PS", + "texts": [ + " The three constraint forces in (1) constrain two translations along x-axis and y-axis with one rotation about z-axis. Thus, the 3(rT)2PS parallel mechanism with parallel constraint screws has three DOFs with two rotations about x-axis and y-axis with one translation along z-axis (2R1T). Altering the (rT)2PS limbs in the previous 3(rT)2PS parallel mechanisms into the phase (rT)1PS will result in various new mechanism topologies with increased mobility. After changing the phase of one limb, the 3(rT)2PS parallel mechanisms become the topology 2(rT)2PS-1(rT)1PS in Fig. 3(a) that has two parallel constraint screws following (1). One constraint less makes the 2(rT)2PS-1(rT)1PS one more DOF than the 3(rT)2PS. Based on the constraint screw analysis, the 2(rT)2PS-1(rT)1PS parallel mechanism has four DOFs with three rotations and one translation along z-axis (3R1T). r S2 A2 A3 B3 B2 S3 r A1 B1 (rT)1 (a) 2(rT)2PS-1(rT)1PS (4DOFs-3R1T) When further changing one more limb phases, the mechanism changes to the topology 1(rT)2PS-2(rT)1PS as in Fig. 3(b), which has only one constraint screw that limited the translation along 3 r S parallel to the bracket axis in limb 3. This mechanism has five DOFs with three rotations and two translations perpendicular to 3 r S , (3R2T). When changing the third limb to phase (rT)1PS, the mechanism becomes another topology 3(rT)1PS as in Fig. 3(c) that does not have any constraint screw with full mobility 6. The above shows one way of changing the limb phases in the order from limb 1 to limb 3 one by one. Actually, when changing the phases in different order or by different number, the mechanism can vary its topology from any one to another with mobility change among 3, 4, 5 and 6. 2. UNIFIED KINEMATICS Since the 3(rT)PS metamorphic parallel mechanism has variable topologies with different mobility each of which is an independent parallel mechanism, how to model those mechanisms in a unified form for applications becomes a challenge" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001946_amm.698.552-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001946_amm.698.552-Figure7-1.png", + "caption": "Fig. 7. Mechanism (-73-65+127-136+67+80) and its direction vectors", + "texts": [ + " The result is a new mechanism (-90-80-70-100-120-80) whose direction vectors are shown in Fig. 6. In this case, the condition of collinearity ofthe first and last direction vectors is satisfied, so the mechanism is assemblable. This assemblability condition of 6R mechanisms can be used for the synthesis of new mechanisms. Let us create a new mechanism choosing parameters of links \u03b11 \u03b12 \u03b13 \u03b14 \u03b15 \u03b16 randomly so that the mechanism satisfies the 6Rmechanism sassemblabilitycondition. In the mechanism shown in Fig. 7:\u03b11= -73\u00b0 \u03b12= -65\u00b0 \u03b13= 127\u00b0 \u03b14= -136\u00b0 \u03b15= 67\u00b0 \u03b16= 80\u00b0. Let us simulate the mechanism in the CAD system and ensure its assemblability. Mobile assemblable mechanisms are more interestingfor practical application. Let us address the mechanism record (\u03b11 \u03b12 \u03b13 \u03b14 \u03b15 \u03b16) to explain the mobility condition of Brikard's linkage modifications). The mechanism consists of six units, which can be divided into three pairs of units: P1, P2, P3 (Fig. 8). Brikard's linkage modification (\u03b11 \u03b12 \u03b13 \u03b14 \u03b15 \u03b16) is mobile if all links turn the direction vector in one direction (clockwise or counterclockwise), and each of the three link pairs rotates the direction vector through 180\u00b0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure2.15-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure2.15-1.png", + "caption": "Fig. 2.15 Maximum principal stress distribution with the arch-shaped implant", + "texts": [ + " Stress distribution of implants with three and four screws fixation showed very similar behavior at their central part and in contact with the mandible body (see Figs. 2.14 and 2.15) whereas the other two conditions had a higher stress concentration at the distal fixation screws. Finally, the arch-shaped implant showed a maximum stress distribution at the ramus condyles bilaterally, whereas theminimal stresswas observed at the osteotomy 2 Experimental and Numerical Evaluation of an Orthognathic Implant \u2026 63 area suggesting that the arch-shaped implant yield a better distribution of stress through the numerical model, see Fig. 2.15. In the last years, maxillofacial surgeons have adapted a so-called hybrid technique for fixation at the mandible structures. The main rationale for this is due to a better mechanical resistance that implants fixated with screws can provide, and the combination of bicortical or monocortical fixation, with and without the use of an implant. Bicortical fixation with screws was first introduced by Spiessl and popularized afterward by Paulus and Steinhauser [9]. Simultaneously, Michelet proposed fixation of the segment after an osteotomy by using sagittal sections and implants with monocortical screws [9]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001216_icit.2015.7125200-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001216_icit.2015.7125200-Figure2-1.png", + "caption": "Fig. 2. Principal IPM flux paths (a) d-axis and (b) q-axis [2]", + "texts": [ + " The price of IPM motors decreases if the cost of the material of the IPM motors decreases through decreasing the size of the IPM motors. The stator outer diameter of the conventional IPM motors are 105mm\u223c112mm. Therefore, the smallest IPM motors whose stator outer diameter is 90 mm have been developed for compressors of 2.2kW and 2.5kW air-conditioners. This motor can be driven using low-price inverters, and has high efficiency to satisfy the Japanese standard of COP. Fig.1 shows the configuration of typical IPM motors[2]. Fig.2 (a) and (b) show the flux paths for d-axis and q-axis, respectively [2]. The torque T is shown as follows[3], T = Pn{\u03c8aIa cos\u03b2 + 0.5(Lq \u2212 Ld)I 2 a sin 2\u03b2} (1) where Pn : a number of pole pairs, \u03c8f : maximum flux-linkage due to permanent magnet per phase, 978-1-4799-7800-7/15/$31.00 \u00a92015 IEEE 825 \u03c8a = \u221a 3 2 \u03c8f Ip : magnitude of the armature current per phase, Ia = \u221a 3 2 Ip \u03b2 = leading angle of armature current from the q-axis Ld, Lq = d- and q-axis components of armature self-inductances. The first term of Eq.(1) shows the magnet torque and the second term the reluctance torque. Conventionally, the optimum location of permanent magnets of IPM motors whose permanent magnets are buried in a single layer has been deep in the rotor core as in Fig. 1 [4] . The reasons described in [4] are shown as follows. The magnetic saturation occurs because q-axis flux flows as shown in Fig.2(b). The reluctance torque decreases as shown in Eq.(1) because of the drop in q-axis inductance caused by the magnetic saturation. Therefore, the locations described in [4] have been deep in the rotor core to make the q-axis magnetic path wide enough. The stator of the developed motor is the distributed winding one which has 24 slots, because the iron loss of the distributed winding stator can be less than one of the concentrated winding stator even if the stator is small and the winding factor of the distributed winding stator is more than one of the concentrated winding stator" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003316_amr.823.84-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003316_amr.823.84-Figure1-1.png", + "caption": "Fig. 1 and Fig. 2 show the models of linear layout (planetary gear) and vertical layout (worm gear reducer). In the experiment, the Yaskawa motor which used as a power input was corotation, meanwhile, Lenze motor was contrarotation which was used to simulate the load side and export different torque and speed setting previously.", + "texts": [], + "surrounding_texts": [ + "Mechanical transmission test platform is widely used these days in adjustable telescope tracking platforms, multiple object tracking, medical equipment movements, etc. Test the integrated test platform performance through different types of test objects to obtaining test data. The servo motor and gear reducers are usually direct-coupled or connected by coupling. The worm and planetary gears which are characterized by reducing speed and increasing torque are assembled individually [1]. However, the gear system displays undesirable characteristics such as the backlash, which becomes a mechanical imperfection. Backlash represents the gap between meshed gears teeth [2, 3], and result in vibration, resonance, decline of precision and noise [4].In this paper, SolidWorks and ANSYS were used to investigate mechanical transmission performance of test platform and compare the two kinds of gear reducer. Design and structure of a test platform Principle and modeling of test platform We use the motion control card and servo driver to accurately control and read the feedback of the output torque, position and speed of the motor. Related measurement device which is capable of real-time measurement will test the vibration and noise in different output curve of the rotational speed, torque in different platform. The platform need to meet the requirements of the experimental accuracy transmission, each component should be easy to operate, disassemble, and easy to convert the layout of the structure to the different transmission system loading tests." + ] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure28.14-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure28.14-1.png", + "caption": "Fig. 28.14. (a) Kinematic model of the mobile platform (b) Unit sphere representation of wheel configurations", + "texts": [ + " 1) Representation of wheel configuration: In order to satisfy the pure rolling and nonslipping condition, all wheel normals must be either parallel or intersect in a single point. The respective wheel configurations are called admissible wheel configuration (AWC). All AWCs can be represented on the surface of a unit sphere and can be described by using two spherical angles: the azimuth angle \u03b7 represents the direction of the translational motion and the altitude \u03b6 is a measure for the amount of rotational motion. In Fig. 28.14b the unit sphere model is illustrated. All configurations on the equator (\u03b6 = 0) correspond to pure translational motion while configurations at one of the poles (\u03b6 = \u00b1\u03c0/2) represent pure rotational motion. As an AWC does not specify the absolute speed of the platform motions, a third variable, the generalized velocity \u03c9 is introduced. From the AWC (\u03b7, \u03b6) the unit vector e(\u03b7, \u03b6) can be calculated. Including the generalized velocity \u03c9 yields the platform velocity in the Cartesian platform coordinate system:\u239b\u239d Px\u0307 Py\u0307 P\u03c8\u0307/\u03baG \u239e\u23a0 = \u03c9 \u239b\u239d cos \u03b6 cos \u03b7 cos \u03b6 sin \u03b7 sin \u03b6 \u239e\u23a0 = \u03c9e(\u03b7, \u03b6), (28" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001497_978-3-319-07572-3_6-Figure6.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001497_978-3-319-07572-3_6-Figure6.2-1.png", + "caption": "Fig. 6.2", + "texts": [ + " In the tensile or compressive stress, the force is applied normal to the cross section under consideration, however if the transverse force is applied to a member the internal forces develop in the plane of the cross section and they are called shearing forces (Fig. 6.1). Since shear distribution in the cross section of the member is not necessarily uniform, an average shear stress is used. Examples of shear stress are found in bolts, rivets and pins that are being used for connecting of various structural members. The relationship between average shear stress in the cross section and the shear (F) can be expressed as: \u03c4ave \u00bc F=A Figure 6.2, shows the rivet connection under tension force F and the shear stress will be developed in the rivet section connecting the plates. In this case, the force F is considered as the shear in the rivet section. The following are the useful terms that are extensively used in engineering generally, and mechanics of materials especially. Allowable Stress Allowable stress is the maximum stress that a member can safely sustain in service. It is also called design stress. Brittleness Brittleness is the property of a material to fail without exces- sive deformation Ductility Ductility is the property of a material that allows the material to elongate under certain stress without breaking" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002699_iemdc.2013.6556225-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002699_iemdc.2013.6556225-Figure2-1.png", + "caption": "Fig. 2. Different winding connections of 6-stator/4-rotor pole SR machine under AC sinusoidal bipolar with DC bias excitation.", + "texts": [ + " The operation principles of such excitation can be simply summarized as follows: the DC excitation flux, which is equivalent to magnet flux in PM machines, interlinks with the AC coils of each phase and induces voltages, which equivalently will be called the back-emf. Consequently, similar to PM machines, an output torque will be generated if the AC coils are excited by AC currents. Further illustrations and investigations will be presented in the following sections. Furthermore, the coils under AC sinusoidal bipolar with DC bias excitation have the possibility to be connected either symmetrically, Fig. 2(a), or asymmetrically, Fig. 2(b). However, it should be mentioned that in the case of asymmetric winding connection only 4 short-pitched DC coils are required, no need for DC coils on phase B teeth, i.e. no DC coils on teeth 2 and 5 in Fig. 1. This is because such coils do not contribute to the output torque, I 978-1-4673-4974-1/13/$31.00 \u00a92013 IEEE but they only generate an extra copper loss, i.e. different DC polarities in the same slot. It should be also mentioned that the asymmetric connection can be also equivalently obtained by two full-pitched DC coils (or actually one full-pitched DC coil), as illustrated in Fig. 2(bII) which actually is wound field doubly salient machines [13]. In this case, the effective length of the machine becomes larger due to longer DC coil endwinding. However, the machine efficiency for both short- and full-pitched DC coil excited machines depends on the total copper loss of all DC coils including the end-winding. In this study, only the short-pitched concentrated DC coils will be considered since both asymmetric winding connections shown in Fig. 2(b) are electromagnetically equivalent. It is also worth mentioning that under AC sinusoidal bipolar with DC bias excitation the AC current frequency is determined by the rotor pole number, i.e. similar to the switched flux PM (SFPM) machines [14]. In other words, one electric cycle completes when the rotor mechanically rotates one rotor pole pitch. As mentioned earlier, the operation principle of the AC sinusoidal bipolar with DC bias excitation is similar to its counterpart of the PM magnet machines", + " The open-circuit torque, i.e. cogging torque, of the two connections are predicted at 30 A/mm2 DC current density and compared in Fig. 5(a). It shows that the cogging torque period of the symmetrical winding connection is 30 mechanical degrees, since the least common multiple between the rotor poles and equivalent stator slots is 12, thus the cogging torque period is 360\u00f712=30. On the other hand, for the asymmetric connection, the DC stator polarity repeats once over the 360 mechanical degrees, Fig. 2(b). This is equivalent to two stator slots from the cogging torque producing point of view. Thus, the cogging torque is produced by the interaction of the 4 rotor poles with these slots. This means the cogging torque waveform is 90 mechanical degrees, as confirmed in Fig. 5(a). In order to examine the influence of the magnetic saturation, the cogging torque is predicted for different current densities and the peak values due to the two winding connections are compared in Fig. 5(b). It shows that the influence of the magnetic saturation is different according to the winding connection", + " The comparison illustrates that for significantly large current densities the flux density distribution becomes relatively more balanced, thus the change of cogging torque with rotor position is relatively small. However, at relatively low current density, Fig. 5(c), such distribution becomes less balanced, thus the cogging torque becomes relatively large. Furthermore, the three phase open-circuit flux linkages, which are seen by the AC coils, are calculated and compared for the two winding connections in Fig. 7. It shows that the flux linkage waveforms of the asymmetric winding connection are non-uniform, Fig. 7(a), since the winding distribution of such connection are also non-uniform, Fig. 2(b). On the other hand, the symmetrical winding connection results in uniform flux linkage waveforms, which are larger than their counterparts of the asymmetric connection, since such connection has larger air gap flux density, Fig. 3. However, the maximum to minimum variation of such waveforms is smaller since for the symmetrical winding connection, the difference between the flux distributions at the maximum flux linkage position of phase A, i.e. when the rotor poles are full aligned with the phase A stator poles, as shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002893_amm.312.345-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002893_amm.312.345-Figure1-1.png", + "caption": "Fig. 1 Schematic of brush seal Fig. 2 CFD model of brush seal", + "texts": [ + " However, less work has been studied the impact of the radial clearance on flow field and leakage, even less work has been done on methods to reduce the leakage under radial clearance condition. In this study,a retaining ring structure is employed to reduce the leakage flow rate on radial clearance conditions of brush seal, geometric configurations effect and disturbance effect of the retaining ring structure on flow field is analyzed. Considering convenience of analysis, axisymmetric model of the brush seal in Fig. 1 is created. As shown in Fig. 2, the sizes (dimensions in mm) of the respective parts of brush seal are from literature [2], c is radial clearance. Defining inlet and outlet boundaries with pressuresPu and Pd=0.1 MPa, lying 30 mm away from the front plate and backing plate, respectively.The air leakage flows from upstream with high-pressure to downstream with low-pressure have three approaches: inward radial flow f1, downward axial flow f2, leakage flow through radial clearance f3. Assuming the bristle pack as porous medium, thuscontinuity RANS equations are used to calculate the axial-radial axisymmetric solution domain of brush seal" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003239_amm.575.250-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003239_amm.575.250-Figure4-1.png", + "caption": "Figure 4: Bending moment at finite rod", + "texts": [ + "199, Purdue University Libraries, West Lafayette, USA-04/07/15,15:13:28) Consider a uniform finite rod in vertical axis having a length L subjected to a propagating axial wave and it shown in Figure 2. By applying the excitation force, eF at 0=y , the rod becomes partial displacement, y\u2206 at 0=y because the boundary condition at that area a free and at Ly = , there is no displacement because the boundary condition of fixed. This phenomenon illustrated at Figure 3. The rotation with small angle \u03b8\u2206 also occurred at free boundary. By principal static and dynamic, bending moments, e M also happen to this rod. Figure 4 is shown the diagram. Practically, dispersive system can be represented as the behaviour of the bending motion in finite rod [4]. According to J.R. Banerjee and Cheng, by using Euler-Bernoulli theorem and Timoshenko beam from axial vibration, it is can be modelled the bending moment effect on uniform finite rod [5-6]. In this study, the model of bending moments is developed from axial vibration. However the methodology is combined with Euler-Bernoulli theorem and wave propagation method. The relationship between angular displacement and rotation can be written as y\u03b4 \u03b4\u03c9 \u03b8 = (1) where \u03b4\u03c9 and y\u03b4 are the partial displacement", + " The general equation of at \u2212y axis displacement in term of time can be represented as Zoom in Figure 2: Finite rod in vertical axis Figure 3: Partial displacement happen at finite rod ( ) ( ) ( ) ( ) ( )yiktiyiktiyktiykti bbbb eAeAeAeAty \u2212+\u2212+ +++= \u03c9\u03c9\u03c9\u03c9\u03c9 4321, (2) where 4321 ,,, AAAA are constants and Bk is a flexural wavenumber with solution equation 4 EIAkB \u03c1\u03c9= where \u03c1 is density, A is cross-sectional area, E is Young\u2019s Modulus and I is a second moment of area. Eq. (1) is valid when differentiate Eq. (2). The new equation can be written as ( )( ) ( )( ) ( )( ) ( )( )Byik B yik B yk B yk ikeAikeAkeAkeA y BBBB \u2212++\u2212+= \u2212\u2212 .... 4321 \u03b4 \u03b4\u03c9 (3) Eq. (3) is known as equation of angular velocity of the finite rod. According to Figure 4, there has bending moment and shear stress relatively by the excitation force. Equation of bending moment can be derived by the second derivative and shear stress derived by third derivative. These two equations are written below in Eq. (4) and Eq. (5). ( )( ) ( )( ) ( )( ) ( )( )22 4 22 3 2 2 2 12 2 .... B yik B yik B yk B yk kieAkieAkeAkeA y BBBB \u2212\u2212 +++= \u03b4 \u03c9\u03b4 (4) ( )( ) ( )( ) ( )( ) ( )( )33 4 33 3 3 2 3 13 3 .... B yik B yik B yk B yk kieAkieAkeAkeA y BBBB \u2212\u2212 ++\u2212+= \u03b4 \u03c9\u03b4 (5) The constant value for four equations can be solved using trivial solution" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002322_1.4029294-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002322_1.4029294-Figure2-1.png", + "caption": "Fig. 2 The kinematic and static parameters of the system shown in Fig. 1", + "texts": [ + " Section 6 concludes with noting that the developments presented in this paper move SESC modeling toward more practical applications. This section reviews the classical determination of the CoM location of a multilink chain in order to define the SESC. Manipulations of the equations to compute the CoM of the original chain as the end-effector of the SESC are also presented. 2.1 Classical CoM Determination. The determination of the CoM for the articulated spatial four-body system shown in Fig. 1 is given as an example to illustrate the method. Figure 2 shows the kinematic parameters including the joint displacement vector d and joint rotation matrix A. Mass property information is also shown, including the mass m and the relative link CoM location c. Attaching a reference frame at each joint, the CoM of the system is defined by the parameters mi, ci;di, and Ai, where the subscript i indicates the numbering scheme used for identifying each body. The CoM location of each link in its local reference frame is indicated by ci 2 R3. The relative location of the origin of the (i\u00fe 1)th reference frame in the previous reference frame is indicated by the displacement vector di 2 R3", + " (1) C \u00bc m1 M \u00f0A1c1\u00de \u00fe m2 M \u00f0A1d1 \u00fe A1A2c2\u00de \u00fe m3 M \u00f0A1d2 \u00fe A1A3c3\u00de \u00fe m4 M \u00f0A1d2 \u00fe A1A3d3 \u00fe A1A3A4c4\u00de (2) Sorting Eq. (2) according to Q Ai C \u00bc A1s1 \u00fe A1A2s2 \u00fe A1A2 bA3s3 \u00fe A1A2 bA3A4s4 (3) where bA3 \u00bc A 1 2 A3 (4) s1 \u00bc m1 M c1 \u00fe m2 M d1 \u00fe m3 \u00fe m4 M d2; s2 \u00bc m2 M c2; s3 \u00bc m3 M c3 \u00fe m4 M d3; s4 \u00bc m4 M c4 (5) Equation (3) becomes C 1 \u00bc A1 0T 0 1 A2 s1 0 1 bA3 s2 0 1 A4 s3 0 1 s4 1 (6) Observe that Eq. (6) is the forward kinematic representation of a serial chain having links s1; s2; s3, and s4, as shown in Fig. 3. This indicates that the CoM location of the branched-chain system shown in Fig. 2 can be modeled by the terminus of an appropriately dimensioned spatial serial chain as seen in Eq. (6). This serial chain is termed the SESC. In a similar manner, any general branched chain composed of rigid bodies articulated by revolute, spherical, or universal joints defines an SESC. The links, or si, of the SESC are determined by the masses (mi), CoM locations of the links (ci), and the distances between the joints \u00f0di\u00de of the original system. The joint angles of the SESC correspond to the joint angles of the original system" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002717_1.4882556-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002717_1.4882556-Figure2-1.png", + "caption": "FIGURE 2. (a) The generating unit of the crease pattern , and (b) the flat foldability types of (a)", + "texts": [ + " Let denote the angles between the crease lines in which are arranged in the same order. If , then the crease lines and which form the angle cannot be both mountains or cannot be both valleys. Theorem 3The sum of alternate angles around a vertex v in a crease patternis . For example, the generating unit u of the unassigned crease pattern arising from the tiling (FIGURE 1(f)) consists of three parallelograms enclosing an isosceles right triangle where the vertex corresponding to the right angle is a vertex of a square and the remaining two vertices are vertices of octagons (FIGURE 2(a)). In this case, the group which acts on to create is where is a counterclockwise rotation centered at one of the octagons and and are two linearly independent translations (FIGURE 3(a)). By applying the necessary conditions of flat foldability to each vertex of the generating unit of , we can determine itsflat foldability types. A parallelogram has two flat foldability types.Since the generating unit is composed of three parallelograms, then it has 8 flat foldability types which are shown in FIGURE 2(b). Assigning three flat foldability types from the eight types given in FIGURE 2(b) to elements of will yield an mv-assignment of given in FIGURE 3(b). This results in a crystallographic flat origami. The folded pattern is shown in FIGURE3(c), which is fixed by the group a plane crystallographic group of type . Note that not all possible assignments of the flat foldability types to the elements of will give rise to crystallographic flat origami. For example, consider the mv-assignment of shown in FIGURE 4(a). This yields a flat origami presented in FIGURE 4(b). Observe that this flat origami is not fixed by any plane crystallographic group" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure9-1.png", + "caption": "Figure 9 Fork suspension model (see online version for colours)", + "texts": [ + " These two suspension units are further described in the following sections. The front suspension is produced by \u00d6hlins and is an upside down type of fork, i.e., the internal tube is in the lower position and is fixed to the wheel. This type of fork has been used in racing since the early nineties. These fork legs are illustrated in Figure 6. To simulate the fork suspension unit realistically, all of the properties of the fork had to be identified and incorporated into the simulation model. The primary spring (number 2 in Figure 9) in each of the fork legs had a stiffness of 9,000 [N/m] and a preload of 0.045 [m]. The preload did not alter the spring stiffness but simulated the effect of a stiffer spring. The preload was used to adjust how much travel resulted from placing a control unit under a certain load. The top out spring that worked against the extension of the fork had a stiffness of 3,500 [N/m] and a length of 0.04 [m]. These top out springs only affected the first 0.04 m of the stroke (which was measured relative to the full extension of the spring)", + " This spring function was used for a free joint between the upper and lower fork. The same method was used to model the rear suspension. During the compression phase, the volume of air contained in the fork diminishes, which serves a double function. 1 The air acts as another spring that is characterised by a high progressive rate. 2 Pressure is created inside the fork, which prevents cavitation problems and foam from forming on the oil-to-air contact surface (Cheney Engineering, 2007). The effect of a progressive air spring (number 1 in Figure 9) was incorporated into the model using a function (see Figure 7) that was used to describe the behaviour of a spring in each of the fork legs. This function included the properties of the gas pressure inside the fork, which depended on the front wheel stroke. This gas pressure created a force that acted to extend the fork; however, the force was zero when the fork reached full extension. This force created by the gas pressure was expressed as a function of the stroke length in Fedem Technology (2013)", + " Using the fork properties given by SRT, this load resulted in an initial fork compression of approximately 40 mm, which was measured relative to the fully extended length of the fork. The sliding fork motion between the lower and upper fork legs was implemented by two prismatic joints that allowed translational motion while preserving the correct fork leg bending stiffness (i.e., these joints were flexible and based on master and slave nodes). The suspension properties of each fork leg were modelled by two springs and a free joint that acted as a spring, and a damper. The spring on top of the model (number 1 in Figure 9) simulated the gas inside the fork. When the fork was fully extended, the gas pressure inside the fork was 0.1013 MPa, which was equal to the air pressure outside the fork: therefore, the gas had no effect. During the fork compression, the gas pressure rose and behaved as a progressive spring (see Figure 7). The largest fork spring (number 2 in Figure 9) was the primary spring and had a stiffness of 9 N/mm. The damper (number 3 in Figure 9) was non-linear and had a different damper coefficient in compression and rebound (see Figure 8). The free joint acted as a spring and performed the same function as the top out spring in the fork (number 4 in Figure 9). This spring only worked during the fork extension, and this free joint also included very stiff springs that acted as \u2018bump stops\u2019 to ensure that the wheel stroke never exceeded 130 mm. The fork could be extended to 40 mm and compressed to 90 mm from its equilibrium position. All of the springs (and the damper) were coupled in parallel in each fork leg because they were not directly connected to each other but between the same components. The detailed properties and tuning of the fork springs and dampers can be found in Giussani (2007) and Weiby (2007)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003608_isciii.2011.6069753-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003608_isciii.2011.6069753-Figure1-1.png", + "caption": "Fig. 1. The flux differential detector", + "texts": [ + " The above mentioned winding faults can be sensed by diverse failure detectors. In [9] several fault detection circuits for SRMs are detailed from very simple ones to complicated diagnosis systems. The most widely used detectors are sensing the overcurrent in the coils, the magnetic flux difference in the two series connected coils or the rate-of-rise of the phase current. Comparing their complexity and costs to performance ratio it seems that the best solution is to apply a flux difference detector (see Fig. 1) for each phase of the SRM. The detector requires additional search coils wrapped around the stator poles. The search coils of each phase are - 117 -978-1-4577-1861-8/11/$26.00 \u00a92011 IEEE connected in series opposing. Hence during normal operation the induced voltages of the search coils are equal and opposite, leaving a zero voltage at the terminals of the series pair. When a fault occur the magnetic flux will be different through the two poles, hence also the electromotive forces induced will be different and this voltage difference can be detected by the bidirectional comparator", + " The mean value of the developed torque during a displacement from the aligned to the unaligned position in the case of the healthy SRM is 3.41 N\u00b7m. When 20% of the turns - 119 - ISCIII 2011\u2022 5th International Symposium on Computational Intelligence and Intelligent Informatics \u2022 September 15-17, 2011, Floriana, Malta of one coil are shorted the torque development capability of the motor is reduced by 15% to 2.92 N\u00b7m. When only half of A coil's turns are working the developed torque of the SRM is 65% of the healthy machine's one (2.22 N\u00b7m). As also the testing of the flux differential winding fault detector given in Fig. 1 was proposed, the variation of the emf induced in the 100 turns search coil wound round pole A was plotted, too (see Fig. 7). As it was expected also this quantity is strongly influenced of the SRM's faults. The numeric field computations performed permitted also the computation of the voltage differences between the electromotive forces induced in two search coils from opposite poles (A and A'). This voltage difference is practically the one sensed by the flux differential detector. In Fig. 8 the input of the detector is plotted versus the angular position of the SRM for three conditions of the winding from pole A" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000869_s00170-021-07749-1-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000869_s00170-021-07749-1-Figure5-1.png", + "caption": "Fig. 5 (a) The position of cuts on the part with defined bending radius 0.7D, (b) outer (left) and inner (right) dimensions of Cut 4, (c) the course of deviations of the 0.7D bend against the CAD model in Cut 4", + "texts": [ + " Furthermore, the wall thickness of the pipe was also analysed in the case of the explosive cladding method presented in the Mr\u00f3z study [34]. The measurement of roundness deviations and deviations from the computer-aided design (CAD) model was performed using an Atos Compact 3D scanner. Pipe bend 1 was subjected to this analysis, and before the measurement it was necessary to degrease the sample, stick the points needed to determine the position and apply an anti-reflective spray for better visibility. For the pipe bend, 7 cuts were made, which are shown in Fig. 5a, analysing the roundness deviations for both the outer and inner pipe dimensions. The cuts are placed 45\u00b0 from the second to the sixth cut, and the cuts are parallel at the extended ends. The individual values of the deviation of the roundness of external and internal dimensions were calculated according to relation (2) and subsequently processed into Table 2. The limit value for external roundness for cold-bent bends 0.7D is not specified in EN 12952-5 [35] or otherwise. Likewise, the standard does not affect the forming of bimetal materials", + " For the internal dimension, the deviation of the roundness is not specified by the standard, and its control is subject only to the ball test passing through the pipe bend, which was successful for the given sample. The analysis of the internal shape and the evaluation of the deviations of the roundness of the cavity were performed only to verify the achieved values after the deformation in the bend, and the calculated values of the deviations of the roundness for the internal dimensions, see Table 2, were used only for control verification. The course of deviations against the CAD model and the individual measured diameters in Cut 4 are shown in Fig. 5b, c. From these deviations, it is clear that in some places, due to the bending process, more significant deformations occurred, but these deformations do not affect the proper functioning of the pipe bend. A similar comparison of the differences of the CAD model from the actual pipe was also performed in the Li study [26]. The hardness analysis was performed on a previously prepared metallographic preparation (prepared from Sample 1) as shown in Fig. 4, where the automatic hardness tester AMH 55 was used for the measurement" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001479_carpathiancc.2015.7145141-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001479_carpathiancc.2015.7145141-Figure1-1.png", + "caption": "Fig. 1 Arrangement of the controllable journal bearing.", + "texts": [ + " Research work was carried out at the VSB Technical University of Ostrava and the research company TECH LAB, Ltd., Prague. The active vibration control of the journal bearings uses the bushing which is free to move and non-rotating. The position of the bushing is governed by a pair of the 978-1-4799-7370-5/15/$31.00 \u00a92015 IEEE perpendicular piezoactuators and the position of the journal is measured by a pair of the proximity probes. The sketch of the controllable journal bearing arrangement is shown in Fig. I. Symbols of the spring elements in this Fig. 1 are as elastic 0- rings. The O-rings together with the piezoactuators determine the connection stiffness of the bushing with the bearing housing. With regard to future industrial applications, we focused on rigid rotors and the standard type of sliding bearings with the only exception that is possibility to control the bushing position with respect to the bearing housing. The journal movement in two directions is measured as close as it is possible to the bearing bushing. Some publications deal with the vibration of the flexible rotors, which have a large deformation in the middle span between the bearings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002019_peac.2014.7038056-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002019_peac.2014.7038056-Figure4-1.png", + "caption": "Fig. 4. The ESM steady vector graph in FOC scheme", + "texts": [ + " Thus, the main electromagnetic torque can be modulated quickly through the fast torque angle \u03b4sm change by rotating the stator flux linkage vector, which is done by switching the appropriate voltage space vector. III. THE FOC AND DTC SCHEME FOR ESM Fig. 3 shows the block diagram of FOC scheme for ESM. The outer Udc closed-loop ensures a stable DC-link voltage during the FES system charging/discharging, with their corresponding to ESM in generating/driving state. Please note that we exchange the closed-loop reference and feedback signal, for a negative isq * must be given in the ESM generating state, and thus a negative power is consuming by the ESM, which is actually supplying power. Fig. 4 shows the vector graph of the ESM in generating state with unity-power factor. In this case, the current and voltage vector of the ESM are in opposite phase. Neglecting the stator leakage inductance Lls, \u03b4sm is the difference between the stator flux linkage angle \u03b8s and the rotor flux linkage angle \u03b8r. \u03b8r can be obtained by the encoder coupled on the rotor shaft. \u03b8s can be calculated through the ESM voltage model, which is shown as follows: ( ) ( ) s s s s s s s s u R i u R i \u03b1 \u03b1 \u03b1 \u03b2 \u03b2 \u03b2 \u23a7\u03a8 = \u2212\u23aa \u23a8 \u03a8 = \u2212\u23aa\u23a9 \u222b \u222b (6) Based on the vector relations as shown in Fig. 4, the ESM phase voltage and current vector can be expressed as follows: ( )s r sq sq s r sd sd s sd sq u L i j e L i i i ji \u03c9 \u03c9= \u2212 + +\u23a7\u23aa \u23a8 = +\u23aa\u23a9 (7) As us and is are in opposite phase, we can get: r sq sqsd sq s r sd sd L ii i e L i \u03c9 \u03c9 \u2212 = + (8) The reference rotor field voltage es * is as follow: 2 2 r sd sd r sq sq* s * sd L i L i e i \u03c9 \u03c9+ = \u2212 (9) The basic DTC scheme for ESM is shown in Fig. 5. The error between the reference torque Te * and the estimated torque Te is the input of a three level hysteresis comparator, and the error between the reference stator flux magnitude \u03a8s * and the estimated stator flux magnitude \u03a8s is the input of a two level hysteresis comparator" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003177_2014-01-0379-FigureA11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003177_2014-01-0379-FigureA11-1.png", + "caption": "Figure A11. Transmission X-member", + "texts": [], + "surrounding_texts": [ + "1. ZIENKIEWICZ, O.C., \u201cThe Finite Element Method\u201d. 3rd ed., McGraw-Hill Book Company, NY, 1979. 2. BATHE, K.J., \u201cFinite Element Procedures in Engineering Analysis\u201d, Prentice-Hall Inc., New Jersy, 1982. 3. BANERJEE, P.K. and BUTTERFIELD, R., Boundary Element Methods in Engineering Science, (McGraw-Hill Book Company, UK), 1981. 4. MALLIKARJUNA and KANT T., \u201cA general fibre reinforced composite shell element based on a refined shear deformation theory\u201d, International Journal of Computers & Structures, Vol.42(3), 381-388, 1992. 5. MALLIKARJUNA and Kant T., \u201cFinite element transient response of composite and sandwich plates with a refined theory\u201d, International Journal of Applied Mechanics American Society of Mechanical Engineers - Vol. 57, No. 4, Pages: 1084-1086, (1990). 6. MALLIKARJUNA and KANT T., \u201cTransient response of isotropic, orthotropic and anisotropic composite-sandwich shells with the superparametric element\u201d, International Journal of Applied Finite Elements and Computer Aided Engineering - Finite Elements in Analysis & Design, Vol.12(1), 63-73, 1992. 7. MALLIKARJUNA and KANT T., \u201cFree vibration of symmetrically laminated plates using a higher order theory with finite element technique\u201d, International Journal for Numeri. Methods in Engg, Vol.28(8), 1875-1889, 1989. 8. MALLIKARJUNA and KANT T., \u201cEffect of cross-sectional warping of anisotropic sandwich laminates due to dynamic loads using a refined theory and C\u00b0 finite elements\u201d, International Journal for Numerical Methods in Engineering, Vol.35(10), 2031-2047, 1992. 9. MALLIKARJUNA and KANT T., \u201cDynamics of laminated composite plates with a higher order theory and finite element discretization\u201d, Journal of Sound and Vibration. Vol. 126(3), 463-475, 1988. 10. Bennur, M., \u201cSuperelement, Component Mode Synthesis, and Automated Multilevel Substructuring for Rapid Vehicle Development,\u201d SAE Int. J. Passeng. Cars - Mech. Syst. 1(1):268- 279, 2008, doi:10.4271/2008-01-0287. 11. Bennur, M., \u201cVehicle Acoustic Sensitivity Performance Using Virtual Engineering,\u201d SAE Technical Paper 2011-01-1072, 2011, doi:10.4271/2011-01-1072. 12. Bennur, M., Posthuma, D., and Lewitzke, C., \u201cHigh Performance Vehicle Chassis Structure for NVH Reduction,\u201d SAE Technical Paper 2006-01-0708, 2006, doi:10.4271/2006-01-0708. 13. Bennur, M. and Weiss, L., \u201cHybrid Technique Based on Finite Element and Experimental Data for Automotive Applications,\u201d SAE Technical Paper 2007-01-0466, 2007, doi:10.4271/2007-01-0466. 14. Bennur, M., Hogland, D., Abboud, E., Wang, T. et al., \u201cMultiDisciplinary Robust Optimization for Performances of Noise & Vibration and Impact Hardness & Memory Shake,\u201d SAE Technical Paper 2009-01-0341, 2009, doi:10.4271/2009-01-0341. 15. OPTISTRUCT, Altair Engineering, Inc., Troy, Michigan. 16. VIRTUAL LAB, LMS International, Leuven, Belgium. 17. Wyckaert, K. and Van der Auweraer, H., \u201cOperational Analysis, Transfer Path Analysis, Modal Analysis: Tools to Understand Road Noise Problems in Cars,\u201d SAE Technical Paper 951251, 1995, doi:10.4271/951251. 18. MSC Nastran, MSC. Software Corp., Santa Ana, CA. 19. BENNIGHOFJ.K. and Kim C.K., \u201cAn adaptive multilevel substructuring method for efficient modeling of complex structures\u201d, Proceedings of the AIAA, 39TH SDM Conference, 33rd SDM Conference, 1631-1639, Dallas, TX 20. BENNIGHOFJ.K. and KAPLAN M.F., \u201cFrequency sweep analysis using multi-level substructuring, global modes and iteration\u201d, Proceedings of the AIAA, 39TH SDM Conference, Long Beach, CA, 1998. 21. BENNIGHOF J.K., KAPLAN M.F., MULLER M.B., and KIM M., \u201cMeeting the NVH computational challenge: automated multilevel substructuring\u201d, Proceedings of the International Modal Analysis Conference XVIII, San Antonio, Texas, 909-915, 2000 22. Bennighof, J., Kaplan, M., Kim, M., Kim, C. et al., \u201cImplementing Automated Multi-Level Substructuring In Nastran Vibroacoustic Analysis,\u201d SAE Technical Paper 2001-01-1405, 2001, doi:10.4271/2001-01-1405. 23. BENNIGHOF J.K. and LEHOUCQR.B., \u201cAn automated multilevel substructuring method for the eigenspace computation in linear elastodynamics\u201d, SIAM Journal Sci. Comput., 25, 2084-2106, 2004. 24. http://boronextrication.com/tag/cadillac/ CONTACT INFORMATION Mallikarjuna Bennur, Ph.D., P.E., MBA General Motors Co. 3300, General Motors Road Milford, Michigan 48380, U.S.A. mallikarjuna.bennur@gm.com www.gm.com" + ] + }, + { + "image_filename": "designv11_84_0003668_10402004.2010.542278-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003668_10402004.2010.542278-Figure1-1.png", + "caption": "Fig. 1\u2014Experimental apparatus for measurement of the static coefficient of friction.", + "texts": [ + " Therefore, most studies utilize visual inspection or other indirect methods to determine when displacement begins and therefore at what point the static friction coefficient should be calculated. To address these issues, a test apparatus was recently developed to measure static friction at high loads with multiple contact shapes and is able to precisely measure displacement of the test specimens (Garcia, et al. (17)). The application of the apparatus to characterize static friction of lubricated metallic line contacts is presented here. The primary focus of this work is to characterize the effect of lubricant viscosity on start-up friction. Figure 1 is a schematic of the test apparatus used for measuring static friction between two cylinders and a flat surface. A detailed description of the test apparatus can be found in a previous article (Garcia, et al. (17)). The clamping mechanism selected for this experiment allows testing at normal loads from 0 to 4,000 N. The flat test plate was made of AISI 304 stainless steel and was 152.4 mm long by 101.6 mm wide. The thickness of the plate was 3.17 mm, and the mass of the plate and fixture used to attach it to the cylinder was 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003435_978-94-017-7300-3_16-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003435_978-94-017-7300-3_16-Figure4-1.png", + "caption": "Fig. 4 Helical shapes driven by residual stress in one layer simulated by finite element method. The residual stress in the top layer is \u03c3t = \u03c30[e1 \u2297 e1 cos2 \u03c6 + (e1 \u2297 e2 + e2 \u2297 e1) cos\u03c6 sin\u03c6]. Here, \u03bd = 0.3, W = 0.02 m, H = 0.005 m and \u03c30 = 10 GPa. (a) A twisted shape forms when \u03c6 = 61.3\u00b0, \u03bd = 0.3, W = 0.02 m, and H = 0.005 m. (b) A twisted shape forms when \u03c6 = 65.5\u00b0, \u03bd = 0.3, W = 0.02 m, and H = 0.005 m. (c) A cylindrical helical shape forms when \u03c6 = 65.5\u00b0, \u03bd = 0. Young\u2019s modulus is E = 100 GPa. The color indicates the total displacement", + "texts": [ + " 3(b)). More specifically, when \u03c30 = 10 GPa, the ratio between the principal curvature is \u03ba2/\u03ba1 \u2248 \u2212(1/14.9)/(1/4.2) \u2248 \u22120.28. The absolute value of this ratio is very close to Poisson\u2019s ratio \u03bd = 0.3, supporting the hypothesis that Poisson\u2019s effect is the main cause of such deformation. Our previous studies [22, 28, 34] have shown that a purely twisted ribbon forms when \u03ba1 cos2 \u03c6 + \u03ba2 sin2 \u03c6 = 0. So here we predict that a purely twisted shape occurs when cot2 \u03c6 = \u2212\u03ba2/\u03ba1 \u2248 \u03bd = 0.3, i.e., \u03c6 \u2248 61.3\u00b0. Figure 4(a) shows the formation of a nearly twisted bilayer ribbon of width W = 0.02 m and thickness H1 = H2 = H = 0.005 m, when the mis-orientation angle is \u03c6 \u2248 61.3\u00b0 and the only non-zero principal component of the Reprinted from the journal 326 residual stress is \u03c30 = 10 GPa along the e1 direction. Through trial-and-error, we found that when \u03c6 \u2248 65.5\u00b0, the shape of the deformed bilayer ribbon is even closer to a purely twisted ribbon (the centerline stays straight, i.e., the helix radius is zero), as shown in Fig. 4(b). By contrast, when Poisson\u2019s effect is absent, the same driving force results in a cylindrical helical ribbon (Fig. 4(c)). 327 Reprinted from the journal From a theoretical perspective, it should hold that |\u03ba2/\u03ba1| = \u03bd [22] in the limit when the width vanishes (i.e., W \u2192 0), as shown by the blue line in Fig. 5. When the ribbon has a finite width, the theoretical values of the principal curvatures can also be numerically calculated [39]. Figure 5 shows the absolute value of the ratio between the principal curvatures (|\u03ba2/\u03ba1|) as a function of the Poisson\u2019s ratio for different width values. In the small width regime (e", + " It is also worth mentioning that the finite element simulations provide convenient access to the strain components as well. In particular, the strain components along the three axes are found to be a linear function of the residual stress as expected. From the plane stress condition, it should hold that \u03b533 = \u2212\u03bd(\u03b511 + \u03b522)/(1 \u2212 \u03bd) [32]. The numerical simulation results indeed show good agreement with this prediction (Fig. 9(c)). It is interesting to note that a shape transition can occur from a purely twisted shape (Fig. 4(a), (b)) to a nearly cylindrical helical shape when Poisson\u2019s ratio changes (Fig. 4(c)) 329 Reprinted from the journal or when the strip becomes wide and thin enough (Fig. 5(a), (b)), even though all the other physical parameters stay the same. The main cause of the latter case is the competition between bending energy (\u223c \u03ba2EWH 3) and stretching energy (\u223c \u03ba4EW 5H ), as has been discussed in recent studies [26, 29, 34, 39]. On the other hand, when the driving force (i.e., residual stress) becomes strong enough, such transition can also occur, because the principal curvatures will become large enough" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003666_robio.2014.7090503-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003666_robio.2014.7090503-Figure1-1.png", + "caption": "Fig. 1. Diagram of the lander, showing the inertial reference frame ( ex, ez), the velocity vector V , the Focus of Expansion (FoE), and the mean thruster force uth and its projections in the Local Vertical Local Horizontal (LVLH) reference frame. \u03c990 and \u03c9135 are presented in red on the lunar ground. Adapted from [16].", + "texts": [], + "surrounding_texts": [ + "Here it is proposed to study autonomous landing during the approach phase extending from the High Gate (HG) -1800m AGL- to the Low Gate (LG) -10m AGL. The mass optimization problem was defined here along with the constraints involved, and its solution was computed in terms of the trajectory and the OF profiles. In order to meet the low computational requirements, the optimal problem was solved offline only once: the OF and pitch profiles were determined and implemented in the form of constant vectors in the lander. Therefore, the guidance strategy is said to be sub-optimal since the offline computed optimal trajectory correspond to the nominal initial conditions which may not be met at the HG. First of all, the optimal control sequence u\u2217 =( u\u2217 th, u\u2217 \u03b8 ) was computed, taking u\u2217 th to denote the braking thrust and u\u2217 \u03b8 to denote the pitch torque (the upper script \u2217 indicates the optimality in terms of the mass, i.e., the fuel consumption). In this paper, optimality refers to the outputs of the optimization problem ( u\u2217 th, u\u2217 \u03b8 ) and the associated reference trajectory ( V\u0307 \u2217 x , V\u0307 \u2217 z , V \u2217 x , V \u2217 z , h\u2217, \u03b8\u2217 ) . Looking for the least fuel-consuming trajectory is equivalent to finding the control sequence u\u2217 that minimizes the use of the control signal (see (1f)). The optimization problem can then be expressed as follows: Solve min uth(t),u\u03b8(t) \u222b tf t0 (uth(t) + |u\u03b8(t)|) dt (7) Subject to Equations (1a)-(1f)\u23a7\u23aa\u23aa\u23aa\u23a8 \u23aa\u23aa\u23aa\u23a9 Vz(t0) = \u221236 m/s, \u2223\u2223Vzf \u2223\u2223 < 1 m/s Vx(t0) = 69 m/s, \u2223\u2223Vxf \u2223\u2223 < 1 m/s h(t0) = 1800 m, hf = 10 m \u03b8(t0) = \u221261\u25e6, |\u03b8f | < 2\u25e6 (8) \u23a7\u23aa\u23aa\u23aa\u23aa\u23a8 \u23aa\u23aa\u23aa\u23aa\u23a9 0 < uth < 3438 N \u221244 < upitch < 44 N \u2200t \u2208 [t0, tf ] (\u2212Vz, Vx, h, x) > 0\u2223\u2223\u2223\u03b8\u0307\u2223\u2223\u2223 < 1.5\u25e6/s (9) This offline sub-optimal guidance strategy was implemented using Matlab optimization software on the nonlinear system under constraints to bring the system from HG to LG. To solve this continuous time optimization problem, many freely available Matlab toolboxes based on various methods can be used. The solution provided by ICLOCS (Imperial College London Optimal Control Software, [20]) based on the IPOPT solver suited our needs for the numerical implementation of a nonlinear optimization problem in the case of the continuous system subjected to boundary and state constraints using the interior point method. The simulation of the open loop under optimal control was therefore run on the nonlinear system to assess the optimal OF and pitch profiles ( \u03c9\u2217 x, \u03c9\u2217 z , \u03b8\u2217 ) . Equation (1a)-(1f) describes the dynamic lander, (8) gives the initial and final conditions and (9) gives the actuator and system constraints imposed along the trajectory. For safety reasons, a 10% margin was added to the thrusters\u2019 physical saturation in order to give the lander greater maneuverability around the predefined trajectory at any point. It is worth noting that a terminal constraint could easily be added if required to the downrange x to make pinpoint landing possible, but this might greatly increase the fuel consumption. Since the case may arise where \u03b8\u0307 = \u2212\u03c9R > \u03c9T and thus \u03c9measured < 0, we had to use a bi-directional version of the 6-pixel VMS adapted for use in the following measurement range: \u03c9measured\u03b5 [\u221220\u25e6/s; \u22120, 1\u25e6/s] \u222a [0, 1\u25e6/s; 20\u25e6/s]. The fuel expenditure decreases the lander\u2019s mass by \u0394m, which is defined as the difference between the initial and final mass of the lander \u0394m = mldr0 \u2212mldr(tf ) where mldr0 = 762 kg and mldr(tf ) = mldr(t0)\u2212 1 gEarth \u222b tf t0 ( uth(\u03b5) Ispth + |u\u03b8(\u03b5)| Isp\u03b8 ) d\u03b5 (10) In order to make sure that the sum \u03c9grd\u2212trh = \u03c9T +\u03c9R does not cancel itself out (i.e. \u03c9T = \u2212\u03c9R), the pitch rate (\u03c9R = \u03b8\u0307) was constrained as follows: \u2223\u2223\u2223\u03b8\u0307\u2223\u2223\u2223 = |\u03c9R| < 1.5\u25e6/s. Under all these conditions, the optimal control sequences (u\u2217 th, u \u2217 \u03b8) were processed: the optimal solution was obtained with tf = 51.46s and a mass change of \u0394m < 33.6 kg (amounting to 4.4% of the initial mass). The trajectory modeled under these constraints can be said to be optimal in the case of a more highly constrained problem. Additional constraints were imposed on \u03b8\u0307 and the 10% margin on the thrust to account for the sensors\u2019 and actuators\u2019 operating ranges, resulting in a more highly constrained problem than the system can actually deal with. In any case, both of these constraints (the saturated pitch rate and the 10% margin added to the thrust) resulted in very similar fuel expenditure predictions to that obtained without these additional constraints (amounting to a difference of only 0.21%)." + ] + }, + { + "image_filename": "designv11_84_0002978_ijvp.2013.058277-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002978_ijvp.2013.058277-Figure1-1.png", + "caption": "Figure 1 GM4DTS US Patent 5,836,850 transmission scheme", + "texts": [ + " In this paper, a mathematical model is established with an open-loop control module which can explain the complex nature of the clutch/brake coordination during the DTSs, and all possible key parameters which can affect shift qualities are studied. This research would also serve as the background work for the future closed-loop control strategy development. The simulation results at different throttle positions are investigated to validate the proposed open-loop control algorithm. The transmission system to be studied in this paper is based on US Patent 5,836,850 as shown in Figure 1 (GM4DTS refers to GM gearbox, 4 speeds, DTS transmission). A powertrain consists of the engine and torque converter (element 12), a final drive (element 14), and a planetary gear train (element 16). The planetary gear train 16 consists of a simple planetary gearset 18 and a compound planetary gearset 20. If the gear ratio of the ring gear 24 and the sun gear 22 is equal to 1.54, and the ratio of the ring gear 42 and the sun gear 40 is equal to 1.714, the drive ratios can be established as: 1st gear ratio 4" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002883_s00542-011-1333-8-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002883_s00542-011-1333-8-Figure3-1.png", + "caption": "Fig. 3 Absorbance measurement setup. Absorbance measurements were carried out with the sample being enclosed by a flow cell. An indicator solution provided both the substrate for the enzyme and a dye sensitive to the reaction products. The change in color intensity of the dye was recorded with the spectrophotometer", + "texts": [ + " Enzyme immobilization was achieved by a 30 min room temperature incubation of small PR samples with 20 ll solution of LOx in phosphate buffered saline (PBS, pH = 7.4) with 25 wt.% poly(ethyleneimine) (PEI) for stabilization of the enzyme on the PR surface (Teramoto et al. 1992; Lin et al. 1996). The concentration of LOx in PEI/PBS was varied between 0 mg/ml (no LOx) and 10 mg/ml, respectively. The samples were then washed with DI water and PBS, and blown dry with N2. We determined the activity of immobilized LOx at room temperature by reflective absorbance spectrometry using the flow cell setup shown in Fig. 3. The samples sat in a 1.6 mm thick cell into which an indicator solution was drawn by a peristaltic pump (Alitea CX-Y, USA). The indicator solution was comprised of 1.0 M sodium L-lactate (enzyme substrate; Fluka, USA), 40 mM 4-aminoantipyrene (hydrogen peroxide indicator; Sigma-Aldrich, USA), 100 mM sodium-4-hydroxybenzoate (Sigma-Aldrich, USA), and 20 U/ml horseradish peroxidase (SigmaAldrich, USA). The color intensity (i.e. absorbance) of the indicator solution increases with rising hydrogen peroxide levels in the chamber, generated as sodium lactate is enzymatically converted by LOx", + " The diameter of the counter electrodes was *1 mm with the width of the metal lines being *75 lm. The width of the working electrode and the spacing between features were *50 lm, respectively. Thereafter, the NR 71 was spin-coated onto the fabricated structures and patterned to expose the electrodes. The sensors were incubated with 20 ll 10 mg/ml LOx in PEI/PBS for 30 min at room temperature, washed and blown dry with N2. We characterized the sensor response to sodium L-lactate in a flow cell setup similar to Fig. 3. A BASi LC epsilon potentiostat (USA) was used to set the working electrode potential (400 mV vs. reference electrode), and to record the current response. The test solutions contained sodium L-lactate in PBS at concentrations ranging from 0 to 5 mM. Before evaluation of the sensing properties, the electrodes were stressed with a potential of 700 mV versus the reference electrode to stabilize their response (Shum et al. 2009). 3.2 Sensor response The operation of the amperometric sensor is based on the generation of hydrogen peroxide supported by the catalytic action of LOx on sodium L-lactate: l-Lactate \u00fe O2 ", + " Hence, we prove the versatility of the immobilization procedure. Furthermore, enzyme functionality was successfully integrated into a biosensing structure based on patterned PR on a SiO2 substrate. We therefore established a simple, but versatile method allowing for the inclusion of enzyme functionality into novel MEMS based devices in a cost-effective manner using already existing microfabrication schemes. Acknowledgments The authors would like to thank S. McQuaide for contribution of the schematic in Fig. 3, and A. Lingley for helpful discussions about the sensor setup. In addition, the authors would like to thank A. Shum and M. Cowan for the preliminary sensor design. Furthermore, we thank the National Science Foundation for the financial support of this work. Conflict of interest The authors declare that they have no conflict of interest. Adams JB (1991) Review: enzyme deactivation during heat process- ing of food-stuffs. Int J Food Sci Techn 26(1):1\u201320. doi: 10.1111/j.1365-2621.1991.tb01136.x Albareda-Sirvent M, Merkoc\u0327i A, Alegret S (2000) Configurations used in the design of screen-printed enzymatic biosensors" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003026_amm.401-403.254-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003026_amm.401-403.254-Figure4-1.png", + "caption": "Figure 4 shows, the radial load Fr of bearing is FGFr += (23) Amplitude of exciting force F is not change, but direction of F is changing; Amplitude and direction of shafting gravity G are constant. So, the amplitude and direction of Fr is changing. Therefore,", + "texts": [ + " (2)Vibration model of rolling element bearing on vibrating screen with single pitting fault Vibrating screen rolling element bearing outer ring fixed on the bearing seat, it can be thought as a rigid support. And the vibration of bearing seat itself subjected by driving of exciting force. Inner ring of rolling element bearing is fixed on the rotating shaft, inner ring of rolling element bearing is fixed on the rotating shaft, it rotate together with the shaft. When pitting corrosion occurred on the outer ring, as Figure 4 shows, amplitude and direction of Fr is changing along with the rotation of the rotating shaft. Relative position of the outer ring pitting and Fr is changing, and relative position of the outer ring pitting and G is not change. In Eq. 24, \u03b8\u2014angle of Fr and G, \u03c6\u2014angle of F and G; From Eq. 24 we can know = ++= ) sin arcsin( cos222 r r F F GFFGF \u03d5\u03b8 \u03d5 (25) From Eq. 11 we can know )( )( )(max \u03b5 \u03d5\u03d5 r r ZJ F Q = (26) From Eq. 12 we can know \u03b1 \u03d5\u03d5 cos )(37.4 )(max Z F Q r= (27) From Eq. 13 we can know \u03b1 \u03d5\u03d5 cos )(5 )(max Z F Q r= (28) From Eq" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003488_amm.592-594.565-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003488_amm.592-594.565-Figure5-1.png", + "caption": "Fig. 5: 3D temperature field distribution at 33sec", + "texts": [ + " Distortion values are measured at some specific points, for which specific grids are made on the plates. Thermo-Metallurgical Results. The weld bead geometry obtained during simulation and experiment has been compared as shown in the fig. 4a and fig. 4b. The weld bead at 15mm and 65mm away from the start point has been given below. The weld bead profiles generated in simulation is found to match with that obtained in experiment. Thermal cycle results. A thermal simulation result provides complete thermal results in all the points of the work piece. Fig. 5 shows the 3D temperature field distribution at 33sec. It shows that the isothermal line presents an ellipse and the isothermal line in front are dense and that in back are dilute. Due to heat transfer phenomena, a tail end to the welding heat source can be observed. The simulated and experimental thermal cycles at three different positions are shown in the fig. 6a, 6b and 6c. A good agreement is found between experimental and simulated results. The small disagreement between peak temperature , however, may be attributed to some inaccuracies in the heat flux distribution, assumed constants in the model, some imprecision in the thermocouple locations with respect to weld line, mesh quality and simplification assumed during numerical solving" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003322_ijsurfse.2014.065818-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003322_ijsurfse.2014.065818-Figure4-1.png", + "caption": "Figure 4 Elastic shakedown and dynamic and kinematic hardening (see online version for colours)", + "texts": [ + " ( ) ( ) 2 2 2 1 2 2 2 2 2 2 2 1 2 ( ) ( ) tan tan tan( ) = \u2212 + = + + = + = \u2212 = + = r x b z r x b z r x z z\u03b8 x b z\u03b8 x b z\u03b8 x (5) Given the angles and distances defined by the equation (5), the solution to the stress field due to the triangular distribution is given by the following expressions (Johnson, 1985): ( ) { } { } ( ) { } 1 2 1 2 2 1 2 2 1 22 1 1 2 1 2 1 2 1 2 ( ) ( ) 2 2 ln 2 ln 2 ln 3 2 ( ) ( ) 2 2 2 ( ) ( ) 2 kk x k k kk z k k k y x z k kk xz p r r\u03c3 x b \u03b8 x b \u03b8 \u03b8x z \u03c0b r \u03c4 r r rx b z \u03b8 \u03b8 \u03b8 \u03c0b r r p \u03c4 z\u03c3 x b \u03b8 x b \u03b8 \u03b8x \u03b8 \u03b8 \u03b8 \u03c0b \u03c0b \u03c3 v \u03c3 \u03c3 p z \u03c4\u03c4 \u03b8 \u03b8 \u03b8 x b \u03b8 x b \u03b8 \u03b8x \u03c0b \u03c0b \u23a7 \u23ab\u239b \u239e= \u2212 + + \u2212 +\u23a8 \u23ac\u239c \u239f \u239d \u23a0\u23a9 \u23ad \u23a7 \u23ab\u239b \u239e \u239b \u239e+ + \u2212 + \u2212\u239c \u239f\u23a8 \u23ac\u239c \u239f\u239d \u23a0 \u239d \u23a0\u23a9 \u23ad = \u2212 + + \u2212 \u2212 + \u2212 = + = \u2212 + \u2212 + \u2212 + + \u2212 + 1 2 2 2 ln r rz r \u23a7 \u23ab\u239b \u239e \u23a8 \u23ac\u239c \u239f \u239d \u23a0\u23a9 \u23ad (6) To find the macroscopic stresses in point A, the contributions of each pk(x) and \u03c4k(x) are added together by applying the principle of superposition of the n components. 1 1 1 1 ; ; ; = = = = = = = =\u2211 \u2211 \u2211 \u2211 n n n n k k k k x x y y z z xz xz k k k k \u03c3 \u03c3 \u03c3 \u03c3 \u03c3 \u03c3 \u03c4 \u03c4 (7) Taking the macroscopic stresses, we then proceed to calculate the mesoscopic stresses (\u03a3), that are related to the macroscopic stresses (\u03c3) through the residual mesoscopic stress tensor (\u03c8), according to Dang Van hypothesis (Fajdiga et al., 2007; Constantinescu et al., 2003). \u03a3 = +\u03c3 \u03c8 (8) Figure 4 illustrates a series of stress-strain cycles (\u03c3 \u2013 \u03b5) where the elastic limit (\u03c3y) is exceeded. The material enters the plastic field undergoing a hardening that is both dynamic and kinematic (Zubizarreta and Ros, 2005; Colak, 2008). After a number of cycles, this leads to a stabilised pure elastic cycle (between the minimum and the maximum stress values: \u03c3min and \u03c3max), represented by the line B\u201d\u2013C\u201d. This phenomenon is known as elastic shakedown and involves a displacement (\u2013\u03c8) of the material yield surface and an increase in the isotropic hardening parameter from K0 up to K, as reflected in Figure 4. Based on the concept of elastic shakedown, Ponter (Johnson, 1985; Conrado et al., 2011) puts forward a theorem to find the mesoscopic stress reflected in the equation (9), where f is the surface or yield criterion using a safety coefficient m. ( ) 2( , ) ( ) 0+ \u2212 \u2264f m\u03c3 x t \u03c8 x K (9) If the Von Mises criterion is used to define the yield surface, the equation (9) is transformed into (10), which represents a six-dimensional hypersphere. 2 2 2 2 2 2 2 2 2 0; or equivalently: 0 2 + +\u239b \u239e \u2212 = + + + \u2212 =\u239c \u239f \u239d \u23a0 x y z xy xz yz s s s J K s s s K (10) where J2 is the second invariant of the deviatoric component of the mesoscopic stress tensor, s" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure6.10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure6.10-1.png", + "caption": "Fig. 6.10 Types of structural loads", + "texts": [], + "surrounding_texts": [ + "the formal exploration given by the algorithms yielded by the explained parameters and reduce the objects mass through the same algorithms that will form the connections between the green areas to be preserved, see Fig. 6.12. 7. Once the manufacturing methods are established it is important to highlight that the generative design it\u2019s not only to be used with additive manufacturing since there are other methods such as subtractive numerical control with 2 to 3 axis in which parameters could be established for a cutter, see Fig. 6.13. 6 Numerical Simulation of Cranial Distractor Components \u2026 171" + ] + }, + { + "image_filename": "designv11_84_0001952_robio.2011.6181704-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001952_robio.2011.6181704-Figure3-1.png", + "caption": "Fig. 3. The 3 joints of LAURON\u2019s legs", + "texts": [ + " This paper will illustrate how the task of controlling LAURON by applying forces is realized: first the motor currents are measured in all joints and the corresponding joint torques are derived. These 18 individual torques are transformed into a 6D force and torque vector. Finally a classifier estimates the intention of the human operator by observing the course of the forces and torques over a short period of time and triggers corresponding motion behaviors. The six-legged walking robot LAURON (Fig. 2) has three joints in each leg (see Fig. 3) summing up to 18 active joints and reaches a weight of about 27 kg including its accumulators. The morphology was derived from the stick insect Carausius morosus ([4]). Custom made UCoM control units (UCoM: Universal Control Module [6]) control the motors using PID controllers and read various sensors such as joint angle sensors, motor encoders and ground contact sensors. The perception capabilities are extended by a camera systems, an IMU and a ToF-Camera. A behavior-based control system enables LAURON to walk autonomously in unstructured and rough terrain (see [2])", + " In the second experiment we have illustrated the effect of adding 15kg payload to the robot, which is shown in the bottom left of Fig. 5. The error of the developed on-board motor current measurement is shown in Fig. 5 (bottom right). This figure also includes the on-board measured motor current and the reference current. For more details on the current measurement and the result for the whole system please refer to the dedicated publication of A.Roennau [5]. Each leg of Lauron IVc is driven by three 12V dc motors each with a mounted gearbox ontop. The alpha joint (see Fig. 3) is driven directly with a bevel gear the other two via drive belts. As dc motors have a known constant correlation (Km) between input current (Imot) and output torque, the resulting motor torque (Mmotor) for each joint can easily be calculated when the input current is known. Because the motor needs to overcome its own friction losses, a constant factor (I0) is subtracted from the measured current. Mmotor := (Imot \u2212 I0) \u2217 Km (1) To calculate the final joint torque (Mjoint) the motor torque is multiplied by the transfer factor (Tjoint) which is the product of the drive belt or bevel gear transfer factor and the gear transfer factor of the motor", + " J(CJ) = ( j1, ..., jn) (13) The resulting Jacobian matrix J(CJ) has 6 rows (for the 3- dimensional case) and n columns. Obviously the complexity of this construction rises linear with the number of joints. In the following we will create the Jacobian for an exemplary joint of an exemplary joint configuration CJ of LAURON as sketched in Fig. 7. We need one of these Jacobian matrixes for each leg of the robot. These are defined for the kinematic chain running from the hip (the alpha joint - see Fig. 3) to the footpoint of the same leg. To include the 3 degrees of freedom of the leg we need 3 columns of the Jacobian. First of all the vectors pi and zi are created corresponding to the sketch from Fig. 7 which then in turn give us the columns of H F J(CJ). z0 = \u239b \u239d 0 0 1 \u239e \u23a0 p0 = \u239b \u239d 0 y0 \u2212z0 \u239e \u23a0 z1 = \u239b \u239d 1 0 0 \u239e \u23a0 p1 = \u239b \u239d 0 y1 \u2212z1 \u239e \u23a0 z2 = \u239b \u239d 1 0 0 \u239e \u23a0 p2 = \u239b \u239d 0 y2 \u2212z2 \u239e \u23a0 z \u00d7 p0 = \u239b \u239d \u2212y0 0 0 \u239e \u23a0 z \u00d7 p1 = \u239b \u239d 0 z1 y1 \u239e \u23a0 z \u00d7 p2 = \u239b \u239d 0 z2 y2 \u239e \u23a0 (14) H F J(CJ) = \u239b \u239c\u239c\u239c\u239c\u239c\u239c\u239d \u2212y0 0 0 0 z1 z2 0 y1 y2 0 1 1 0 0 0 1 0 0 \u239e \u239f\u239f\u239f\u239f\u239f\u239f\u23a0 (15) By adding the individual 6D force and torque vectors acquired from the six individual limbs, while taking their corresponding point of application into account, we gain the resulting 6D force and torque vector for the robot\u2019s body" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001946_amm.698.552-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001946_amm.698.552-Figure3-1.png", + "caption": "Fig. 3. Bricard\u2019s linkage chain", + "texts": [ + " The link turning the direction vector clockwise through \u03b1 degree is (-\u03b1). If the rotation is counterclockwise, the link is (+\u03b1). In Brikard's linkage each of the six links turns the direction vector through 90 \u00b0 clockwise therefore the mechanism can be designated as (-90-90-90-90-90-90). Links ABand BC were joined so that the direction vector of the B hinge as a part of the AB link coincided with the direction vector of the B hinge as a part of the AB link. When the mechanism is open, the links built in one line form the chain AB-BC-CD-DE-ED-FA\u2032 shown in Fig. 3. The direction vector turns through 90\u00b0 clockwise when passing each hinge from A to A \u2032. Closing the chain by aligning hinges A and A\u2032 forms Bricard\u2019s mono-mobile linkage (Fig. 1). The mechanism is assemblable if its chain (Fig. 3) can be closed at least for one value of the driving link rotation angle. The mechanism is mobile if its chain can be closed for a continuous range of driving link rotation angles. The mechanism has a crank if its chain can be closed for any value of the driving link rotation angle. It has been shown above that Brikard's linkage (-90-90-90-90-90-90) is assemblable and mobile. Within this work Brikard's linkage modifications (\u03b11 \u03b12 \u03b13 \u03b14 \u03b15 \u03b16) assemblability condition is defined. Let us arrange the chain of Brikard\u2019s linkage (Fig. 3) so that the assembly direction of the mechanism has the same direction as the view vector. The result is a projection of the direction vectors of Brikard's linkage hinges (Fig. 4). In the mechanism (-90-90-90-90-90-90), direction vectors of hinges A and A\u2032 are collinear. The modification of Brikard\u2019s linkage (\u03b11\u2022\u03b12\u2022\u03b13\u2022\u03b14\u2022\u03b15\u2022\u03b16) is assemblable, if direction vectors of the first and last hinges in the chain are collinear. The mechanism (-90-80-70-100-120-150) and the projection of its direction vectors is shown in Fig 5" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001497_978-3-319-07572-3_6-Figure6.3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001497_978-3-319-07572-3_6-Figure6.3-1.png", + "caption": "Fig. 6.3", + "texts": [ + " Some materials represent better resistance than others. For instance, compressive and tensile strengths of steel are the same, whereas concrete is strong in compression and weak in tension. Toughness Toughness is the resistance of material to high forces without breaking. As we discussed earlier, tensile and compressive internal forces develop in the structural members when they are under the axial loads. The internal forces will distribute over the entire cross section and create normal stresses in the cross section of the member. Figure 6.3 shows the force distribution on the cross section of the bar directed to the left when the force F is directed to the right. When the member is under tension, the stress is called tensile stress. If the member is under compression, then the stress is called compressive stress. Tensile and compressive stresses are called normal stresses, since the force F and the force distribution are both perpendicular to the cross section of the member (Fig. 6.4). Normal stress formula \u03c3\u00bcF/A discussed earlier, can be applied for tensile or compressive stresses conditionally the force F is centrally applied to the cross section of the member" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001826_ecce.2015.7310634-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001826_ecce.2015.7310634-Figure8-1.png", + "caption": "Fig. 8. IPM-type model and equivalent magnetizing distribution modeling.", + "texts": [ + " EXPERIMENTAL RESULT In this study, we performed an analysis for the case with the N-pole 50% demagnetization model for different N- and S-pole demagnetization ratios. Therefore, we determined that the presence of 2nd and 4th harmonics enabled irreversible demagnetization detection of the PM. Therefore, we performed an experiment with a test motor to verify that the proposed detection technique can be used in BLDCM applications. Table I gives specifications of the test motor used in the analysis. The experiment was performed under the same condition as in the simulation with 3,300 rpm. Fig. 8 is the normal condition model used in the experiment. The cross section of the test motor is shown in Fig. 9. Because the test motor is small and it is very difficult for its demagnetized condition to be realized using the division method of the PM, the demagnetized magnet portion is fabricated in air, and has a step shape in order to realize the N-pole 50% demagnetization condition. We compared the simulation result of the rotor transformation model and magnet division model in order to justify the use of the rotor transformation model" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001502_intmag.2015.7156611-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001502_intmag.2015.7156611-Figure2-1.png", + "caption": "Fig. 2. Optimal results with skewing operation. (a) Flux distributions without skewed rotor. (b) Flux distributions with skewed rotor. (c) Back-EMF waveforms. (d) Cogging torque waveforms.", + "texts": [], + "surrounding_texts": [ + "INTERMAG 2015\nCogging Torque Optimization of a Novel Transverse Flux Permanent Magnet Generator with Double C-hoop Stator for Wind Power Application. Z. JIA, H . LIN Southeast University, Nanjing\nI . INTRODUCTION Transverse flux permanent magnet generator (TFPMG) is especially suitable for wind power application for the merits of large pole numbers, decoupled magnetic circuit, and high power density . The distinguishing feature of TFPMG is the magnetic flux existed in three-dimensional space and three-dimensional finite element method (3-D FEM) is employed to analyze its characteristics . Such as flux-switching TFPM generator [1], many TFPMGs with new topologies have been proposed . However, they commonly have a drawback that only half of PMs do work at the same time and the cogging torque vibrations are unacceptable and desiderated to be optimized . The proposed 12 pole-pairs TFPMG overcomes these shortcomings, which schematic structure is shown in Fig . 1 (a) . The generator is constructed by the double C-hoop stator cores inserted into machined cavities in the stator holder, the doubled PMs screwed onto two rotor disks with opposite polarities to enable the flux-concentrated effect, and the armature winding bundling all stator hoops . II . COGGING TORQUE ANALYSIS The sizing optimization of the stator and the rotor to reduce the cogging torque is analyzed and the side view of the rotor and the stator with indicated size parameters is shown in Fig . 1(b) . The previous investigations demonstrated that the cogging torque is significantly influenced by the ratios of ks and kr, which denote the ratios of circumferential widths of stator hoop and rotor core to pole pitch, respectively [1-2] . The 3D-FEM is employed to investigate the magnetic field distributions and the size relationship . The ratios comply with the inequation of (kri > 1 - ksi) (kri >kso) = kri > 1 - 0 .788 * ksi are described in detail to optimize the cogging torque performance . Fig . 1(c) shows that the amplitude of cogging torque of a single phase TFPMG with optimized ratios of kr and ks is significantly reduced . The circled effect indicates that the optimized cogging torque varies smoother than the original one due to the increased ks with decreased kr . In view of the mentioned factors the optimum values of ks and kr are 0 .76 and 0 .71 to achieve the best performance of the cogging torque and the flux linkage . III . COGGING TORQUE OPTIMIZATION According to the above analysis, the method of skewing the rotor core is applied to improve the flux linkage and minimize the cogging torque of the generator . The top view of the skewed rotor core versus the stator hoop is shown in Fig . 1(d) . The static and transient characteristics of the skewed models at a series of rotor positions from 0 to 6 mechanical degrees are calculated . The no-load flux density distributions under unskewed and skewed rotor core of 3 .6 degree based on 3-D FEM are obtained . As shown in Fig . 2(a) the unskewed rotor core is easier to saturation and the maximum magnetic field intensity is 2 .15 T . On the contrast, as shown in Fig . 2(b) the skewed rotor core of 3 .6 degree does benefit to bring fairly well-distributed flux at the average magnetic field intensity of 1 .5T . Meanwhile, the effect of the skewing method does benefit to the utilization of uniform distribution of magnetic field intensity in both stator and rotor cores . The back-EMF waveforms of skewed rotor core are shown in Fig . 2(c) . It can be seen that when the rotor core is skewed too deep to 4 .8 degree the amptitude of back-EMF will decline sharply . Apparently, when the rotor core is skewed to 3 .6 degree, the back-EMF waveform is the closest to the sinusoid with relatively low total harmonic distortion (THD) . The influences of skewed rotor core on the cogging torque waveforms without and with skewed to 3 .6 degree rotor core are compared in Fig . 2(d) . Obviously, the amplitude of the cogging torque of the skewed rotor core is reduced significantly over 50% of the unskewed structure with minimized cogging torque ripple . The predicted cogging torque waveforms without and with skewed to 3 .6 degree rotor core verify that the optimization is feasible . Hence, the electromagnetic torque will increase enormously if the load current is applied and then the cogging torque is negligible . So the influences of skewing method on static and transient characteristics are equilibrated to consider the \u201cSkewed 3 .6\u201d as the optimal choice .\n1) Jianhu Yan, Heyun Lin, and et al . \u201cCogging torque optimization of flux-switching transverse flux permanent magnet machine,\u201d IEEE Trans. Magn., 49(5) 2169-2172, (2013) . 2) A . Masmoudi, A . Njeh, and et al . \u201cOptimizing the overlap between the stator teeth of a claw pole transverse flux permanent magnet machine,\u201d IEEE Trans. Magn., 40(3) 1573\u20131578, (2004) ." + ] + }, + { + "image_filename": "designv11_84_0003259_gt2013-94105-Figure14-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003259_gt2013-94105-Figure14-1.png", + "caption": "Figure 14: Seeded fault types", + "texts": [ + " The test approach was to first run the lathe rig with undamaged gears to establish baseline data. Following that, the primary gear wheel was replaced with a gear seeded with various faults, to look for changes in the eddy current sensor data relative to the baseline condition. A further test was conducted to assess the sensor response to radial misalignment of the gear wheels. A number of controlled defects simulating damage to the gear wheel or teeth were created in the gear wheel, as follows (illustrated in Figure 14): 1. Single transverse groove across the mid-point of the tooth face 2. Single shallow transverse groove across the root of the tooth 3. Single deep transverse groove across the root of the tooth 4. Three transverse punch dots across the mid-point of the tooth face 5. Single punch dot on one side of the tooth face 8 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/asmep/76989/ on 07/08/2017 Terms of Use: http://www.asme" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001497_978-3-319-07572-3_6-Figure6.9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001497_978-3-319-07572-3_6-Figure6.9-1.png", + "caption": "Fig. 6.9", + "texts": [ + " Load \u00bc mass acceleration due togravity \u00bc 5, 000kg\u00f0 \u00de 9:8m=s2\u00f0 \u00de \u00bc 49, 000N A1 \u00bc \u03c0 0:5m\u00f0 \u00de2=4 \u00bc 0:196m2 A2 \u00bc \u03c0 0:75m\u00f0 \u00de2=4 \u00bc 0:442m2 \u03c31 \u00bc F=A1 \u00bc 49, 000N=0:196m2 \u00bc 2:5 105 Pa \u03c32 \u00bc F=A2 \u00bc 49, 000N=0:442m2 \u00bc 1:1 105 Pa Example 6.4 Calculate the tensile stress in a steel bar 1.5 in. 1.5 in. cross section shown (Fig. 6.8) if it is subjected to an axial load of 110 kips. Solution A \u00bc 1:5 in: 1:5 in: \u00bc 2:25 in:2 \u03c3 \u00bc F=A \u00bc 110kips=2:25 in:2 \u00bc 48:89ksi or 48, 890psi Example 6.5 A circular tube with an outside diameter of 40 mm and inside diameter of 20 mm shown (Fig. 6.9) is under a compressive force of 50,000 N. Determine the compressive stress in the tube. Solution The effective area of cross section is the difference between outside circle area and the inside circle area. Aeffective \u00bc \u03c0Do 2=4 \u03c0Di 2=4 \u00bc \u03c0=4 Do 2 Di 2 \u00bc \u03c0=4 0:0016m2 0:0004\u00f0 \u00de \u00bc 9:4 10 4m2 \u03c3\u00bc 50, 000N/9.4 10 4m2\u00bc 5.31 107 Pa In Sect. 6.3 we discussed tensile and compressive stresses and we called them normal stresses, because they act perpendicularly to the surface. Shear stresses, however, are developed in a parallel direction, or tangentially to the surface (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003678_1.4024704-Figure16-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003678_1.4024704-Figure16-1.png", + "caption": "Fig. 16 x-direction deformation (in meters) of the workpiece. Welding toward the fixed face CC0D0D.", + "texts": [], + "surrounding_texts": [ + "The GTA welding process and the various transport processes involved are discussed in detail in Part I [2]. The mathematical model can be divided into two parts: (a) weld pool dynamics modeling and (b) structural analysis modeling. In the weld pool dynamics modeling, the melting/solidification problem is handled using the enthalpy-porosity formulation. The molten metal flow in the weld pool is obtained using the governing equations of continuity, momentum and energy, based on the assumption of incompressible laminar flow. The Navier\u2013Stokes (N\u2013S) momentum equation takes into account the mushy zone through the momentum sink term, and includes the electromagnetic (Lorentz) force as a body force term. The Lorentz force is determined using the current continuity equation in association with the steady state version of the Maxwell\u2019s equation in the domain of the workpiece for the current density and magnetic flux. The structural analysis model is developed based on isotropic material behavior. The elastic response is handled using the isotropic Hooke\u2019s law with temperature dependent Young\u2019s modulus and Poisson\u2019s ratio. For the inelastic response or plasticity, incompressible plastic deformation is assumed with rate-independent plastic flow and vonMises yield criterion. The yield strength is considered as a function of temperature only. Also, the bilinear isotropic hardening model is employed to consider the material strain-hardening behavior. The mathematical models for both weld pool dynamics and structural analysis have been discussed in extensive detail in Part I and hence is not represented here. However, it is to be noted that the analysis in this study ignores the influences from the arc pressure and a flat weld pool surface is assumed. These assumptions are reasonable for the present study and discussed in detail in Part I of the present study. Also, the boundary and initial conditions used in the mathematical model are described in detail in Part I." + ] + }, + { + "image_filename": "designv11_84_0002771_ecticon.2011.5947899-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002771_ecticon.2011.5947899-Figure1-1.png", + "caption": "Fig. 1. The relation of the accelerometer axis and gravity.", + "texts": [ + " The compensation of two sensors and the final state of attitude estimation correlates these angle estimations with those resulting from the accelerometer and gyroscopes. In this paper, the hardware development board has been implemented and successfully tested on the vibration platform equipped with low-cost sensors. Experimental result are presented and discussed in the last section. II. Accelerometer The 2-axis accelerometer can be used to determine the VA V's attitude (roll and pitch). The 8th Electrical Engineering/ Electronics, Computer, Telecommunications and Information Technology (ECTI) Association of Thailand - Conference 2011 Figure 1 shows an accelerometer which can measure the earth's gravity. The axis of this accelerometer is perpendicular to the VA V. The angle Theta between the actual gravity vector and the measured gravity is related to the pitch of the VA V. We can calculate the pitch orientation of V A V as follows: Pitch = Theta + 90 0 Accelerometer = cos(Theta)*gravity Theta = Acos (Output_accelerometer / gravity) then Pitch = Asin (Output_accelerometer / gravity) (1) For calculation of the roll angle, another axis of accelerometer which perpendicular to the pitch accelerometer is needed to determine" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003403_amm.419.438-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003403_amm.419.438-Figure2-1.png", + "caption": "Fig. 2 The new grip design.", + "texts": [ + " Used SOLIDWORK software to draw 3D models, and analysis different materials of golf head and shaft by ANSYS. Design Fig.1. The new golf wood club All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 132.174.254.159, Pennsylvania State University, University Park, USA-24/05/15,08:04:02) This paper adopted of the new golf wood club is design, including head (1), shaft (2) and grip (3), is shown in Fig.1. The new grip desgin in this paper is shown in Fig.2. The detail explanation is as follow. The cover (4) is provided with screw holes (6). The shaft extends into the inner screw holeswith threaded connection. Grip (5) has a big end (7) and a small end (8) of the truncated conical shape. The grip (5) includes theouter grip (9) and the inner grip (10). The outer grip is made of thermoplastic elastomer (TPE), the inner grip is made of natural rubber. Compared with existing technology, the advantages are that the assembly process is simple. The physical parameters were shown in Table 1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003816_amr.291-294.2115-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003816_amr.291-294.2115-Figure1-1.png", + "caption": "Fig. 1 The 3-DOF parallel manipulator", + "texts": [ + " The approach has been validated and applied to dynamic variable structures or nonlinear systems [4-5]. There are some natural frequencies and damping ratios to multi-mode system. A classic multi-mode input shaper is setup when several same kinds of input shapers are convolved [6]. The performance of multi-mode input shaper is decreased when several frequency bandwidths and amplitudes of vibration modes are very different. This disadvantage can be overcomed by using the hybrid multi-mode positive input shaping strategy. A 3-DOF parallel manipulator (shown in Fig. 1) is composed of three symmetrical closed-loop linkages [7]. Light weight linkages are used for better performance with high speed. However, lightweight linkages are more likely to vibrate, and the trajectory accuracy of the system motion is sacrificed. All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 152.14.136.77, NCSU North Carolina State University, Raleigh, USA-10/04/15,03:33:54) deformed linkages are shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure4-1.png", + "caption": "Figure 4 Ilmor X3 joint modelling (see online version for colours)", + "texts": [ + " These modes had no effect on the simulated Ilmor handling performance. To prepare the links for simulation in Fedem, it was necessary to determine where the different links should be connected by joints. A Fedem joint connected at least one master and one slave node (triad) from two links to facilitate constrained 3D translational or rotational motion between the links. The master or slave node could be attached to other nodes on the same link via RBE2 or RBE3 elements to improve the structural load distribution. Figure 4 illustrates how one of two revolute joints (shown in yellow) enable rotational motion between the Ilmor engine block and the crank shaft. In this study, both rotating and counter-rotating crankshafts were tested to reduce the wheelie effect. This master and slave-based joint modelling technique is very robust, and the reduced FE models eliminated the traditional problems that are encountered with over-constrained rigid bodies. In the example given above, the crankshafts were constrained using two revolute joints (only one joint is shown), thereby imposing ten constrained DOFs" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002107_146.111076-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002107_146.111076-Figure4-1.png", + "caption": "Fig. 4. The Idealized geometry of fracture depicting the four-point bending test condition.", + "texts": [ + " The infiltrated part was normalized at 900 8C, air cooled and then austenitized in a neutral salt bath at 900 8C for 0.5 hrs. The specimens were then tempered at three different temperatures; at 177 8C/1 h/water quenched, 428 8C/ 1 h/water quenched and 704 8C/1 h/water quenched. 3.2. Fatigue-crack propagation test The fatigue-crack propagation test of RapidSteel was carried out on a 40 kN electro-servo-hydraulic test machine in four point bending mode at a stress ratio of R = 0.1 and a frequency of 100 Hz. Single-edge notched RapidSteel specimens (see Fig. 4) dimensioned at L = 65, W = 17, and B = 13 mm were used for fatigue-crack propagation measurements at room temperature. The load-shedding procedure was used to determine the crack growth thresholds according to the British Standard BS 6835-1:1998 [12]. The reported stress-intensity factor thresholds correspond to a crack growth rate of 10\u20138mm cycle\u20131. Conventional potential-drop technique and optical microscopy were used to measure and observe the fatiguecrack. During the experiments, the instantaneous crack length a, the number of cycles N, and applied load range (DP = Pmax \u2013 Pmin) were continuously monitored" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure6-1.png", + "caption": "Figure 6 RELATION BETWEEN AXIAL FEED AND ANGULAR POSITION", + "texts": [ + " As a result of the cutter being indexed relative to the gear, the speed ratio is no longer equal to the ratio N N of the number of teeth between thec g\u00ce cutter and gear. The equivalence of normal pitches determines 3 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use the relation between the axial positions and . The speedw wc i and feed of the cutter are pre-specified. Depicted in Figure 6 is the transverse curve for one rotation of the input gear element. f is the feed or axial displacement per revolution of the geari element. The following relation is obtained from the law of sines (N p ):i ni \u0153 2 cos( )1 # b2 + a2 2a . We need to find the shortest path through which a UAV can reenter the common surveillance region once the UAV goes outside it. From the analysis in [2], we can conclude that circular paths of radius R, which completes the longest paths inside the ellipse, maximises the ratio of time spent inside the ellipse to that spent outside it. There are two such paths with total length 2\u03c0R, with the length of path inside the ellipse being 2Rsin\u22121(b/R). These paths are shown in Fig. 5. Hence, to ensure that at least one UAV is always present inside the ellipse, the minimum number of UAVs required is given by: N = \u2308 \u03c0 sin\u22121(b/R) \u2309 where, is the ceiling function. In this paper, we studied the case where fixed wing UAVs with bounded turning radius are used to monitor two different stationary ground targets simultaneously. We approximated the feasible region with an ellipse, and determined the longest possible path for the UAV inside it using Pontryagin\u2019s minimum principle. The paper not only generalizes the single target protection scheme given in [2] to two targets but also generalizes the optimal flight trajectory of a UAV inside a circular region (as seen in [2]) to trajectories in a more general elliptical region" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003537_sces.2014.6880104-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003537_sces.2014.6880104-Figure1-1.png", + "caption": "Fig. 1 Force Balance diagram of L", + "texts": [ + " where and \u03b2 mgl From \u201c(1),\u201d and \u201c(2),\u201d (M I sec / cos ) = g = u \u2013 h where [7] has been applied on LIP [8] [9] is a highly ariable system which has t at upright position. LIP which is stabilized by variables [11] are the position and its derivative ive backstepping will be is stabilized at upright RTED PENDULUM n under actuated system osition are controlled by ated by servo motor .This more complex to control sed loop control system actuator, pendulum as a der to detect the angle of erted pendulum system is hich is hinged to cart of s shown in Fig. 1. The stem can be described by ned through conservation inear Inverted Pendulum u (1) (2) M \u03b2 tan / sin (3) (4) g = M I sec / cos h M \u03b2 tan / sin u = g ( (5) where p = h/g I sec / I sec / \u03b2 tan / (6) x u/m \u2013 b (7) where m = k g/I sec x b pI sec x /k \u03b2 tan x /k So the complete dynamic model of linear inverted pendulum can be represented as, x ug p x xx um b (8) where \u201cg\u201d, \u201cp\u201d, \u201cm\u201d and \u201cb\u201d are system parameters which are combination of system parameters such as inertia \u201cI\u201d ,mass of cart \u201cM\u201d and mass of pendulum \u201cm\u201d and length of rod \u201cL\u201d of linear inverted pendulum" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003924_elektro.2014.6848891-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003924_elektro.2014.6848891-Figure9-1.png", + "caption": "Fig. 9. Axial cross section of induction machines.", + "texts": [ + " As a result we needed to decide whether it was necessary to change the drive (replace the existing drive by a new one with a higher rated power) or which impact and influence the overloading had on the existing motor. 978-1-4799-3721-9/14/$31.00 \u00a920 141EEE 222 II. PROBLEM DESCRIPTION The continuous 24-hour power measurement was performed with an independent measurement system. The machine operates approximately at constant frequency, voltage and power factor. Due to this fact the current is proportional to the power (Fig.l) and also proportional to the torque (fig. 9). For this given (measured) duty cycle, any other magnitudes and dependencies were calculated. The starting torque does not cause any problems, because the pump starts generally with a small torque, and the required initial torque pulse (kick-on pulse) has always the same amplitude. Just with the increasing speed the torque increases approximately quadratic. The wastewater pump was located in the concrete dungeon, almost completely under the surface. The motor ambient temperature was relatively low, 10-15 \u00b0C all the year", + " But unfortunately, this simplification cannot be done in the study of complex thermal analysis. In the view of temperatures distribution the model must be most complex. Previous fact leads to made complex 3D model of induction machines. Boundary conditions have been estimated using analytical solution [3], [4], [5]. The equivalent thermal conductivity of construction elements is used the next relation n I I1=1 apr, (8) Where t5 is length of i material, ,1 is thermal conductivity of i material. V. SIMULATION RESULTS Fig. 9. and 10. shows the distribution of temperatures. Fig. 9. represent axial cross section of induction motor. Fig. 10. represent cross section of 1M. Temperature of stator winding, rotor shorting rings and rotor bars is important of thermal analysis. Highest temperature is placed at rotor bars. Maximal temperatures of rotor reach up to 160 0e. Temperature of stator winding is 135\u00b0e. VI. CONCLUSION The motor temperature depends on the type of the loading cycle and ambient temperature. The self-ventilated system is insufficient for the given loading cycle, because the temperature is over the class B, although under the class F" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure6.4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure6.4-1.png", + "caption": "Fig. 6.4 Views and details from the arch-support", + "texts": [], + "surrounding_texts": [ + "work space of Autodesk\u00ae Fusion 360\u00ae Generative Design which can be found on the upper-left corner of the window, see Fig. 6.5. At this point, the work flow is very similar to the design proposed by Krish [9] where the designer modifies the model features and the restrictions to generate the iterations, see Fig. 6.6. Once the program is in the environment frame, tools are activated and buttons will be selected from left to right, see Fig. 6.7." + ] + }, + { + "image_filename": "designv11_84_0003223_detc2013-12237-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003223_detc2013-12237-Figure1-1.png", + "caption": "FIGURE 1. SEVEN-LINK BIPED MODEL.", + "texts": [ + " The control torques/forces are obtained from \u03c4 = (DcBT )\u22121Dc(My\u0308\u2212 h\u2212CT b \u03bb b). (5) The reactions of constraints \u03bb c can be written as \u03bb c(y\u0307,y, t) =\u2212(CcM\u22121CT c ) \u22121[Cc0 +CcM\u22121(h+BT \u03c4 +CT b \u03bb b)], (6) where Cc0(y\u0307,y, t) = \u2212(C\u0307cy\u0307+ d/dt(\u2202\u03c6 c/\u2202 t)). The reactions of constraints are often denoted as generalized reaction forces, too. 2 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 04/15/2018 Terms of Use: http://www.asme.org/about-asme/terms-of-use BIPED MODEL A planar seven-link biped model is considered as shown in Fig. 1. The planar biped is composed of two identical legs and a HAT. Each leg consists of a femur, a tibia and a rigid foot. The assumptions of this study are that 1) the HAT maintains a constant upright posture during motion; 2) all joint actuators are revolute, massless, frictionless and can only move in the sagittal plane; 3) the supporting foot is in contact with a hard surface terrain, and both feet stay parallel with the stairs; 4) there is enough frictional force to prevent slippage between the foot and the floor during walking", + " The \u201cflying\u201d biped has nine DOF, n = 9. Its position coordinates are y = [xH , yH , \u03b8R1, \u03b8R2, \u03b8R3, \u03b8L1, \u03b8L2, \u03b8L3, \u03b8T ] T , and control inputs are \u03c4 = [\u03c4R1, \u03c4R2, \u03c4R3, \u03c4L1, \u03c4L2, \u03c4L3] T . The biped starts from the state of rest when both feet contact the ground and the whole body weight is supported by the left (L) leg. The initial step which is the only one different from the following is done by the right (R) leg. Table 1 contains the inertial and geometrical parameters of the biped where the notation follows from Fig. 1. Constraints The constraints (a) represent the trajectories of hip position and HAT orientation, \u03c6 a = xH \u2212 xHn(t) yH \u2212 (h0 + xH \u00b7 (hst/dst)) \u03b8T = 0, (7) whereas constraints (b) and (c) are \u03c6 b,c = xH + l1 \u00b7 sin\u03b8i1 + l2 \u00b7 sin\u03b8i2 \u2212 xAi yH \u2212 l1 \u00b7 cos\u03b8i1 \u2212 l2 \u00b7 cos\u03b8i2 \u2212 yAi \u03b8i3 \u2212\u03c0/2 = 0, (8) where the number of constraints is n = 9 with m = 3 for each leg and l = 3 for the HAT, and xAi and yAi are ankle coordinates, with i = R or L. Rheonomic Constraints. The values of the hip coordinates for the first step are obtained from equation xHn(t) = a5t5 + a4t4 + a3t3 with boundary conditions, xHn(0) = 0, x\u0307Hn(0) = 0, x\u0308Hn(0) = 0, xHn(T0) = s0, x\u0307Hn(T0) = v0, x\u0308Hn(T0) = 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000926_gt2008-50641-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000926_gt2008-50641-Figure4-1.png", + "caption": "Figure 4 \u2013 Boundary conditions for calculating the elementary stiffnesses", + "texts": [ + " 3 has six degrees of freedom, namely two vertical displacements, 1v and 3v , and four horizontal displacements 1u \u2026 4u ; this is the minimum required for describing the coupling between two successive bumps. The elementary characteristics of the bump strip model are the stiffnesses ik and the transmission force angle d\u03b8 that can be analytically expressed for each type of bump. The elementary stiffnesses and transmission force angles are determined by using energetic methods for two kinds of boundary conditions (Fig. 4): no rotation or free rotation. This distinction is necessary in order to correctly take into account the last bump of the strip located at the free end. The analytic calculations use the classical linear elasticity approach. First the bending moments and normal loads are calculated in all regions of the elementary structures. Then the elastic deformation energy is estimated and Castigliano\u2019s theorem is used to express the elementary spring stiffness or the transmission force angles. The complete mathematical development and the resulting relations for ik and d\u03b8 are nloaded From: http://proceedings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002923_gt2013-95585-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002923_gt2013-95585-Figure7-1.png", + "caption": "Figure 7. RE-CIRCULATING FLOW UNDER LEAF TIP", + "texts": [ + " To allow a better investigation of the second flow effect, the recirculation due to viscous shear by the rotor, a linear pressure profile decaying from the high pressure inlet to the low pressure outlet, has been subtracted from these to create the contours (b), (d), and (f) respectively. These resulting contour plots show that the inlet and exit effects have almost no impact on the bulk of the flow field under the leaf tip and that the viscous recirculation effect is nearly independent of ambient pressure and axial flow through the gap. This is as expected since viscosity is not a function of density. The circumferential variation in pressure (left to right in the figures) is as expected: a cartoon of the flow field is sketched in Fig. 7. At the leftmost edge, the sudden increase in gap height from G to G+h causes a low pressure region. As the gap converges, a high pressure region is created approximately 50% to 80% along the ramp. Interestingly the low pressure region from the next increase in gap height on the neighbouring leaf spills over, causing a reduction in pressure close to the righthand end of the leaf tip. By integrating any of the corrected pressure profiles, and resolving perpendicular and parallel to the rotor surface, the hydrodynamic lift and drag acting on the leaf tip can be calculated", + " Figures 9 and 10 are surface plots of leaf tip lift and drag for a constant speed condition. As expected the lift force in Fig. 9 quickly decays as the gap height G increases, confirming that a significant air-riding force only exists for gaps below 2.0\u00b5m. Interestingly there is also a dependence on slope height h, with peak lift forces being created at h \u2248 1\u00b5m. This can be explained by the fact that shallow gaps (small h) are poor at viscously compressing the flow, implying low peak pressures and that steep gaps (large h) suffer from large low pressure regions (see Fig. 7). The relationship between leaf tip drag and gap shape, shown in Fig. 10, is significantly different, showing a peak at minimum G and h. This is the case as laminar shear drag is proportional to 1 Gap . Geometries used for the investigation were chosen to be more densely clustered at low values of G and h, where lift and drag forces change more rapidly. The results provide a good overview of how these are affected by changes in gap geometry. By interpolating in the resulting matrices containing values for lift and drag for the various combinations of G, h, and VR, the forces acting on the leaf tips for a set of given conditions can be determined" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002176_amm.712.81-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002176_amm.712.81-Figure1-1.png", + "caption": "Fig. 1. The analyzed gear wheel", + "texts": [ + " Gear wheel which is a part of one of the mechanical systems of an automotive vehicle made of sinter Sint-D 32 and alloy structural steel 42CrMo4 for quenching and tempering has been analyzed from structure and functional properties point of view. All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 130.237.29.138, Kungliga Tekniska Hogskolan, Stockholm, Sweden-11/07/15,19:09:50) As a material for the research were used gear wheels which form part of one of the mechanical systems in an automotive vehicle, schematically shown in Fig.1. This element is made of two types of materials representing the various technologies. The first material is made of sinter by powder metallurgy technology (technology parameters reserved) based on a powder purchased commercially Sint-D 32 (Table 1) from the serial production process. The second material is alloy steel structural for quenching and tempering 42CrMo4 (Table 2). Metallographic examination have been performed using optical light microscope Axiovert 25 and scanning electron microscope Jeol JSM-6610LV in order to determine the structure and chemical composition of the tested materials" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure22-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure22-1.png", + "caption": "Figure 22 Adjustable pivot axes (see online version for colours)", + "texts": [], + "surrounding_texts": [ + "The aerodynamic drag force is the most important factor affecting the behaviour of the bike. A lifting force from air flow also acts on the bike. This effect was not included in the simulation model. The aerodynamic drag force was calculated using the following equation: 2 2 o X AD \u03c1 V C A F \u22c5 \u22c5 \u22c5 = where \u03c10 denotes the density of air at 20 deg./1 atmosphere pressure (1.2 [kg/m3]) V denotes the bike velocity in [m/s] CX denotes the drag coefficient A denotes the frontal area of the bike. SRT provided a drag coefficient multiplied by the frontal area A of the Ilmor X3 bike of 0.27 m2 (Giussani, 2007). The aerodynamic drag force at each time step in the simulation was calculated by the control system. This drag force was applied at the front of the bike and was always opposite to the direction of travel. The control system measured the velocity of the frame CG, and the aerodynamic drag force was calculated using the formula given above." + ] + }, + { + "image_filename": "designv11_84_0003708_gt2014-26673-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003708_gt2014-26673-Figure2-1.png", + "caption": "Fig. 2 A FE model of the cantilever beam", + "texts": [ + " The algorithm developed for the selection of the step size allows the capture of all variations of the LCO amplitudes even where these variations are abrupt, and provide high computationally efficiency. The method numerical properties have been tested on a number of structures. Some of the examples considered and obtained results are discussed below. Analysis of LCO sensitivity for simplified test cases One of the test cases considered was a cantilever beam with sides 1000200100mm and with the following material properties: elasticity modulus E=10 5 N/mm 2 ; density =4.43*10 -9 Mg/mm 3 . The three-dimensional solid finite element model shown in Fig. 2 is used. One end of the beam is free, the other end is clamped and the total number of DOFs in the model is 12500. The modal damping factors of the beam are assumed to be 0.02 for all mode shapes, except of the modal damping factor for the first mode, which is assumed to be the fluttering mode, and in most calculations it has the value 1 0.02 . Three types of the nonlinearities applied to a node at the free end of the beam on amplitude and frequencies of LCO are explored: (i) friction damper; (ii) cubic nonlinear spring and (iii) gap nonlinearity" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001652_978-3-319-10891-9_10-Figure10.4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001652_978-3-319-10891-9_10-Figure10.4-1.png", + "caption": "Fig. 10.4 Magnetic levitator and its model", + "texts": [ + " The setup uses two computers, one for the controller and the other for simulating the plant with sensors and actuators. The computers are connected through a real Ethernet link. The experimental network setup is such that 1 \u00d7 10\u22123 [s] \u2264 \u03c4m i+1 \u2212 \u03c4m i \u2264 5 \u00d7 10\u22123 [s] 1 \u00d7 10\u22123 [s] \u2264 \u03c4 c i+1 \u2212 \u03c4 c i \u2264 5 \u00d7 10\u22123 [s] . (10.36) Figure 10.3 shows the measured round trip time. Based on the measurements, we can consider the maximum delays1 to be T m, T c = maxRTT 2 \u2248 26 \u00d7 10\u22123 [s]. The plant parameters (equations are shown in Fig. 10.4) are \u03b1 = \u03c0 6 , g = 9.8 [ m s2 ] , m = 0.05 [ Kg ] , c = 0.5 [Hm] , L = 1 [H], R = 10 [ ]. According to Sect. 10.2.1, we assume that a stabilising controller is given for the nominal plant. In our case, the nominal controller is the result of the straightforward application of numerical self-tuning routines inMatlab, and has the transfer function: 1 The measurements include both the network-induced delays and some additional delays which have been added via software in order to simulate the effects of additional traffic" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001263_s11771-014-2355-z-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001263_s11771-014-2355-z-Figure1-1.png", + "caption": "Fig. 1 Impact model of gear meshing spur gears", + "texts": [ + " In this work, the main motivation is to conduct the simulation of nonlinear impact damping model with noninteger compliance exponent, which is different from the research [24\u221225] with a simplification at n=1. The nonlinear impact damping model was analyzed, in which the physical parameters were determined under two types of compliance exponent conditions. Combining with Runge-Kutta numerical method, the numerical analyses were conducted. Light and heavy load conditions were considered, which exhibited different responses in different cases. The single-degree-of-freedom gear impact model composed of a pinion and a gear is depicted in Fig. 1. An external torque Tp is exerted upon the pinion, and the output torque Tg upon the gear is equal to the input torque. mi (i=p, g) and Ii (i=p, g) are the mass and the moment of inertia of the pinion and the gear, respectively. Ri (i=p, g) are the base radii. Some key assumptions for the analyses of the nonlinear impact damping gear system are included as follows: 1) the pinion and gear are rigid disks; 2) the motions considered are in the same plane and only torsion vibrations are studied. Backlash B is adopted in the dynamic model to prevent tooth jamming and thermal expansion for the manufacturing errors and assembly errors" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure26.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure26.2-1.png", + "caption": "Fig. 26.2. Motion with restrictions", + "texts": [ + " The controller inputs are either position or velocity commands, sent from the master site. According to the nature of the task the controller can receive the restrictions from the Geometric Reasoning Module and the variables corresponding to the allowed DOF. When position or velocity commands are selected from the local command center depending on the task, the robot can move strictly in the restriction subspace (xr) or with a deviation from it (xhd), allowed by the stiffness and damping implemented in the Force Guidance Module. In Fig. 26.2, vector d represents the deviation of the position or velocity command produced by the operator. 26 Force Reflecting Teleoperation Via IPv6 Protocol 449 The position control scheme of the remote robotic cell is stable. Then, if the input references (position/velocity) of the controller are bounded the overall system is also stable. The drawback of this straightforward approach is that in some cases transparency of the overall system is sacrificed for the sake of system stability. Most geometric constraint solvers come from the CAD world and deal with the general problem of positioning multiple objects in such a way that they satisfy a set of predefined geometric constraints", + " There are two possible combinations of position/velocity of the master in this scheme. Depending on the task, the operator must decide whether the real position/velocity of the master or the projection into the restricted subspace rss of these values is send to the slave. In the first case the slave motion is strictly within the restricted subspace (gs = xr or gs = x\u0307r), and, in the second, it moves along the actual values of the master (gs = xhd or gs = x\u0307hd) meaning that the master\u2019s position/velocity could deviated by a certain value |d| from the restricted subspace (Fig. 26.2). \u2022 Scheme 2. In this scheme, instead of holding the master\u2019s position/velocity data, the guidance signal carries the restriction subspace information along with the current position/velocity in the subspace, gs = (rss,pm) or gs = (rss, p\u0307m). Since the movements only take place in rss, it is not necessary to send all six position/velocity components but only the ones concerning the available DOF. \u2022 Sensor force. The raw force measurement fe that comes from the sensor\u2019s data is filtered in the local command center at a cutoff frequency of 500 Hz. The resulting sensor force fs is calculated as follows: fs = Tsfe where Ts is a transformation between the force sensor frame and the haptic frame. \u2022 Restriction force. This is an attractive force that tends to fix the haptic position to the restriction subspace. The direction of this force is calculated by vector d (Fig. 26.2) and its magnitude is given by: frk = KPek + Dk \u2022 Viscous force. If velocity mode is selected and the velocity of the master is high enough, the slave may not be able to follow the velocity commands. To avoid this situation an additional restriction has been implemented: above a certain velocity value, which depends on the maximum velocity achievable by the slave, the motion restricting force is a function of the master velocity x\u0307hd = v\u0302, and this force is zero below that value. The resulting force of this effect fv at instant k is given by: fvk = Kvv\u0302k where Kv is a gain that fits the needs to restrict velocity" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure22.4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure22.4-1.png", + "caption": "Fig. 22.4. (left) Experimental setup, (right) Geometric relation for the task \u201cpathfollowing\u201d", + "texts": [ + " In the robotic procedure adopted by systems such as Robodoc [28], Acrobat [29] and CASPAR [30], a mill is used for the cutting procedure, and the blade is required to cut along the planned path on the bone. Meanwhile, the cutting edge of the tool should be kept perpendicular to the cutting plane in order to provide more efficient force. Modeling of Task We model the femur cutting task as a task to guide the tip of a long straight tool following a 2D b-spline curve C1 in plane \u03a0 while keeping the tool shaft 392 M. Li, A. Kapoor, and R.H. Taylor perpendicular to the plane. The geometric relation is shown in Figure 22.4. We assume that the path C1 and the cutting plane are known in the robot coordinate frame by using an appropriate registration method. During the procedure, the tip of the tool (task frame {t}) is allowed to move along the planned path C1. At the same time, a point, xp,s on the tool shaft (task frame {s}) is only allowed to move along the second path C2, which is a translation of C1 above the target plane. xclp,t is the closest point to the tip of the tool on C1 and xclp,s is the projection of xclp,t on C2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002384_2011-01-1424-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002384_2011-01-1424-Figure5-1.png", + "caption": "Fig. 5. Test components in the TTR.", + "texts": [ + " Indeed, the most severe contact conditions in the MTM would only lead to power transmission of approximately 30W in the contact, whereas the power in the variator contact may exceed 45kW. The Torotrak Traction Rig (TTR) Torotrak has developed a novel disc traction rig for the measurement of the traction properties of fluids under conditions representative of the variator [17]. Unlike conventional 2-disc rigs, a third cylindrical disc rotates in a plane perpendicular to the driving and driven crowned rollers, as illustrated in Figure 5. The purpose of the third component is to introduce spin into the contact and thereby simulate the full-toroidal variator kinematics. The TTR embraces the operating envelope of the variator contact up to 25kW and was used in the current study to assess any changes in the traction performance of the test traction fluid by measuring the peak traction coefficient of the fluid before and after the three parts of the fluid life tests. SAE Int. J. Engines | Volume 4 | Issue 12138 Shown in Figure 6 are the results of the three endurance tests along with a number of S-N tests at much higher variator powers (typically 210kW) than previously presented" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002754_icelmach.2014.6960214-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002754_icelmach.2014.6960214-Figure9-1.png", + "caption": "Fig. 9. CP-MGG for hub dynamos.", + "texts": [ + " As was mentioned, to satisfy JIS requirements, the multipole claw pole structure is needed. However, because a higher pole number will cause magnetic short-circuiting, it cannot be said that the claw pole structure only by itself is effective. Therefore this paper suggests the CP-MGG, which is essentially a claw pole generator that has been combined with the magnetic gear to increase the rotation speed and increase the reactance but at the same time also reduce the number of the poles. The proposed CP-MGG is shown in Fig. 9 and its specifications are shown in Table III. The proposed CPMGG has three major parts; a low-speed rotor, high-speed rotor and stator. The low-speed rotor is composed of 30 pole pieces made of laminated silicon steel sheets. The highspeed rotor has ferrite magnets which form 10 pole pairs. The stator consists of circumferentially arranged claw poles, a single coil, and ferrite magnets between the claw poles. These permanent magnets are all magnetized in the same direction. The operational principle of the CP-MGG is described" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002069_cdc.2013.6760116-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002069_cdc.2013.6760116-Figure4-1.png", + "caption": "Fig. 4. Picture of the robot considered in the simulations.", + "texts": [ + " This task, however, is reasonably straightforward since the vector o associated with an object of known position, namely the set of all the cubes occupied by it, can be easily determined and, as a consequence, the set RO(o). Note finally that the method can easily be extended to the case in which the obstacles move. To show the effectiveness of the method explained in the previous sections we consider the path planning problem for a robotic manipulator the end-effector of which has to reach a target avoiding an obstacle which lies in the straight line connecting the initial and the final positions. We focus on the first three d.o.f. of a 6 d.o.f. manipulator (see Figure 4). Therefore, we have a total of three degrees of freedom and hence a three-dimensional problem. The obstacle consists in a vertical plate located in the robot workspace. As described above, the first step is to sample the possible configurations and the working environment and to construct the corresponding matrix C. In this case the edge of the cube corresponding to the environment is 3 meters long, while the edge of the sub-cubes is 20 centimeters long. Hence there are a total of l = 3375 sub-cubes" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003889_kem.584.142-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003889_kem.584.142-Figure2-1.png", + "caption": "Fig. 2 (Up Right): Traditional three views 2-Dimensional drawing", + "texts": [ + " It is required by AICHE that the appropriate size of the car had to fit in the shoebox parameters when disassembled, and no larger than 400mm long, 300mm wide and 18mm high. Secondly, it was necessary to take into consideration the material of the wheels would be used. The main focus was on using rubber wheels. Those wheels could efficiently transfer electrical and mechanical power from the driving motor to the wheels without having to worry much about the wheels slipping on the ground at the start of each run. The Fig. 1 and Fig. 2 below, shows the preliminary design of the chemical \u2013 electric model car. The body had exactly 400mm length, 300mm width and 5mm thickness. This body was made from alloy steel, which had very high Modulus of Elasticity (Young\u2019s Modulus, around 200GPa). Fig. 3 shows the simulated deformation of the car body by using alloy steel under 100 N force uniformly added to the surface of the car body. The maximum displacement is 0.09248mm, which means that this 5mm thick car body can carry all of the chemical equipment on it with nearly zero meter displacement" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003727_2015-01-1524-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003727_2015-01-1524-Figure1-1.png", + "caption": "Figure 1. Carcass translation deformation (Left), bending deformation (middle), twisting deformation (Right).", + "texts": [], + "surrounding_texts": [ + "Tire carcass lateral deformation can be divided into three kinds of deformation: translation, bending and twisting deformation [7]. The tire carcass lateral translation deformation is (1) Where, Kcy0 is the carcass lateral translation stiffness. Carcass lateral bending deformation is: (2) where, Fy is tire lateral force, Kcb is lateral bending stiffness, a is half of the tire contact patch length, x is the longitudinal coordinate of each tread element in the contact patch coordinate system, \u03be(x/a) is the general carcass bending function, representing the tire carcass lateral bending deformation. Carcass lateral twisting deformation is: (3) Where, Mz is tire aligning moment, \u03b8 is tire twisting angle, N\u03b8 is tire carcass twisting stiffness. Therefore, the tire carcass lateral deformation can be expressed as: (4)" + ] + }, + { + "image_filename": "designv11_84_0001387_1432891715z.0000000001469-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001387_1432891715z.0000000001469-Figure9-1.png", + "caption": "Fig. 9. The strain on the contact area is then small enough, ensuring that the contact area is within the range of linear elasticity theory. Using equations (2), (3) and (5) and ignoring the influence of the frictional force in the x direction to the displacement in the y direction, the displacement of the contact central point o in the y direction is as follows:", + "texts": [], + "surrounding_texts": [ + "coefficient of a metallic lenticular ring gasket joined by a bolted flange H. X. Gong*, Z. G. Xu and S. Wang The metallic lenticular ring gasket seal belongs to the half self-tightening category. The selftightening capability of the gaskets directly determines the sealing performance of the pipeline joined by a bolted flange. Self-tightening coefficients of the lenticular gasket are defined by a mathematical model for the maximum contact pressure and preloaded force of the joint. Following this assumption, new formulae for the gasket stiffness and its member stiffness were derived using the theory of elastic mechanics. The relationship of their deformation coordination to different working conditions was analysed. The theoretical results indicate that the selftightening capability of the metallic lenticular ring gasket is a function of the member stiffness." + ] + }, + { + "image_filename": "designv11_84_0002069_cdc.2013.6760116-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002069_cdc.2013.6760116-Figure6-1.png", + "caption": "Fig. 6. Two representations of the trajectory in the workspace. The initial and the final configurations are depicted with dark and light gray, respectively. The solid and the dashed arrows correspond to i(t) = 1 and i(t) = 0, respectively.", + "texts": [ + " The result of this operation is represented in Figure 5 with the stars points. Note that, given C and the cubes occupied by the obstacles, this step is as fast as a matrix multiplication. The final step is the construction of the Hamiltonian function (7) and to select the matrix J in (4). In this particular case J has been chosen in such a way that the trajectory lies on a plane. Figure 5 represents the trajectory in the space of the joints from the initial point (the small circle) to the target (the small cross). Figure 6, instead, show a sampling of the motion of the robotic arm in the real space. The initial (dark gray) and the final (light gray) positions are shown together with other samples of the trajectory. A movie showing the whole trajectory in the workspace, made with the help of the OpenRAVE software [17], is available on-line at the URL [18]. The contribution of the present paper is twofold: we have extended a previous result concerning the recently developed Hamiltonian-based path planning method to the case on ndof robot; we have analyzed the implementation aspects and shown the practical implementability by simulating a realistic scenario" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003981_icems.2011.6073407-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003981_icems.2011.6073407-Figure4-1.png", + "caption": "Fig. 4 Magnet dimensions of IPM CMMF motor. \u03b81: magnet effective opening= 30[deg], \u03b82: magnet angle= 15[deg], lg: gap length= 0.5[mm], l2: gap to flux barrier length= 0.6[mm], lm: magnet thin= 2.7[mm], lmh: magnet width= 10.0[mm])", + "texts": [ + " The interior type PMSM, in generally, has some advantages such as a short air gap length and generating reluctance torque compared to the SPMSM. Adding the above advantages, the proposed interior type CMMF motor (IPM CMMF motor) has a possibility to utilize magnetic coupling between p1 and p2 characteristics. So, we reveal these characteristics and advantages in this section. A. Interior type CMMF motor model and concept The models of interior type and surface type 4 poles and 12 poles/ 18 slots CMMF motor (IPM and SPM CMMF) are shown in Fig. 4. The motor dimensions are shown in TABLE I, the magnet shape of IPM CMMF motor is expanded as shown in Fig. 4 (c). The magnet effective opening is same to the SPM model (30 [deg]) and magnet angle is 15 [deg]. B. Basic characteristics The basic characteristics such as motor parameters and torque waveform are shown in this section. Because the motor has two kinds of magnetic parameters, we need to reveal each magnetic parameter respectively. Fig. 5 shows the current and inductance characteristics by adding each 4 poles or 12 poles synchronized current. As shown in this figure, the 4 poles property has a saliency in the wide current range. However, the 12 poles property has little saliency between d and q axis inductances. This is because the 4 poles (magnet) flux path is remained as shown in Fig. 4 (a), but a path of 12 poles magnet flux is mostly cancelled by CMMF method. The summary of its electric parameters are shown in TABLE II. The magnet linkage fluxes are calculated by using no-load FEA results. Fig. 6 shows the torque waveforms by adding each 4 poles or 12 poles synchronized current and its compounded current for IPM CMMF motor (Fig. 4 (a)). In this figure, we found that the 4 poles property (T4poles) has lower torque ripple than 12 poles (T12poles), because the 4 poles property is able to utilize not only the magnet but also the reluctance component to generate the output torque. In Fig. 6, we also show the linear summation (TL) of T4poles and T12poles and FEA result (TComp) by adding 4poles and 12poles synchronized current simultaneously. From this figure, we found the difference between TComp. and TL. This means that the compounded property has a coupling between 4 poles and 12 poles characteristics" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002757_imece2011-63090-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002757_imece2011-63090-Figure3-1.png", + "caption": "Figure 3. The Full Vehicle Ride Model (7-DOF)", + "texts": [ + " The governing equations of motion can be obtained by considering the forces and moments as follows: 0 M Z C Z l Z C Z l Zs f wf s r wrb s f r k Z l Z k Z l Zs f wf s r wrf r (3) 0 I C Z l Z l C Z l Z ls f wf s r wryy f f r r k Z l Z l k Z l Z ls f wf s r wrf f r r (4) 0 m Z C Z l Z k Z l Zs f wf s f wfuf wf f f C Z Z k Z Zwf rf wf rftf tf (5) + 0 m Z C Z l Z k Z l Zs r wr s r wrur wr r r C Z Z k Z Zwr rr wr rrtr tr (6) rkfk rcfc trk trctfk tfc wrZwfZ ufm urm rrZrfZ sZ fl rl fl rl bM Figure 2. The Half Car Model (4-DOF) 2.3 The Full Vehicle Ride Model (7-DOF) The full-vehicle suspension model shown in Fig. 3 consists of sprung mass Mb with three degrees of freedom namely; bounce zs roll and pitch motions. Another four masses are added to represent the unsprung masses mi . For simplicity all tires are replaced with equal stiffness kt and tire damping coefficient Ct . Each suspension is modeled by a linear spring in parallel with shock absorber. Using the Newton s second law of motion, the following equations of motion can be derived. 5 1 6 2 7 3 8 41 2 3 4 05 1 6 2 7 3 8 41 2 3 4 M Z C Z Z C Z Z C Z Z C Z Zb s k Z Z k Z Z k Z Z k Z Z (7) 5 1 6 2 7 3 8 41 2 3 4 05 1 6 2 7 3 8 41 2 3 4 I C Z Z a C Z Z b C Z Z b C Z Z ayy b k Z Z a k Z Z b k Z Z b k Z Z a (8) 5 1 6 2 7 3 8 41 2 3 4 05 1 6 2 7 3 8 41 2 3 4 I C Z Z c C Z Z c C Z Z c C Z Z cxx b k Z Z c k Z Z c k Z Z c k Z Z c (9) 05 1 1 1 5 1 1 11 1 1 1m Z C Z Z C Z q k Z Z k Z qt t (10) 06 2 2 2 6 2 2 22 2 2 2m Z C Z Z C Z q k Z Z k Z qt t (11) 07 3 3 3 7 3 3 33 3 3 3m Z C Z Z C Z q k Z Z k Z qt t (12) 08 4 4 4 8 4 4 44 4 4 4m Z C Z Z C Z q k Z Z k Z qt t (13) 5Z Z c ab (14) 6Z Z c bb (15) 7Z Z c bb (16) 8Z Z c ab (17) Downloaded From: http://proceedings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002287_amr.680.422-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002287_amr.680.422-Figure3-1.png", + "caption": "Fig. 3 Dynamic multi-body model of heavy vehicle", + "texts": [ + " In this paper, the concept of a discrete beam in SIMPACK software was used to establish the leaf spring flexible body model. The \u2018magic formula\u2019 tire model which was proposed by professor Pacejka was used to simulate the tire characteristics [12]. The front axle, mid-axle, rear axle, front leaf spring, rear balanced suspension leaf spring, steering and tires subsystem are established in SIMPACK. And each subsystem can be connected by zero degrees of freedom connection to frame. Then a rigid-flexible coupling virtual prototype model of the truck will be established which is shown in Fig.3. Then a driving dynamic model of heavy vehicle was established which combined virtual prototype vehicle with bilateral tracks\u2019 spatial domain road excitation (see Fig.3). The dynamic tires load of the vehicle\u2019s each axle can be calculated by simulation. The normal force of tire. Under the conditions of B-grade bilateral tracks\u2019 spatial domain road excitation, full load and vehicle speed of 60 km/h, the change curves along with driving distance of tire normal load of front axle, mid-axle and rear axle are shown in Fig.4. Seen from Fig.4, under the conditions of bilateral tracks\u2019 road excitation, tire dynamic normal load variation and size are not the same, and the position of the maximum value is also different" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.26-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.26-1.png", + "caption": "Fig. 2.26. Anthropomorphic arm with spherical wrist", + "texts": [ + "24 reveals that the direct kinematics function cannot be obtained by multiplying the transformation matrices T 0 3 and T 3 6, since Frame 3 of the anthropomorphic arm cannot coincide with Frame 3 of the spherical wrist. Direct kinematics of the entire structure can be obtained in two ways. One consists of interposing a constant transformation matrix between T 0 3 and T 3 6 which allows the alignment of the two frames. The other refers to the Denavit\u2013Hartenberg operating procedure with the frame assignment for the entire structure illustrated in Fig. 2.26. The DH parameters are specified in Table 2.6. Since Rows 3 and 4 differ from the corresponding rows of the tables for the two single structures, the relative homogeneous transformation matrices A2 3 and A3 4 have to be modified into A2 3(\u03d13) = \u23a1 \u23a2\u23a3 c3 0 s3 0 s3 0 \u2212c3 0 0 1 0 0 0 0 0 1 \u23a4 \u23a5\u23a6 A3 4(\u03d14) = \u23a1 \u23a2\u23a3 c4 0 \u2212s4 0 s4 0 c4 0 0 \u22121 0 d4 0 0 0 1 \u23a4 \u23a5\u23a6 while the other transformation matrices remain the same. Computation of the direct kinematics function leads to expressing the position and orientation of the end-effector frame as: p0 6 = \u23a1 \u23a3 a2c1c2 + d4c1s23 + d6 ( c1(c23c4s5 + s23c5) + s1s4s5 ) a2s1c2 + d4s1s23 + d6 ( s1(c23c4s5 + s23c5) \u2212 c1s4s5 ) a2s2 \u2212 d4c23 + d6(s23c4s5 \u2212 c23c5) \u23a4 \u23a6 (2", + "70) and n0 6 = \u23a1 \u23a3 c1 ( c23(c4c5c6 \u2212 s4s6) \u2212 s23s5c6 ) + s1(s4c5c6 + c4s6) s1 ( c23(c4c5c6 \u2212 s4s6) \u2212 s23s5c6 ) \u2212 c1(s4c5c6 + c4s6) s23(c4c5c6 \u2212 s4s6) + c23s5c6 \u23a4 \u23a6 s0 6 = \u23a1 \u23a3 c1 ( \u2212c23(c4c5s6 + s4c6) + s23s5s6 ) + s1(\u2212s4c5s6 + c4c6) s1 ( \u2212c23(c4c5s6 + s4c6) + s23s5s6 ) \u2212 c1(\u2212s4c5s6 + c4c6) \u2212s23(c4c5s6 + s4c6) \u2212 c23s5s6 \u23a4 \u23a6 (2.71) a0 6 = \u23a1 \u23a3 c1(c23c4s5 + s23c5) + s1s4s5 s1(c23c4s5 + s23c5) \u2212 c1s4s5 s23c4s5 \u2212 c23c5 \u23a4 \u23a6 . By setting d6 = 0, the position of the wrist axes intersection is obtained. In that case, the vector p0 in (2.70) corresponds to the vector p0 3 for the sole anthropomorphic arm in (2.66), because d4 gives the length of the forearm (a3) and axis x3 in Fig. 2.26 is rotated by \u03c0/2 with respect to axis x3 in Fig. 2.23. Consider the DLR manipulator, whose development is at the basis of the realization of the robot in Fig. 1.30; it is characterized by seven DOFs and as such it is inherently redundant. This manipulator has two possible configurations for the outer three joints (wrist). With reference to a spherical wrist similar to that introduced in Sect. 2.9.5, the resulting kinematic structure is illustrated in Fig. 2.27, where the frames attached to the links are evidenced", + "94) it is pWxc1 + pWys1 \u2212pWz = d3s2 \u2212d3c2 , from which \u03d12 = Atan2(pWxc1 + pWys1, pWz). Notice that, if d3 = 0, then \u03d12 cannot be uniquely determined. Consider the anthropomorphic arm shown in Fig. 2.23. It is desired to find the joint variables \u03d11, \u03d12, \u03d13 corresponding to a given end-effector position pW . Notice that the direct kinematics for pW is expressed by (2.66) which can be obtained from (2.70) by setting d6 = 0, d4 = a3 and replacing \u03d13 with the angle \u03d13 +\u03c0/2 because of the misalignment of the Frames 3 for the structures in Fig. 2.23 and in Fig. 2.26, respectively. Hence, it follows pWx = c1(a2c2 + a3c23) (2.95) pWy = s1(a2c2 + a3c23) (2.96) pWz = a2s2 + a3s23. (2.97) Proceeding as in the case of the two-link planar arm, it is worth squaring and summing (2.95)\u2013(2.97) yielding p2 Wx + p2 Wy + p2 Wz = a2 2 + a2 3 + 2a2a3c3 from which c3 = p2 Wx + p2 Wy + p2 Wz \u2212 a2 2 \u2212 a2 3 2a2a3 (2.98) where the admissibility of the solution obviously requires that \u22121 \u2264 c3 \u2264 1, or equivalently |a2\u2212a3| \u2264 \u221a p2 Wx + p2 Wy + p2 Wz \u2264 a2 +a3, otherwise the wrist point is outside the reachable workspace of the manipulator" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003530_detc2011-48476-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003530_detc2011-48476-Figure2-1.png", + "caption": "Fig. 2. Kinematic chain of the 3-CUP parallel manipulator", + "texts": [ + " NOTATION AND GEOMETRY DESCRIPTION It is well known in literature that parallel mechanisms are named according to the number of limbs and type of joints contained in a kinematic chain. Hence, 3-CUP is a parallel mechanism constructed of three identical and equally distributed legs (i. e. kinematic chains). Consider that each leg has a cylindrical joint (denoted as C) perpendicular to the fixed base, followed by a universal joint (henceforth denoted as U), and finally a prismatic (denoted as P) joint that connects the limb to the moving platform. A schematic diagram of a CUP kinematic chain is shown in Fig. 2. For the convenience of the reader, scalar quantities are represented in this paper by lower case lettering in italic type (e. g. a), vectors by lower case lettering in bold italic type (e. g. di), and matrices by upper case lettering in bold upright type (e. g. R). Cartesian coordinate reference frames are denoted by upper case lettering (e. g. G). Furthermore, upper case lettering in italic type is used for orthogonal axes, centroids, and points fixed in links. Consider as abbreviations of cosine, sine and tangent, C, S and T respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002291_s1064230713020056-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002291_s1064230713020056-Figure4-1.png", + "caption": "Fig. 4. Speed\u2013torque characteristics of impulse modes.", + "texts": [ + " 2 2013 AUTOMATION OF CONSTRUCTION OF CHARACTERISTIC CURVES 255 Ta is the armature time constant; \u03a9avg\u2217 = \u03a9avg/\u03a9id is the average speed in terms of relative units; \u03a9id is the no load speed; and \u03b3 = timp/T is the PWM duty cycle. Based on (2.2), the current pulsation amplitude within the PWM period is (2.3) where = 1/ is the hyperbolic cosecant. Given the constant \u03c4a, the maximum current pul sation amplitude is attained at \u03b3 = 1/2: \u03b4Imax = IS \u03c4a/4). The analytical expression of the average torque in terms of the relative units is (2.4) where Tavg\u2217 = Tavg/TS; TS is the stall torque. Expression (2.4) describes the family of speed\u2013torque char acteristics consisting of parallel straight lines (Fig. 4a). The system of equations (2.5) \u03b4I 2IS \u03c4a 2 \u03b3\u03c4a 2 \u03c4a 1 \u03b3\u2013( ) 2 sinh ,sinhcosech= xcosech xsinh (tanh Tavg* \u03b3 \u03a9avg*,\u2013\u00b1= \u03a9B. I III*, \u03b3( ) 1 e \u03b3\u03c4a\u2013 1 e \u03c4a\u2013 ,\u00b1= TB. I III*, \u03b3( ) \u03b3 1 e \u03b3\u03c4a\u2013 1 e \u03c4a\u2013 \u2013 \u239d \u23a0 \u239c \u239f \u239b \u239e \u00b1= \u23a9 \u23aa \u23aa \u23a8 \u23aa \u23aa \u23a7 256 JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL Vol. 52 No. 2 2013 KRIVILEV is obtained under the condition iI\u2217(0) = iII\u2217(T) = 0; it is used for the construction of borderline curves in the first and third quadrants of the speed\u2013torque plane. The borderline curves in the second and fourth quadrants are described using the system (2", + " By the width of the alternating current region, we mean the distance between the maximum load torque values in the motor and the regeneration regions beginning with which the current in the motor armature has the same sign (positive or negative) during the PWM period at any average angular speed of the motor. The mathematical expression (in terms of relative units) used to determine the width of the alternating current region is (2.11) Therefore, the width of the alternating current region depends on the PWM period. When we have T = 0, the width of the alternating current region is W\u2217 = 0. If \u03c4a \u2264 1/2, then the width of the alternating current region is W\u2217 \u2248 \u03b4Imax\u2217. The alternating current region is outlined by the dotted line in Fig. 4a. 3. THE SECOND IMPULSE MODE Differential equations for impulse mode II are as follows: (3.1) The solution of (3.1) for iI(0) = iII(T) and iI(\u03b3T) = iII(\u03b3T) is (3.2) \u03a9B. II IV*, \u03b3( ) 1 e \u03b3\u03c4a\u2013 1 e \u03c4a\u2013 e 1 \u03b3\u2013( )\u03c4a,\u00b1= TB. II IV*, \u03b3( ) \u03b3 1 e \u03b3\u03c4a\u2013 1 e \u03c4a\u2013 e 1 \u03b3\u2013( )\u03c4a\u2013 \u239d \u23a0 \u239c \u239f \u239b \u239e ,\u00b1= \u23a9 \u23aa \u23aa \u23a8 \u23aa \u23aa \u23a7 TB. I* 1 \u03c4a 1 \u03a9 B. I* 1 e \u03c4a\u2013\u239d \u23a0 \u239b \u239e\u2013\u239d \u23a0 \u239b \u239eln \u03a9 B. I* ,\u2013= TB. II* 1 1 \u03c4a e \u03c4a \u03a9 B. II* 1 e \u03c4a\u2013\u239d \u23a0 \u239b \u239e+\u239d \u23a0 \u239b \u239eln \u03a9 B. II* ,\u2013\u2013= TB. III* 1 \u03c4a \u2013 1 \u03a9 B. III* 1 e \u03c4a\u2013\u239d \u23a0 \u239b \u239e+\u239d \u23a0 \u239b \u239eln \u03a9 B. III* ,\u2013= TB. IV* 1\u2013 1 \u03c4a e \u03c4a \u03a9 B", + "3) taken into account, the current pulsation amplitude for the second impulse mode is two times as large as the current pulsation amplitude for the first impulse mode. Based on (3.2), expressions for the speed\u2013torque characteristics in terms of relative units (3.4) and for the borderline curves between the currents with the same and opposite signs are obtained in para metric and explicit form (3.5) (3.6) The speed\u2013torque characteristics and the borderline curves between the currents with the same and opposite signs for the second impulse mode are shown in Fig. 4b. The width of the alternating current region for the second impulse mode is two times as large as the width of the same region for the first impulse mode. 4. THE THIRD IMPULSE MODE The system of equations for impulse mode III is (4.1) where t0 is the time interval from the beginning of the PWM period to the instant at which the current in the armature changes its sign with zero PWM signal (the lead edge of PWM signal is meant here). \u03b4I 4IS \u03c4a 2 \u03b3\u03c4a 2 \u03c4a 1 \u03b3\u2013( ) 2 .sinhsinhcosech= Tavg* 2\u03b3 1\u2013( )\u00b1 \u03a9avg*\u2013= \u03a9B", + "+ln= UPOW\u00b1 RiI t( ) LiI' t( ) ce\u03a9avg for 0 t timp,\u2264 \u2264+ += UPOW+\u2212 RiII t( ) LiII' t( ) ce\u03a9avg for timp+ + t t0,\u2264 \u2264= 0 RiIII t( ) LiIII' t( ) ce\u03a9avg for t0 t T,\u2264 \u2264+ +=\u23a9 \u23aa \u23a8 \u23aa \u23a7 258 JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL Vol. 52 No. 2 2013 KRIVILEV The solution of Eqs. (4.1) for iI(0) = iIII(T), iI(\u03b3T) = iII(\u03b3T), and iII(t0) = iIII(t0) = 0 is (4.2) The time interval t0 is determined as To determine the current pulsation amplitude, the following expression is to be used Based on (4.2), the description of the speed\u2013torque characteristics for impulse mode III is obtained (Fig. 4c): (4.3) If t0 = T, then (4.3) coincides with (3.4); if t0 = \u03b3T, then it coincides with (2.4). The first variant of (4.3) is correct for 1 > \u03a9avg\u2217 \u2265 0, and the second one for 0 \u2265 \u03a9avg\u2217 > \u20131. In case of the first variant of (4.1), the borderline curve of pulse mode III coincides in the first quadrant (t0 = T, iI(0) = iII(T) = 0) with the borderline curve between the currents with the same and opposite signs for impulse mode II (3.5), and the borderline curve in second quadrant (t0 = \u03b3T, iI(\u03b3T) = iIII(\u03b3T) = 0) coincides with the borderline curve for pulse mode I (2", + "1) for iI(0) = iII(t0) = iIII(T) = 0 and iI(\u03b3T) = iII(\u03b3T) is (5.2) The following expression makes it possible to calculate the length of the interval t0: The current pulsation amplitude under the steady state conditions is calculated using the expression (5.3) The maximum value of \u03b4I is at the boundary of the region of impulse mode IV when \u03b3 = 1/2. If the load torque is zero, then the current pulsation amplitude is zero as well. Based on (5.2), the description of speed\u2013torque characteristics is obtained in the form (Fig. 4) (5.4) For t0 = T, expression (5.4) coincides with (3.4). Since the borderline curves of the pulse mode IV are obtained under the condition t0 = T, their descrip tion coincides with that of the corresponding borderline curves of the currents with the same and opposite signs for impulse mode II. The width of the region of impulse mode IV is equal to the width of the alter nating current region for impulse mode I (2.11). 6. THE FIFTH IMPULSE MODE The system of equations for impulse mode V is (6.1) where t0 is the time interval within the PWM period during which the current is continuous", + "\u2013\u00b1= UPOW\u00b1 RiI t( ) LiI' t( ) ce\u03a9avg for 0 t timp,\u2264 \u2264+ += 0 RiII t( ) LiII' t( ) ce\u03a9avg for timp+ + t t0,\u2264 \u2264= UM ce\u03a9avg for t0 t T,\u2264 \u2264=\u23a9 \u23aa \u23a8 \u23aa \u23a7 iI* t( ) 1\u00b1 \u03a9avg*\u2013( ) 1 e t/Ta\u2013 \u2013\u239d \u23a0 \u239b \u239e for 0 t timp,\u2264 \u2264= iII* t( ) e \u03b3\u03c4a t/Ta\u2013 e t/Ta\u2013 \u2013\u239d \u23a0 \u239b \u239e\u00b1 \u03a9avg* 1 e t/Ta\u2013 \u2013\u239d \u23a0 \u239b \u239e for timp\u2013 t t0,\u2264 \u2264= iIII* t( ) 0 for t0 t T.\u2264 \u2264=\u23a9 \u23aa \u23aa \u23aa \u23a8 \u23aa \u23aa \u23aa \u23a7 t0 Ta 1 e \u03b3\u03c4a 1\u2013 \u03a9avg* \u00b1 \u239d \u23a0 \u239c \u239f \u239b \u239e .ln= 260 JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL Vol. 52 No. 2 2013 KRIVILEV The analytical expression for the calculation of the current pulsation amplitude coincides with (5.3) used for impulse mode IV. Based on (6.2), the description of the speed\u2013torque characteristics for impulse mode V is obtained in the form (Fig. 4e) (6.3) For t0 = T, expression (6.3) coincides with (2.4). Since the borderline curves of impulse mode V on the speed\u2013torque characteristic plane are obtained under the condition iI(0) = iII(T) = 0, their mathematical description coincides with that of the borderline curves of the currents with the same and opposite signs for impulse mode I (2.7) and (2.9). The width of the region of impulse mode V is half the width of the alternating current region for impulse mode I (2.11). 7. COMMUTATION METHODS The expressions for the speed\u2013torque characteristics for five impulse modes and the corresponding borderline curves obtained above may be used for the derivation of the mathematical description of the speed\u2013torque characteristics of arbitrary commutation methods of switching elements", + " The first group includes the S method, the second group includes SN and SA methods, the third, N and A methods, the fourth, ND and AD methods, and the fifth, the D method. 7.1. N and A Methods The N and A methods are characterized by the presence of unipolar voltage pulses on the motor wind ing, which corresponds to impulse mode I. Therefore, the speed\u2013torque characteristics of two variants of the N and A methods are described by expression (2.4), and their graphical representation is shown in Fig. 4a. 7.2. S Method Presence of bipolar voltage pulses on the winding is peculiar for S method, which is peculiar for the pulse mode II. Thus, the description of speed\u2013torque characteristics of the S method coincides with the description of speed\u2013torque characteristics of the pulse mode II (3.4). One of two possible variants of (3.4) has to be used for construction of the speed\u2013torque characteristics (Fig. 4b). 7.3. D Method In the case of D method, there are bipolar pulses (UM = {+UPOW, \u2013UPOW} or UM = {\u2013UPOW, +UPOW}) on the winding or bipolar pulses with the back emf at the end of the PWM period (UM = {+UPOW, \u2013UPOW, } Tavg* \u03b3 \u03a9avg*t0/T.\u2013\u00b1= JOURNAL OF COMPUTER AND SYSTEMS SCIENCES INTERNATIONAL Vol. 52 No. 2 2013 AUTOMATION OF CONSTRUCTION OF CHARACTERISTIC CURVES 261 or UM = {\u2013UPOW, +UPOW, }) depending on the load torque and the PWM duty cycle. Therefore, the expression for the speed\u2013torque characteristics of the D method (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002117_2013-01-2011-Figure13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002117_2013-01-2011-Figure13-1.png", + "caption": "Figure 13. Tunnel Matchboxing mode at 1.5Hz", + "texts": [ + "2mm respectively while the front right and the rear right move down by 5.4mm and 7.4mm respectively. A similar roll mode is observed at every time slice of the data. Likewise, a time slice at any point in the lateral displacement data shows that all four cab mounts are moving in phase at all times in the lateral direction. Refer Figure 12. The lateral swaying of the cab does not deform the cab and only the effect of mass inertia might lead to any damage. The roll mode causes tunnel Matchboxing of the cab as shown in figure 13. PSD plot of the displacements measured on the cab show that the roll mode frequency is 1.5Hz as shown in Figure 14. Since the cab or the components mounted in the cab do not have any mode below 2Hz, it is safe to assume that modal amplification of the cab is not likely for the Automatic Side Loading operation. Due to the tunnel Matchboxing caused by the roll mode of the cab, the corner of the tunnel lights up on the back panel of the cab. The stresses obtained from the Inertia Relief analysis are at a single time slice of the Automatic Side Loading operation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002782_ecj.10179-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002782_ecj.10179-Figure1-1.png", + "caption": "Fig. 1. Diagram of rotary inverted pendulum.", + "texts": [ + " The proposed control thus proved effective as an adaptive control method for unstable unknown-parameter systems allowing superposition of disturbance signals. A simplified motion diagram of a rotary inverted pendulum and its experimental model are shown in Figs. 1 and 2, respectively. The rotary arm of the first link is coupled directly with a DC motor (100 V, 250 W) and the pendulum portion can be extended or contracted by a mini DC motor (24 V, 6.4 W) via a rack-and-pinion mechanism. The equation of motion for the controlled object in Fig. 1 can be derived by using the Euler\u2013Lagrange equation as follows: Here \u03b8, \u03b8 . , and \u03b8 .. are respectively the angle, angular velocity, and angular acceleration, \u03c4 is the torque, M(\u03b8) is the inertial term, C(\u03b8, \u03b8 . )\u03b8 . is the nonlinear term, G(\u03b8) is the gravity term, B\u03b8 . is the viscous friction term, and D(\u03b8 . ) is the Coulomb friction term: Here J1 = I1 + m1a1 2 + m2l 2, l = l1/2, J2 = I2 + m2a2 2, r = m2a2, S2 = sin\u03b82, C2 = cos\u03b82. In addition, g is the gravitational acceleration, \u03c41 is the input torque from the motor, b1 and b2 are viscous friction coefficients, and d1 is the Coulomb fiction coefficient of the arm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003423_humanoids.2014.7041380-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003423_humanoids.2014.7041380-Figure1-1.png", + "caption": "Fig. 1. Definition of variables [7]", + "texts": [ + " In this way, several joints of the robot will be given a required high priority motion, while the remaining ones may be given constant values, or any low priority motion. Then, once all these motions are defined (or proposed) and specifically their velocity profiles, the total linear and angular momenta of the robot with respect to some reference point (as it is the origin of the world frame 0), P0 and L0 respectively, will also be inherently given. Let us denote by \u03c4p the net moment of the vertical ground reaction force fp about the ZMP, whose position is denoted by rp/0 as it is taken from the origin of the world frame, 0 (see Fig. 1). The moment of fp about the origin of the world coordinate frame, \u03c40, can be calculated as \u03c40 = rp/0 \u00d7 fp + \u03c4p. (1) 978-1-4799-7174-9/14/$31.00 \u00a92014 IEEE 329 In addition, we know that the following relationships hold: P\u03070 = m\u0303g + fp, (2) L\u03070 = rc/0 \u00d7 m\u0303g + \u03c40, (3) where g = [ 0 0 \u2212g ]T is the gravity vector and g is the acceleration due to gravity, whereas m\u0303 stands for the total mass of the robot and rc/0, for the position of the center of mass (CoM) of the robot with respect to the origin of the world coordinate frame" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001195_978-81-322-1656-8_29-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001195_978-81-322-1656-8_29-Figure2-1.png", + "caption": "Fig. 2 Configuration of a typical bump type GFB [12]", + "texts": [ + " The underlying compliant structure (bumps) provides a tenable structural stiffness source [3, 4] resulting in a larger film thickness than the rigid wall bearings [5, 6] enabling high-speed operation and larger load capacity including tolerance for shaft misalignment [2]. Also damping of coulomb type arises due to the relative motion between the bumps and the top foil, and between the bumps and the bearing wall [7]. Foil Bearings generally operates with ambient air. However, some specific applications use other fluids such as helium, xenon, liquid nitrogen, and liquid oxygen among others [8]. Significant improvements in hightemperature limits Fig. 2 are obtained by using coatings (solid lubricants) [9, 10]. Since late 1960s, gas foil bearings are common in air cycle machines, the heart of the environment control system in aircraft. Current applications of foil bearings include oil-free cryogenic turbo-expanders for gas separation plants, auxiliary power units for various aerospace and ground vehicles, automotive gas turbine engines, vapour-cycle centrifugal compressors, and commercial air/gas compressors [7]. Peng and Carpino [6] developed a finite difference formulation, coupling hydrodynamic and elastic foundation effects to calculate stiffness and damping force coefficients in foil bearings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001967_iet-cta.2014.0667-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001967_iet-cta.2014.0667-Figure2-1.png", + "caption": "Fig. 2 Inter-connected pendulum system", + "texts": [ + " A surprising fact observed in [16] is that as the number of agents increases a graph is more likely to be controllable. This fact, combined with Theorem 3, gives a counterintuitive result that, when H = A, the output sum ys or average ya is more and more likely to have enough information for collective stabilisation by a SAC controller, as the number of agents increases. For an illustration of our developments and findings up to now, we present a physical example in this section. Consider a multi-agent system composed of three identical inverted pendulums shown in the right side of Fig. 2. The pendulums are mechanically coupled by springs whose characteristic constants are the same k . In addition, all springs are neither stretched nor compressed when \u03b81 = \u03b82 = \u03b83 = 0. Assuming that only the leftmost pendulum is able to accept an external force input f , our aim is to stabilise three pendulum angles under the SAC scheme. For small angles, one can easily find from the left diagram of Fig. 2 that each pendulum is an unstable second-order system \u03b8i(s) ui(s) = 1/mL s2 \u2212 g/L (17) where ui(s) denotes the horizontal force. Merging the external force f and spring forces, we can model the pendulum 932 IET Control Theory Appl., 2015, Vol. 9, Iss. 6, pp. 929\u2013934 \u00a9 The Institution of Engineering and Technology 2015 doi: 10.1049/iet-cta.2014.0667 system as a consensus multi-agent system d(s) n(s) [ \u03b81 \u03b82 \u03b83 ] = \u2212H [ \u03b81 \u03b82 \u03b83 ] + [ 1 0 0 ] ue 1, H = [ 1 \u22121 0 \u22121 2 \u22121 0 \u22121 1 ] (18) where n(s) = k m , d(s) = s2 \u2212 g L , ue 1 := f kL (19) The symmetric matrix H in (18) has three eigenvalues \u03bc1 = 0, \u03bc2 = 1, \u03bc3 = 3 and eigenvectors v = [v1 | v2 | v3] = 1\u221a 6 \u23a1 \u23a3 \u221a 2 \u221a 3 \u22121\u221a 2 0 2\u221a 2 \u2212\u221a 3 \u22121 \u23a4 \u23a6 (20) From Pk = vkvT k , the parameters {\u03b1ik} in (7) can be found to be (\u03b1ik) = 1 6 [ 2 3 1 2 0 \u22122 2 \u22123 1 ] (21) and thus the transfer functions from the external input ue 1 to pendulum angle are given as g1(s) = \u03b81/ue 1 = 1 6mL \u00d7 ( 2 s2 \u2212 g/L + 3 s2 \u2212 g/L + k/m + 1 s2 \u2212 g/L + 3k/m ) g2(s) = \u03b82/ue 1 = 1 6mL ( 2 s2 \u2212 g/L + \u22122 s2 \u2212 g/L + 3k/m ) g3(s) = \u03b83/ue 1 = 1 6mL \u00d7 ( 2 s2 \u2212 g/L + \u22123 s2 \u2212 g/L + k/m + 1 s2 \u2212 g/L + 3k/m ) From Theorem 1, we can conclude that any SAC controller stabilising g1 with a measurement of \u03b81 stabilises all three pendulum angles collectively, whereas controllers stabilising g2 with a measurement of \u03b82 may not stabilise {\u03b81, \u03b83}" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002968_gt2011-45903-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002968_gt2011-45903-Figure3-1.png", + "caption": "Fig. 3. Scheme of the coupling misalignment.", + "texts": [ + " Otherwise, like described in this paper, it is necessary to take into account the effect of rigid coupling misalignment on the static centreline and, as a consequence, on the reactions of the bearing, considering that these reactions are changing due 2 Copyright \u00a9 2011 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/30/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use to the rotation of the shaft, i.e. to the orientation of the misalignment with respect to the phase reference. Effect of rigid coupling misalignment on the shaft-line Let\u2019s consider Fig. 3 in which a close up of a rigid coupling of the machine is shown. For the sake of simplicity, only a coupling is considered, being the model presented easily generalizable to Cn couplings. In the general case, the coupling faces are connected in correspondence of the jC th node and both radial and angular misalignment may occur as a consequence of wrong mounting or imperfect machining. However, not only the magnitudes of these misalignments have to be considered, but also the relative phase with respect to the phase reference and the fact that the shaft is rotating with rotational speed " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003721_detc2014-34297-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003721_detc2014-34297-Figure2-1.png", + "caption": "FIGURE 2. Serially connected kinematic chain corresponding to the direct sum of two subalgebras t\u22a5u\u03021 \u2295 sO.", + "texts": [ + " This section shows the application of screw systems of locally constant rank in the synthesis of legs of parallel platforms. This task faces two possible scenarios: 1. Direct sum of subalgebras. In this case dim(A1\u2295A2) = dim(A1)+dim(A2) (1) The number of screws \u2014and therefore the number of kinematic pairs\u2014 associated with the mechanical generators of the subalgebra is equal to the corresponding dimension of the subalgebra. On this regard, consider a serially connected kinematic chain shown in Figure 2. In this case, the direct sum of two subalgebras is given by t\u22a5u\u03021 \u2295 sO, where t\u22a5u\u03021 , represents the subalgebra associated with translations in a plane perpendicular to u\u03021 while sO, represents the subalgebra associated with the spherical displacements around the point O. It can be proved that the dimension of the screw system associated with the serially connected chain is 5, furthermore, the ordered screw system have constant rank, see definition 2. 2. Not-direct sum of subalgebras. In this case dim(A1 +A2)< dim(A1)+dim(A2) (2) It should be noted that, in the synthesis of legs of parallel platforms, is undesirable to have passive or redundant degrees of freedom", + "org/about-asme/terms-of-use with the serially connected chain is 5, furthermore, the ordered screw system have locally constant rank, see definition 2 and corollary 1. Figure 3(b), shows the serially connected kinematic chain where the redundant kinematic pair has been removed. It can be proved that the dimension of the screw system associated with the serially connected chain is 5, furthermore, the ordered screw system have locally constant rank, see proposition 2. Consider the serially connected kinematic chains shown in Figure 2, where the ordered screw system of locally constant rank is generated by the direct sum t\u22a5u\u03021 \u2295 sO. Simply changing the direction of the first revolute pair of the subalgebra sO, could be considered like a serially connected kinematic chains generated by the not-direct sum gu\u03021 +sO. Figure 4 shows this example after eliminating the redundant kinematic pair. It can be proved that the dimension of the screw system associated with the serially connected chain is 5, furthermore, the ordered screw system have locally constant rank, see proposition 2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure8.37-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure8.37-1.png", + "caption": "Fig. 8.37 Equivalent stress (von Mises). Distribution of loads in the lateral side of the healthy foot", + "texts": [], + "surrounding_texts": [ + "The results shown in Tables 8.3 and 8.4 indicate that the ortheses, printed in 3D in CILAB, are adequate for the patient. The maximum principal stress in the ortheses doesn\u2019t exceed the yield stress of the material which is equal to 30.70 MPa. This suggested that there is no plastic deformation that affects the ortheses. When the results of FEM and interferometry are compared, we observe the zones, where the results match. The displacements that were been registered in the interferometry" + ] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.13-1.png", + "caption": "Fig. 2.13. Description of the position and orientation of the end-effector frame", + "texts": [ + " It was previously illustrated that the pose of a body with respect to a reference frame is described by the position vector of the origin and the unit vectors of a frame attached to the body. Hence, with respect to a reference frame Ob\u2013xbybzb, the direct kinematics function is expressed by the homogeneous transformation matrix T be(q) = \u23a1 \u23a2\u23a3n b e(q) sbe(q) abe(q) pbe(q) 0 0 0 1 \u23a4 \u23a5\u23a6 , (2.48) where q is the (n\u00d7 1) vector of joint variables, ne, se, ae are the unit vectors of a frame attached to the end-effector, and pe is the position vector of the origin of such a frame with respect to the origin of the base frame Ob\u2013xbybzb (Fig. 2.13). Note that ne, se, ae and pe are a function of q. The frame Ob\u2013xbybzb is termed base frame. The frame attached to the endeffector is termed end-effector frame and is conveniently chosen according to the particular task geometry. If the end-effector is a gripper, the origin of the end-effector frame is located at the centre of the gripper, the unit vector ae is chosen in the approach direction to the object, the unit vector se is chosen normal to ae in the sliding plane of the jaws, and the unit vector ne is chosen normal to the other two so that the frame (ne, se,ae) is right-handed", + "50) yields T 0 3(q) = A0 1A 1 2A 2 3 = \u23a1 \u23a2\u23a3 c123 \u2212s123 0 a1c1 + a2c12 + a3c123 s123 c123 0 a1s1 + a2s12 + a3s123 0 0 1 0 0 0 0 1 \u23a4 \u23a5\u23a6 (2.63) where q = [\u03d11 \u03d12 \u03d13 ]T . Notice that the unit vector z0 3 of Frame 3 is aligned with z0 = [ 0 0 1 ]T , in view of the fact that all revolute joints are parallel to axis z0. Obviously, pz = 0 and all three joints concur to determine the end-effector position in the plane of the structure. It is worth pointing out that Frame 3 does not coincide with the end-effector frame (Fig. 2.13), since the resulting approach unit vector is aligned with x0 3 and not with z0 3. Thus, assuming that the two frames have the same origin, the constant transformation T 3 e = \u23a1 \u23a2\u23a3 0 0 1 0 0 1 0 0 \u22121 0 0 0 0 0 0 1 \u23a4 \u23a5\u23a6 . is needed, having taken n aligned with z0. Consider the parallelogram arm in Fig. 2.21. A closed chain occurs where the first two joints connect Link 1\u2032 and Link 1\u2032\u2032 to Link 0, respectively. Joint 4 was selected as the cut joint, and the link frames have been established accordingly", + " The homogeneous transformation matrices defined in (2.52) are for the single joints: A0 1(\u03d11) = \u23a1 \u23a2\u23a3 c1 0 s1 0 s1 0 \u2212c1 0 0 1 0 0 0 0 0 1 \u23a4 \u23a5\u23a6 Ai\u22121 i (\u03d1i) = \u23a1 \u23a2\u23a3 ci \u2212si 0 aici si ci 0 aisi 0 0 1 0 0 0 0 1 \u23a4 \u23a5\u23a6 i = 2, 3. Computation of the direct kinematics function as in (2.50) yields T 0 3(q) = A0 1A 1 2A 2 3 = \u23a1 \u23a2\u23a3 c1c23 \u2212c1s23 s1 c1(a2c2 + a3c23) s1c23 \u2212s1s23 \u2212c1 s1(a2c2 + a3c23) s23 c23 0 a2s2 + a3s23 0 0 0 1 \u23a4 \u23a5\u23a6 (2.66) where q = [\u03d11 \u03d12 \u03d13 ]T . Since z3 is aligned with z2, Frame 3 does not coincide with a possible end-effector frame as in Fig. 2.13, and a proper constant transformation would be needed. Consider a particular type of structure consisting just of the wrist of Fig. 2.24. Joint variables were numbered progressively starting from 4, since such a wrist is typically thought of as mounted on a three-DOF arm of a six-DOF manipulator. It is worth noticing that the wrist is spherical since all revolute axes intersect at a single point. Once z3, z4, z5 have been established, and x3 has been chosen, there is an indeterminacy on the directions of x4 and x5", + "50) yields T 3 6(q) = A3 4A 4 5A 5 6 = \u23a1 \u23a2\u23a3 c4c5c6 \u2212 s4s6 \u2212c4c5s6 \u2212 s4c6 c4s5 c4s5d6 s4c5c6 + c4s6 \u2212s4c5s6 + c4c6 s4s5 s4s5d6 \u2212s5c6 s5s6 c5 c5d6 0 0 0 1 \u23a4 \u23a5\u23a6 (2.67) where q = [\u03d14 \u03d15 \u03d16 ]T . Notice that, as a consequence of the choice made for the coordinate frames, the block matrix R3 6 that can be extracted from T 3 6 coincides with the rotation matrix of Euler angles (2.18) previously derived, that is, \u03d14, \u03d15, \u03d16 constitute the set of ZYZ angles with respect to the reference frame O3\u2013x3y3z3. Moreover, the unit vectors of Frame 6 coincide with the unit vectors of a possible end-effector frame according to Fig. 2.13. The so-called Stanford manipulator is composed of a spherical arm and a spherical wrist (Fig. 2.25). Since Frame 3 of the spherical arm coincides with Frame 3 of the spherical wrist, the direct kinematics function can be obtained via simple composition of the transformation matrices (2.65), (2.67) of the previous examples, i.e., T 0 6 = T 0 3T 3 6 = \u23a1 \u23a2\u23a3n 0 s0 a0 p0 0 0 0 1 \u23a4 \u23a5\u23a6 . Carrying out the products yields p0 6 = \u23a1 \u23a3 c1s2d3 \u2212 s1d2 + ( c1(c2c4s5 + s2c5) \u2212 s1s4s5 ) d6 s1s2d3 + c1d2 + ( s1(c2c4s5 + s2c5) + c1s4s5 ) d6 c2d3 + (\u2212s2c4s5 + c2c5)d6 \u23a4 \u23a6 (2" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001730_0954406215603740-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001730_0954406215603740-Figure2-1.png", + "caption": "Figure 2. Representation of the positioning of the spline coupling (a) and the iso-Ruiz contour map (b).", + "texts": [ + " The first Ruiz parameter, in this paper, was calculated for a rectangular array of points on the spline tooth surface, yielding an iso-Ruiz contour map (see Figures 2 and 3). Figure 3 shows a typical iso-Ruiz map obtained for a spline coupling under a 700N m torque with a misalignment angle of 1.5 mrad. at Middle East Technical Univ on January 17, 2016pic.sagepub.comDownloaded from Ruiz parameter values were calculated in a rectangular array of points covering contact area, the largest values of Ruiz parameter identifying critical points in terms of wear damage. Taking the reference axes given in Figure 2, the adopted analytical formula is R X,Y\u00f0 \u00de \u00bc COF P X,Y\u00f0 \u00de X,Y\u00f0 \u00de \u00f02\u00de where for each point defined by the coordinates X and Y, the relevant pressure P(X,Y) between the teeth in contact and the slip amplitude d(X,Y) is considered. Teeth geometries (shaft and hub) may be represented as convex surfaces in contact with a plane; Hertz formulas9 were accordingly used to calculate pressures at contact points P X,Y\u00f0 \u00de \u00bc P0 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 X b 2 Y a 2s \u00f03\u00de where P0 is the maximum value of the contact pressure obtained by means of the following equation P0 \u00bc 3 2 F a b \u00f04\u00de where a and b are the semi-axes of the elliptic contact area, calculated as a \u00bc q \u00f05\u00de and b \u00bc q: \u00f06\u00de To determine the q term, the following equation was used q \u00bc ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3 8 1 \u00fe 2P F 3 s \u00f07\u00de The coefficients of the Hertzian formulations l and t were determined by five equations system and i take into account the properties of the materials in contact; the equation to calculate the latter is i \u00bc 1 2i Ei 4 \u00f08\u00de being vi the Poisson coefficient and Ei the Young modulus of the material, while P is the curvature of the two surfaces in contact and the correspondence equation is X \u00bc 1 R2 1 R1 \u00fe 1 Rb 1 sin \u00f09\u00de where R1 is the real pitch radius of the shaft, Rb is the crowning radius along the face width of the shaft, R2 is the pitch radius of the hub and a is the pressure angle" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002582_0954406211404854-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002582_0954406211404854-Figure4-1.png", + "caption": "Fig. 4 The generalised Altmann loop\u2019s link-based overall frame of reference", + "texts": [ + " Substitution of these results into equation (4) with i\u00bc 2 gives a34 a34 \u00fe a12c 1\u00f0 \u00de \u00fe a23 ffiffiffiffiffi F1 p a23 ffiffiffiffiffi F1 p \u00fe a34 a34 \u00fe a12c 1\u00f0 \u00de \u00bc a12 a12 \u00fe a34c 1\u00f0 \u00de \u00fe a23 ffiffiffiffiffi F1 p a23 ffiffiffiffiffi F1 p \u00fe a12 a12 \u00fe a34c 1\u00f0 \u00de from which it is immediate that \u00bc Hence, generally, the loop possesses two linesymmetric closure modes. 4 INTERSECTIONS OF THE REVOLUTE AXES By the nature of the trihedral linkage, as exemplified in Fig. 2, axes 1, 3, and 5 intersect at the single-point P, as do axes 2, 4, and 6 at point Q. These locations are readily calculated with the aid of the rotation matrices of Section 2, and in accordance with the layout of Fig. 4, as follows. Use is made throughout of equations (1). n1 \u00bc 0 0 1 0 B@ 1 CA, n2 \u00bc s 1 c 1 0 0 B@ 1 CA, n3 \u00bc c 1s 2 s 1s 2 c 2 0 B@ 1 CA, n4 \u00bc s 2c 3 c 2 s 2s 3 0 B@ 1 CA, n5 \u00bc s 3 0 c 3 0 B@ 1 CA, n6 \u00bc 0 1 0 0 B@ 1 CA n12 \u00bc c 1 s 1 0 0 B@ 1 CA, n23 \u00bc c 1c 2 s 1c 2 s 2 0 B@ 1 CA, n45 \u00bc c 2c 3 s 2 c 2s 3 0 B@ 1 CA, n56 \u00bc c 3 0 s 3 0 B@ 1 CA, n61 \u00bc 1 0 0 0 B@ 1 CA Proc. IMechE Vol. 225 Part C: J. Mechanical Engineering Science at MCMASTER UNIV LIBRARY on July 1, 2015pic.sagepub.comDownloaded from Denoting the relevant axis offsets by 1\u2013 6, then, it is clear that 1 0 0 1 0 B@ 1 CA \u00bc a12 c 1 s 1 0 0 B@ 1 CA\u00fe a23 c 1c 2 s 1c 2 s 2 0 B@ 1 CA \u00fe 3 c 1s 2 s 1s 2 c 2 0 B@ 1 CA \u00bc a34 1 0 0 0 B@ 1 CA a23 c 3 0 s 3 0 B@ 1 CA\u00fe 5 s 3 0 c 3 0 B@ 1 CA, from which it is found that 1 \u00bc a23 \u00fe a12c 2 s 2 , 3 \u00bc a12 \u00fe a23c 2 s 2 , 5 \u00bc a34 \u00fe a23c 3 s 3 : Similarly, a34 1 0 0 0 B@ 1 CA\u00fe 6 0 1 0 0 B@ 1 CA \u00bc a12 c 1 s 1 0 0 B@ 1 CA\u00fe 2 s 1 c 1 0 0 B@ 1 CA \u00bc a34 1 0 0 0 B@ 1 CA a23 c 3 0 s 3 0 B@ 1 CA a12 c 2c 3 s 2 c 2s 3 0 B@ 1 CA \u00fe 4 s 2c 3 c 2 s 2s 3 0 B@ 1 CA which, with the help of equations (4), yield 6 \u00bc 3, 2 \u00bc 5, 4 \u00bc 1, as expected" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003982_saci.2014.6840063-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003982_saci.2014.6840063-Figure5-1.png", + "caption": "Figure 5. The slip-way control", + "texts": [ + " The manipulator is consist from a linear actuator (cylinder), which is lifting and lowering the manipulator with a 4/2 electro-pneumatically valve, and a rotational \u201cpneumatically motor\u201d, what is turning the gripper in 900 degrees, from the conveyor position to the selected spout position. The grippers are actuated through a 3/2 valve with spring returning. 6. The slip-way positioning The last (fourth) pneumatically unit is consist from two 4/2 valves situated oppositely to each other. Each cylinder can have 2 positions (\u00c9511, \u00c9512 and \u00c9522, \u00c9521); through the slip-way can have 4 defined positions. The pneumatic control circuits see on Fig. 5. IV. THE MECHATRONIC SYSTEM OPERATION The equipment is a PLC controlled selecting machine, which are sorting the specimens based on their material. The specimens are moved by pneumatic cylinders and valves, and the valves are controlled by the PLC. The material of the specimens is stated based on evaluation of three different sensory data (inductive sensor, capacitive sensor and an opticsensor). In this section the basic movements of the cylinders are discussed. First of all let\u2019s have a look to the basic position, when all the cylinders are in starting position" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001999_jrc2013-2545-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001999_jrc2013-2545-Figure4-1.png", + "caption": "Figure 4: Unitruck running gear and expanded view of its components.", + "texts": [ + " It is proportional to the contact pressure \ud835\udc5d(\ud835\udc41/\ud835\udc5a2), the sliding distance for that element \u2206\ud835\udc60(\ud835\udc5a) and inversely proportional to the hardness of the worn material \ud835\udc3b(\ud835\udc41/\ud835\udc5a2). \u2206 \u2206 (2) The proportionality is represented by the Wear Coefficient (\ud835\udc58). This has been determined in laboratory measurements and depends on sliding velocity and contact pressure. It is represented by a wear chart (Figure 3) with four different regions that represent different wear mechanisms. Vehicle model The proposed vehicle for this study is a two axle vehicle with Unitruck running gear (Figure 4) composed of two short coupled units with UIC standard side buffers. The axle load is 5.8 tons for an unladen vehicle and 22.5 to 25 tons for a laden vehicle. The multibody simulation software GENSYS is used for the analysis. The model is composed of fifteen bodies: carbody, two wheelsets, four saddles and eight massless wedges. All bodies are rigid. The model has 90 degrees of freedom (dof). The flexibility of the track in the lateral direction is also considered for a total of 94 dof, and track irregularities based on experimental measurements are introduced" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.36-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.36-1.png", + "caption": "Fig. 2.36. SCARA manipulator", + "texts": [ + " Prove that the quaternion product is expressed by (2.37). 2.10. By applying the rules for inverting a block-partitioned matrix, prove that matrix A1 0 is given by (2.45). 2.11. Find the direct kinematics equation of the four-link closed-chain planar arm in Fig. 2.34, where the two links connected by the prismatic joint are orthogonal to each other. 2.12. Find the direct kinematics equation for the cylindrical arm in Fig. 2.35. 2.13. Find the direct kinematics equation for the SCARA manipulator in Fig. 2.36. 2.14. Find the complete direct kinematics equation for the humanoid manipulator in Fig. 2.28. 2.15. For the set of minimal representations of orientation \u03c6, define the sum operation in terms of the composition of rotations. By means of an example, show that the commutative property does not hold for that operation. 2.16. Consider the elementary rotations about coordinate axes given by infinitesimal angles. Show that the rotation resulting from any two elementary rotations does not depend on the order of rotations", + " Further, define R(d\u03c6x, d\u03c6y, d\u03c6z) = Rx(d\u03c6x)Ry(d\u03c6y)Rz(d\u03c6z); show that R(d\u03c6x, d\u03c6y, d\u03c6z)R(d\u03c6\u2032 x, d\u03c6 \u2032 y, d\u03c6 \u2032 z) = R(d\u03c6x + d\u03c6\u2032 x, d\u03c6y + d\u03c6\u2032 y, d\u03c6z + d\u03c6\u2032 z). 2.17. Draw the workspace of the three-link planar arm in Fig. 2.20 with the data: a1 = 0.5 a2 = 0.3 a3 = 0.2 \u2212\u03c0/3 \u2264 q1 \u2264 \u03c0/3 \u2212 2\u03c0/3 \u2264 q2 \u2264 2\u03c0/3 \u2212 \u03c0/2 \u2264 q3 \u2264 \u03c0/2. 2.18. With reference to the inverse kinematics of the anthropomorphic arm in Sect. 2.12.4, discuss the number of solutions in the singular cases of s3 = 0 and pWx = pWy = 0. 2.19. Solve the inverse kinematics for the cylindrical arm in Fig. 2.35. 2.20. Solve the inverse kinematics for the SCARA manipulator in Fig. 2.36." + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001136_978-3-319-19788-3_1-Figure1.16-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001136_978-3-319-19788-3_1-Figure1.16-1.png", + "caption": "Fig. 1.16 Examples of spatial robots with 2 DOF able to position a device in a plane with a constant orientation of the platform. a The Par2 from the LIRMM, France (Pierrot et al. 2009). b The IRSBot-2 robot from the IRCCyN, France (Briot et al. 2012b)", + "texts": [ + " However, in order to increase the stiffness of robots with 1Throughout this book, when we mention the number and types of DOF of the PKM, we refer to the number and types of DOF of its mobile platform. The number of DOF of the mobile platform is denoted ndof while the number of DOF of the entire robot is denoted Ndof planar motions of the platform, especially in the direction normal to the displacement plane, a recent idea was to design spatial robots able to achieve planar motions of their platform (Fig. 1.16). The large majority of PKM have been designed in order to be able to move their platform in the space. We call them the Spatial Parallel Manipulators (SPM). The robots of this category are too numerous to mention all of them. However, we can cite: \u2022 robots with three translationalDOF (also called translational parallel manipulators (TPM)): among them, we can mention the Delta robot (Clavel 1990) (Fig. 1.17a), the Orthoglide (Chablat and Wenger 2003) (Fig. 1.17b), the Tripteron (Gosselin 2009) (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003999_s12283-014-0158-y-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003999_s12283-014-0158-y-Figure2-1.png", + "caption": "Fig. 2 A general oblique impact between balls (x0 T direction is given by the right-hand grip rule)", + "texts": [ + " As potting accuracy and precisely predicting final cue ball positions after a shot is of paramount importance in pocket billiards, the high deviation between the actual and predicted ball trajectories will affect the accuracy of the simulated game play. Therefore, analysis is required regarding the very general impact between snooker or billiards balls to provide realistic simulations. In the following analysis, initially, a general solution will be derived for the problem of two identical balls colliding obliquely, and at the end, the values applicable for snooker will be substituted. Here, the problem of the cue ball, C, obliquely impinging on to another object ball O is analyzed (see Fig. 2). Both the cue and object balls are assumed to have equal mass and radius. In Fig. 2, it is important to note that ball C does not spin about its frontal axis (about the direction of V0), this condition is only prevalent during a masse\u0301 shot and is not normally encountered in snooker. When two spheres collide, a contact is made over a finite size of area on their surface, due to the deformation present at the interface. The contact area between the spheres during impact is usually estimated through the Hertz theory. However, a point contact is assumed here. The assumption of a point contact has also been used by other researchers such as Kondic [22] and Domenech [23]", + " 4 is used in conjunction with these numerical simulations to obtain the values of the coefficient of restitution and the value of sliding friction. The fundamental idea is to replicate the experimental results by numerical simulations, using two random numerical values for the above parameters by a trial-and-error procedure. The experimental plot shown in Fig. 4 was obtained under the conditions of cut angle (h), sidespin (xS 0) and topspin (xT 0 ) assuming the values of, 0 , 0 and V0 R , respectively (see Fig. 2). For each of the incident speed values, V0, given in Fig. 4, the value of centroid velocity of the object ball at the termination of impact, _yO G P f I , was found numerically for the restitution coeffi- cient between the balls, e, in the range 0.7\u20131.0 and the friction coefficient (lbb) between 0 and 0.2, both in 0.01 increments. For given values of friction and restitution coefficients between the balls, i.e., lbb and e, the RMS value of all the errors between the experimental and the numerical values for each of the incident velocities given in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002868_s12206-015-0143-9-Figure24-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002868_s12206-015-0143-9-Figure24-1.png", + "caption": "Fig. 24. Schematics and notations for CP Guidance.", + "texts": [ + " The altered schematic and notations are shown in Fig. 23. The LOS is calculated as the heading angle referenced to the northerly direction, and the heading command cmdy is tilted by the arctangent of /N cmd cV V . If the aircraft\u2019s flight path is a series of waypoints, and if the direction of the course is changed by a certain amount of angle, the PN guidance may create an abrupt change in heading angle and side slip, which results in abrupt roll rate changes. Therefore, circular path guidance is used during a change in path direction - see Fig. 24. During waypoints #1 and #2, the aircraft was flying using PN guidance. As it reached waypoint #2, and the direction of the course changed suddenly to waypoint #3, a new point for the center of the circle was provided with a radius of R, and a target turn angle of Tz where, for example, 180\u00b0 of Tz means a half circle turn. cz is a completed turn angle that starts from zero degree when circular turn is initiated, and it determines the exit condition of c Tz z\u00b3 . In this case, the center of the circle was treated as a reference point for \u03bb, and in any position, the heading command should be tangential to the circular path given by: 2cmd cIpy l= + \u00b4 ( cI = +1 for ccw, -1 for cw)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002212_icgccee.2014.6922332-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002212_icgccee.2014.6922332-Figure2-1.png", + "caption": "Figure 2 Flux controller Figure 3 Torque controller", + "texts": [ + " DTC is a vector control method used to control the torque and speed of the motor by controlling the switching sequence of the inverter transistors. The Figure1shows the block diagram of conventional DTC. The actual and reference values of electromagnetic torque and stator flux are compared and the errors are used as inputs for the hysteresis controllers, which maintain the torque and flux errors within upper and lower limits allowed. Depending on the output from the controllers and the sector number, the switching logic determines the optimum switching vector to the inverter. The Figure 2 shows two level flux hysteresis controller and the Figure 3 shows three level torque hysteresis controller. The calculated magnitude of stator flux and electromagnetic torque are compared with their reference values in their corresponding hysteresis comparators. The entire section is divided into 6 sectors and each section having the angle of 60\u00b0. There are eight vectors from V0 to V7 among whichV1 to V6 are the active vectors, V0 and V7 are null vectors. According to the operation principle of DTC, the selection of a voltage vector is made as per in Table1 to maintain the electromagnetic torque and stator flux within the limits of two hysteresis bands" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002357_amm.571-572.326-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002357_amm.571-572.326-Figure1-1.png", + "caption": "Fig. 1 Bicycle model of the vehicle", + "texts": [ + " No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans Tech Publications, www.ttp.net. (ID: 142.103.160.110, University of British Columbia, Kelowna, Canada-04/07/15,16:49:20) Vehicle kinematic model The vehicle travels at a low speed in parking process, during which the side force will not appear. So there is no lateral sliding when the wheel is rolling. Thus, the vehicle is simplified as a non-complete model, namely vehicle kinematic model in the paper. As shown in figure 1, \u03b4 is front wheel angle; L is wheel base; \u03b8 is heading angle; R is turn radius. Left and right front wheel angle is not the same, we assume that: 1 2( ) / 2\u03b4 \u03b4 \u03b4= + (1) When the vehicle is steering, steering radius and vehicle front wheel angle have the following relationship: tan L R \u03b4 = (2) Feedforward and feedback tracking algorithm There exists certain deviations, including heading deviation \u03c8 (the deviation between the actual and the desired heading) and distance deviation d (the distance between the center of rear axle and the section of path), when the vehicle is tracking one segment of the path" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003520_detc2014-34759-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003520_detc2014-34759-Figure1-1.png", + "caption": "Fig. 1. Three components of a harmonic drive", + "texts": [ + " From the results, applying explicit dynamics analysis for the harmonic drive is doable and has several unique advantages than traditional TCA. Due to the harmonic drive\u2019s generality, it is possible that more researchers and technicians will adopt this method for gear industry in the future. HARMONIC DRIVE The harmonic drive\u2019s theory is based on elastic dynamics and utilizes the flexibility of metal. The mechanism has three basic components: a wave generator, a flex spline, and a circular spline (Fig. 1). The wave generator is made up of two separate parts: an elliptical disk called a wave generator plug and an outer ball bearing. The gear plug is inserted into the bearing, giving the bearing an elliptical shape as well. The flex spline is like a shallow cup. The sides of the spline are very thin, but the bottom is thick and rigid. This results in significant flexibility of the walls at the open end due to the thin wall, but in the closed side being quite rigid and able to be tightly secured (to a shaft, for example)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003068_cacs.2014.7097182-Figure7-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003068_cacs.2014.7097182-Figure7-1.png", + "caption": "Fig. 7. System trajectories: (a) start from the region I and (b) start from the region II.", + "texts": [ + " The convergence of these trajectories is illustrated in Fig. 6(b). In persuing this control objective, we define the region I and II by dividing the phase plane into its upper plane and lower plane as illustrated in Fig. 6(b). By considering the trajectory of the system where C = \u03b8T , it follows from (39)-(40) that \u03b8 can be written as \u03b8 = \u2212 I 2\u03c4 \u03b8\u0307|\u03b8\u0307|+ \u03b8T . (41) We introduce curve \u03c8 as the system trajectory that converges to \u03b8T as illustrated in Fig. 6(b). We observe the trajectories in those two regions with initial conditions shown in Fig. 7 and we can summarize the control strategies [10] as follows \u2022 Suppose the initial condition A is located in the region I, the control command is set to negative so that it begins to slow down and enters the region II. When the system trajectory intersects the curve \u03c8 at point B, the control command is switched to positive so that the system trajectory converges to \u03b8T (Fig. 7(a)). 978-1-4799-4584-9/14/$31.00 \u00a92014 IEEE \u2022 Likewise in previous case; if the initial condition A is located in the region II, the control command is set to positive so that it begins to speed up and enters the region I. When the trajectory intersects the curve \u03c8 at point B, the control command is set to negative and hence, the system trajectory converges to \u03b8T (Fig. 7(b)). Based on the graphical analysis shown in Fig. 7, it follows that the control command is given by u = \u23a7\u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23a8 \u23aa\u23aa\u23aa\u23aa\u23aa\u23aa\u23a9 +1, as { \u03b8 + I 2\u03c4 \u03b8\u0307|\u03b8\u0307| \u2212 \u03b8T < 0 \u03b8 + I 2\u03c4 \u03b8\u0307|\u03b8\u0307| \u2212 \u03b8T = 0, \u2200\u03b8\u0307 < 0 \u22121, as { \u03b8 + I 2\u03c4 \u03b8\u0307|\u03b8\u0307| \u2212 \u03b8T > 0 \u03b8 + I 2\u03c4 \u03b8\u0307|\u03b8\u0307| \u2212 \u03b8T = 0, \u2200\u03b8\u0307 > 0 (42) Note that (42) gives the conditions on the control command so that the system trajectories converge to \u03b8T . So far, we have derived the control for one dimensional case only by assuming that \u03b8 is a general rotation angle. To design the controller for the cold gas dynamic system, by recalling the dynamics model in (29)-(33), we see that if the displacements of body center in x, y and z axes are very small, then (31)-(33) can be neglected" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003671_detc2011-48794-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003671_detc2011-48794-Figure2-1.png", + "caption": "Figure 2. Contact surface and contact node.", + "texts": [ + " The penalty technique is used for imposing the constraints in which a normal reaction force (Fn) is generated when a node penetrates in a contact body whose magnitude is proportional to the penetration distance. In the present formulation, the force is given by [11-13]: \u23a9 \u23a8 \u23a7 < \u2265 += 0 0 d d dcs dcAdkAF pp p pn & & & & (13) iin nvd =& (14) where A is the area associated with the contact point, kp and cp are the penalty stiffness and damping coefficient per unit area; d is the closest distance between the node and the contact surface (Fig. 2); d& is the signed time rate of change of d; sp is a separation damping factor between 0 and 1 which determines the amount of sticking between the contact node and the contact surface at the node (leaving the body); nr is the normal to the surface and invr is the velocity vector in the direction of nr . The normal contact force vector is given by: niin FnF = (15) The total force on the node generated due to the frictional contact between the point and surface is given by Eq. (10). Using the EHL (elasto-hydrodynamic lubrication) theory [3, 6], we can express the shear-stress (\u03c4 ) between two elastic contact surfaces as: Downloaded From: http://proceedings" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001185_gt2015-42329-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001185_gt2015-42329-Figure2-1.png", + "caption": "FIGURE 2: Airflow in rig", + "texts": [ + " This is possible during operation to simulate rotor/stator movements due to temperature changes. The axial movement is possible with a speed of up to 175 mm/s and a step size of 5x10\u22125 mm. The radial movement works with a speed of up to 6 mm/s and a step size of 9x10\u22125 mm. This is more than an order of magnitude faster than in the real engine with about 6x10\u22122 mm/s radial movement. The casing consists out of two units. The stationary parts consist mainly of the bearing chamber (4), the bearings and a ring structure. This is shown in Figure 2. The rotating parts are shown in gray and consist of the shaft and the overhung rotor, which allows for quick changes of the rotor structure and the installed seal. It is also easily possible to optically examine the rotor surface to detect possible friction and wear. The traversable casing consists of three parts. The seal holder with the measured seal is installed in this casing and is therefore movable. The main parts of the instrumentation are installed in the casing to measure for example the front and back pressure or the rotor elongation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002903_s1068798x11100248-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002903_s1068798x11100248-Figure2-1.png", + "caption": "Fig. 2. Combinations of the deviations of sinusoidal pin and bearing contact surfaces, with phase shift.", + "texts": [ + "022 mm, the maxi mum diameter of the slip bearing after assembly and subsequent boring of the upper and lower bushes in the crankshaft bearing. We assume that the pin cross sections are circles, while the longitudinal generatrix is sinusoidal. We investigate the gaps between the bush and the shaft when the generatrices of the contact surfaces are, respectively, sinusoidal and hyperboloid or sinusoidal and linear, as well as other combinations. The combi nation of sinusoidal bush and hyperboloid pin surfaces is best in terms of eliminating oil leakage through the end gaps of the frictional pair. In Fig. 2, we show a combination in which the shaft and bush surfaces are solids of revolution with phase shifted sinusoidal generatrices. This combination also exists in many variants and, correspondingly, has not been much studied. Our results show that surfaces with sinusoidal generatrices correspond to a steadier oil volume in the gap. A combination with closed ends of the gap is best in terms of minimum end gap and reduced end pressure at the bearing (Figs. 1d and 3). Such a combination may be obtained by matching the sinusoidal genera trices at the ends\u2014that is, by appropriate selection of Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001433_pamm.201410021-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001433_pamm.201410021-Figure1-1.png", + "caption": "Fig. 1: Top view of the considered vehicle structure.", + "texts": [ + " KGaA, Weinheim 1 Introduction Vehicle safety is an important research topic, especially with focus on heavy commercial vehicles. According to study [1], accidents with such vehicles cause more seriously injured persons and fatalities and a much higher average physical damage, than car accidents. Therefore, improving their vehicle safety has a great potential in improving the overall traffic safety. In this article the authors outline the necessary steps for the design of a nonlinear vehicle dynamics controller using electronic steering axles of truck semitrailer combinations, cf. Fig. 1. The controller will not act at the front axle steered by the driver, but the last trailer axle. These electronic steering axles are available in some trailers, to improve their maneuverability. It is shown in [2] using linear control methods, that steering all trailer axles can improve tracking performance as well as vehicle stability for several vehicle combinations. The present paper presents how nonlinear control design techniques can be used, when just one of the trailer axles is steered. The used control structure is shown in Fig", + " reference generator second order sliding mode control exact input-output linearization vehicle \u03b4f x\u0303 vEA \u03b4t \u03b4f x Fig. 2: Block diagramm of the controller to be designed 2 Modeling In this section the focus lays on the model used for the control design. For the sake of brevity the kinematic model used for the reference generator and the three dimensional simulation model built in Simpack are omitted. The modeling follows the methods and notations described in [3]. The truck semitrailer shown in Fig. 1 consists of two bodies building a serial chain with two joints. The first joint between the earth and the tractor has three degrees of freedom (two translational, one rotational) and the second joint between tractor and semitrailer one rotational degree of freedom. These four degrees of freedom are described by the vector of joint coordinates \u03b2 and joint velocities \u03b7. \u03b2 = (x1, y1, \u03c81, \u03c812) T and \u03b7 = ( 1vx, 1vy, \u03c8\u03071, \u03c8\u030712 )T (1) Using the systematic multi-body approach from [3] leads to the equations of motion of the vehicle" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002117_2013-01-2011-Figure15-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002117_2013-01-2011-Figure15-1.png", + "caption": "Figure 15. High Stress location at 5second time slice of Automatic Side Loading operation on Back Panel", + "texts": [ + " Since the cab or the components mounted in the cab do not have any mode below 2Hz, it is safe to assume that modal amplification of the cab is not likely for the Automatic Side Loading operation. Due to the tunnel Matchboxing caused by the roll mode of the cab, the corner of the tunnel lights up on the back panel of the cab. The stresses obtained from the Inertia Relief analysis are at a single time slice of the Automatic Side Loading operation. nCode DesignLife was used to create a Duty cycle file for the loading and the repeats were provided to calculate the life of the cab. Figure 15 shows the high stress location at the corner of the tunnel at the 5 second time slice during the Automatic Side Loading operation. In order to make sure that the forces obtained from CVM-D are correct, the accelerations measured on the cab were correlated between analyses and testing. Time series data, PSD plots and Level crossings were used to compare the analysis results with physical testing. Since strains were not measured during physical testing, they could not be correlated. Due to the cab being in roll mode induced by the frame dynamics, the lateral accelerations on the cab back panel top have been compared between analysis and physical testing in this paper" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001077_978-1-4419-7979-7_6-Figure6.28-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001077_978-1-4419-7979-7_6-Figure6.28-1.png", + "caption": "Fig. 6.28 Pointwise construction of the exciting current if\u00f0t\u00de of a magnetic circuit", + "texts": [], + "surrounding_texts": [ + "Consider the increasing branches of B(t) and [B(t) versus if(t)] characteristic (Fig. 6.27): At t\u00bc t1\u00bc 0 the flux density has the average value of B(t)\u00bc 0 T resulting in if\u00f0t1\u00de \u00bc 1:0A; At t\u00bc t2 the flux density has a value of B(t)\u00bc 1.0 T resulting in if\u00f0t2\u00de \u00bc 3:5A, At t\u00bc t3 the flux density has a value of B(t)\u00bc 0.5 T resulting in if\u00f0t3\u00de \u00bc 0A; At t\u00bc t4 the flux density has an average value of B(t)\u00bc 0 T resulting in if\u00f0t4\u00de \u00bc 1A; At t\u00bc t5 the flux density has a value of B(t)\u00bc 1.0 T resulting in if\u00f0t5\u00de \u00bc 3:5A; 250 6 Magnetic Circuits: Inductors and Permanent Magnets At t\u00bc t6 the flux density has a value of B(t)\u00bc 0.5 T resulting in if\u00f0t6\u00de \u00bc 0A; At t\u00bc t7\u00bc t1\u00bc 0 the flux density has an average value of B(t)\u00bc 0 T resulting in if\u00f0t1\u00de \u00bc 1:0A: Apparent Power Requirements for AC Excitation of Cores: The AC excitation characteristics of core materials are usually expressed in terms of apparent power (volt-amperes) Sc rather than a magnetization curve relating B and H, see Fig. 6.29a. The apparent power Sc required to excite the core to a specified flux density is Sc \u00bc ErmsIfrms \u00bc 4:44fNAcBcmax \u2018cHcrms N : (6.122) For a magnetic material of density gc\u00bc 7.86 (kg-force)/dm3\u00bc 7,860 (kg-force)/m3 the weight of the core is weightcore \u00bc Ac\u2018cgc (6.123) if(A) B(T) 6.3 Magnetic Properties of Materials 251 252 6 Magnetic Circuits: Inductors and Permanent Magnets and the apparent power per unit weight is (see Fig. 6.29a) sc \u00bc Sc weightcore \u00bc 4:44f gc BcmaxHcrms: (6.124) The excitation apparent power measured in volt-amperes at a given frequency f (e.g., f\u00bc 60 Hz) is dependent only on Bcmax within the core because Hcmax is a unique function of Bcmax and is independent of turns and geometry. As a result the AC excitation requirements for a magnetic material are often given in terms of apparent power sc expressed in volt-amperes per unit weight, as is illustrated in Fig. 6.29a for M45 Fully Processed 26 Gauge electrical steel. This figure also shows the (real) power loss per unit weight pc within the iron core, and the magnetizing force Hcmax required. Figure 6.29b depicts the DC single-valued B\u2013H characteristic for the same electrical steel [8, 9]. Core Losses due to AC Excitation of Cores: The apparent power Sc required by cores can be separated into real power (Pc) and reactive power (Qc) at a certain frequency and a given Bcmax: S \u00bc Sc \u00bc Pc \u00fe jQc: (6.125) The reactive power Qc is cyclically supplied and absorbed by the excitation source and contributes to the excitation current iF delivered by the source. In Chap. 7 we will call the reactive current causing the reactive power the \u201cmagnetization\u201d current im. The real power Pc absorbed by the core is due to two loss mechanisms which are associated with time-varying fluxes in magnetic materials. l I2R heating associated with eddy currents in cores. l Second loss mechanism is due to the hysteretic nature of magnetic materials. To reduce eddy current losses the iron cores must be laminated and consist of electrical steel sheets as indicated in Fig. 6.30b. Due to the core flux f t\u00f0 \u00de \u00bc fmaxsinot; (6.126) and Faraday\u2019s law in integral form e(t\u00de \u00bc \u00de C ~E(t) d~l \u00bc Nd( \u00d0 S ~B(t) d~S\u00de/dt \u00bc N fmaxocosot the induced voltage becomes 6.3 Magnetic Properties of Materials 253 where ~E(t)is the electric-field strength within/outside the core due to the changing flux f(t) as depicted in Figs. 6.31a, b. Note e(t) is the induced voltage with its amplitude Emax and its rms value Erms. 254 6 Magnetic Circuits: Inductors and Permanent Magnets currents flow within the cross-section Asolid, and within one lamination (Fig. 6.31d) the eddy currents flow within the cross-section Alaminated. This means the ratio Rlaminated/Rsolid 1,000. If the induced voltage in the solid core is Erms solid\u00bc 1 V then the induced voltage in one lamination is Erms laminated Erms solid. As a result the loss within the core laminations is much less than that of the solid core because Rlaminated 1,000 Rsolid." + ] + }, + { + "image_filename": "designv11_84_0002897_amm.86.301-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002897_amm.86.301-Figure4-1.png", + "caption": "Fig. 4 The sample piece material of self-lubricating after oil loss test", + "texts": [ + " The sliding friction shim was improved to composite structure that composed of three layers, two pieces of self-lubricating material beside and steel material in the middle. According to no oil test compared, the old material tin bronze sample piece whose temperature rise quickly and wear intensified. So the test was forced to stop (the sample piece condition after 300min test see Fig. 3). On the other hand, the material self-lubricating sample piece pass the no oil test for 400min, and the sample piece condition after test is better, see Fig. 4. The conclusion of the test is that using self-lubricating material can clear raise the friction parts no oil capability. After improvement the weakness of MGB and the material of the sliding friction shim change to the self-lubricating one, we take a test to validate the MGB oil loss capability. After the oil losing during the test, the time from oil press alarming to end of the test was 34min 44sec (the time for oil loss test availability). The MGB turned limberly when it was inspected by hand-turning after test" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002509_carpathiancc.2011.5945811-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002509_carpathiancc.2011.5945811-Figure1-1.png", + "caption": "Figure 1 A four-degree-of-freedom model", + "texts": [ + " Torsional vibration models of gears system are also classified according to time-invariant and time-variant such as linear time invariant (LTI) models with stiffness, linear time-varying (LTV) models with stiffness, time-varying models with backlash, as well as time-invariant average stiffness and time-invariant models in both backlash and stiffness simultaneously [13]. A. Equation of Motion of Gears System A four-degree-of-freedom model of the pinion gear\u2013 wheel gear system is considered for simplicity. The pinion gear body and wheel gear body are assumed to be rigid. The teeth are assumed to be elastic and parallel spring\u2013damper combinations are assumed to exist between the teeth and the gear body. A four-degree-of-freedom model is shown in Fig. 1. The equations of motion of the pinion gear\u2013wheel gear system are written in terms of the four-degree-of- freedom model as follows [7\u201310, 15]: pS , p\u03c6 , wS and 37978-1-61284-361-2/11/$26.00 c\u00a92011 IEEE w\u03c6 Equation (1) is written for tooth i of a pinion gear as follows: ( ) ( ) pppdpeppdpeppth TSrKSrDSJ +\u2212+\u2212= \u03c6\u03c6 22 (1) where pthJ is the moment of inertia of tooth i of the pinion gear [kg.mm2], dpr is the pitch circle radius of the pinion gear [mm], p\u03c6 is the rotational position of the pinion gear body [rad], pS is the rotational position of tooth i of the pinion gear [rad], p\u03c6 is the rotational velocity of the pinion gear body [rad/s], pS is the rotational velocity of the tooth i of the pinion gear [rad/s], pS is the rotational acceleration of the tooth i of the pinion gear [rad/s2], and pT is the contact torque applied to tooth i [N" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002235_humanoids.2011.6100821-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002235_humanoids.2011.6100821-Figure9-1.png", + "caption": "Fig. 9. Reference posture", + "texts": [ + " For each reference posture, we assigned finger links contacting the object and the reference posture for small, mid and large sized objects for each reference grasping motion. We manually generate the reference grasping posture for three fingered fingertip, four-fingered fingertip, and four-fingered enveloping. Actual grasping posture is generated by modifying the reference posture as detailed in the next section. For each reference posture, we define the grasping rect angular convex (GRC) to select the feasible grasp type for the given object. The GRC is defined in the hand coordinate system as shown in Fig. 9. The superscript Pa is defined as a vector a in the hand coordinate system. The GRCmax is the maximum size of the object without interfering with the finger links and GRCmin and GRCdes is the minimum, desired sizes of the object. For the i-th grasp type with the GRCmax has posi tion/orientation PPmax,i/P Rmax,i (i = 1\"\", n), and edge length vectors of GRCmax, GRCdes, the GRCmin defined as emax,i, edes,i, and emin,i' The elements of emax,i are sizes of GRCmax and shown in Fig. 9 as (emax,i)x, (emax,i)y, and (emax,i)z. Pdi is defined as the vector having the direction of approach to the object and is an outer unit normal vector of a GRC surface. We define maximum and minimum mass, mmax,i and mmin,i of the grasped object using the i-th grasp type. Given the object shape to be grasped, our planner calculates the object convex polygon (OCP) including the grasped object in object coordinate system. In this paper, we consider the rectangular box as the OCP. For complex object shape, we split the object into several regions and calculate the OCP for each region" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.1-1.png", + "caption": "Fig. 7.1 A general parallel robot (the gray pairs denote the actuated joints)", + "texts": [ + " These robots have been chosen because they are typical examples found in the literature and/or their dynamic model will be defined later in the book. Let us consider a general PKM composed of a rigid fixed base (denoted as the elementB0 onwhich is attached the global frameF0(O, x0, y0, z0)), a rigidmoving platform (elementBp) and n legs. Each leg is a kinematic chain (which is serial most of the time, but can also be composed of closed-loop or tree-structure sub-chains1) composed of bodies connected by mi joints located at points Aij (revolute, prismatic, universal, etc.\u2014i = 1, . . . , n, j = 1, . . . , mi ) (Fig. 7.1\u2014in this figure, the gray pairs denote the actuated joints). The j th link of the leg i will be denoted in what follows as the linkBij. Moreover, the joint located at point Aij will be parameterized by the variable qij. 1At the end of the Chap.7 on the kinematics of PKM, we will present the geometry and kinematic equations by considering PKMmade of serial legs only. However, the methodology can be extended to any types of legs. Moreover, the equations of the dynamics presented in Chap.8 are general and can be used for any types of legs made of serial, closed-loop or tree-structure chains", + " An additional problem is to obtain the passive joint coordinates qd (which are needed for the computation of the dynamic model) as a function of the active joint variables qa , i.e. qd = Gd(qa). The main idea that is usually followed is to adequately rearrange the Eq. (7.4) in order to suppress the translation parameters of the vector x so that a polynomial depending on the tangent, sine and/or cosine of the rotation parameters of the vector x can be obtained. To simplify the calculations, it is generally necessary to set the base frame origin O at one robot base anchor point (e.g. point A11 in Fig. 7.1) and the moving platform frame origin P at one robot platform anchor point (e.g. point A1m1 in Fig. 7.1). Note that this choice can also be taken while solving the IGM. In order to simplify and/or clarify understanding of the problem, the following geometrical approach could be also used. The idea is to virtually disconnect a leg from the robot moving platform, e.g. without loss of generality at the joint located at point A1m1 (Fig. 7.12a). This joint will be denoted as joint A1m1 . In that case, even when all the actuators are fixed, the moving platform gains one or more DOF and the joint A1m1 can freely describe a configuration loci (translations plus rotations) denoted asS ", + " This problem is still an open problem for research on PKM, even if some methodologies have already been proposed. To get the assembly mode knowledge, it is possible to use additional encoders mounted in the passive joints (Arai et al. 1990; Inoue et al. 1985). Such additional information can help to find the real posture of the robot, and can also help to simplify the computation of the pose. For example, if all active and passive joint coordinates of the leg k of the general robot presented in Fig. 7.1 are measured, the problem remains to find the direct geometric model of a serial structure, which has a direct and unique solution (Khalil and Dombre 2002). Another solution is to use exteroceptive sensors such as cameras. The most common approach consists of the direct observation of the end-effector pose (Espiau et al. 1992; Horaud et al. 1998; Martinet et al. 1996). However, some applications prevent visual observation of the end-effector of a parallel mechanism. For instance, it is not wise to imagine observing the end-effector of a machine-tool while it is generally not a problem to observe its legs that are most often designed with slim and rectilinear rods (Merlet 2014)", + " In this section, the kinematic relations linking the active joint velocities to the platform twist and passive joint velocities are defined and analyzed. The kinematic relations linking the active joint velocities to the platform twist could be obtained by differentiating (7.4) w.r.t. time. However, this solution may not be computationally efficient. Therefore, we propose to use the following methodology which can take advantage of the recursive algorithms defined in Sect. 5.2.4. Let us consider the input-output relation of the chain i (Fig. 7.1) which expresses the platform twist (which will be denoted below as 0tp, the superscript \u201c0\u201d denoting that the vector is given in the reference frameF0) as a function of all joint velocities q\u0307i for the considered chain. From (5.6) and (5.7), we have 0tp = 0Jpi q\u0307i = [ 0$i1 . . . 0$i mi ] q\u0307i . (7.53) where 0$ik is a unit twist representing the displacement of the end-effector when joint ik is moving only and mi is the number of joints in the considered chain.2 Let us rewrite (7.53) by reorganizing matrix 0Jpi so that we can group: \u2022 in a sub-matrix 0$ia the unit twists corresponding to the active joints of velocities q\u0307ai, and \u2022 in a sub-matrix 0$id the unit twists corresponding to the passive joints of velocities q\u0307di", + " All joint velocities could be obtained by differentiating (7.3) w.r.t. time. However, this solution may not be computationally efficient. Therefore, we propose to use the following methodology which is based on equating the twist at the terminal frame of each chain as a function of the platform twist 0tp from one side, and as a function of the joint velocities of the chain from the other side. This method can take advantage of the recursive algorithms defined in Sect. 5.2.4. Let us consider the chain i of the PKM, which is composed of mi joints (Fig. 7.1). From (3.2), we can compute the twist of the platform at point Ai mi (that will be denoted as ti p) as: 0ti p = 0tp + [0\u03c9p \u00d7 0rP Ai mi 0 ] (7.81) which can also be rewritten in the matrix form: 0ti p = [ 13 \u22120r\u0302P Ai mi 0 13 ] 0tp = Jti 0tp (7.82) where Jti is a (6 \u00d7 6) matrix. As the joint located at Ai mi also belongs to the chain i , its twist can be obtained by using the relation (5.8) as: 0ti p = 0Ji mi q\u0307i = [ 0$i1 i mi . . . 0$i mi i mi ] q\u0307i (7.83) where q\u0307i represents the vector of all joint velocities of the chain i , 0Ji mi = 0Ri mi i mi Ji mi is the chain i kinematic Jacobian matrix of dimension (6 \u00d7 mi ) and $ik i mi is a unit twist representing the displacement of the chain tip Ai mi when joint ik is moving only", + "172) Computation of the passive joint accelerations is necessary for the computation of the dynamic model. Therefore, the way to compute them is defined in this section. Once again, all joint accelerations could be obtained by differentiating (7.3) w.r.t. time twice. However, this solution may not be computationally efficient. Therefore, we propose to use the following methodology which can take advantage of the recursive algorithms defined in Sect. 5.3. Let us consider the chain i of the PKM, which is composed of mi joints (Fig. 7.1). From (3.20), we can compute the acceleration of the platform at point Ai mi (that will be denoted as t\u0307i p) as: 0 t\u0307i p = [ 13 \u22120r\u0302P Ai mi 03 13 ] 0 t\u0307p + [0\u03c9p \u00d7 (0\u03c9p \u00d7 0rP Ai mi ) 0 ] = Jti 0 t\u0307p + di . (7.173) From (5.30), and as the joint located at Ai mi also belongs to the chain i , we also have 0 t\u0307i p = 0Ji mi q\u0308i + 0bi mi (7.174) where q\u0308i represents the vector of all joint accelerations of the chain i , 0Ji mi = 0Ri mi i mi Ji mi is the chain i Jacobian matrix also found in (7.83), and 0bi mi = 0Ri mi i mi bi mi , where i mi bi mi can be obtained by the recursive algorithm (5" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.10-1.png", + "caption": "Fig. 2.10. Rotation of an angle about an axis", + "texts": [ + " Let r = [ rx ry rz ]T be the unit vector of a rotation axis with respect to the reference frame O\u2013xyz. In order to derive the rotation matrix R(\u03d1, r) expressing the rotation of an angle \u03d1 about axis r, it is convenient to compose 5 The ordered sequence of rotations XYZ about axes of the fixed frame is equivalent to the sequence ZYX about axes of the current frame. elementary rotations about the coordinate axes of the reference frame. The angle is taken to be positive if the rotation is made counter-clockwise about axis r. As shown in Fig. 2.10, a possible solution is to rotate first r by the angles necessary to align it with axis z, then to rotate by \u03d1 about z and finally to rotate by the angles necessary to align the unit vector with the initial direction. In detail, the sequence of rotations, to be made always with respect to axes of fixed frame, is the following: \u2022 Align r with z, which is obtained as the sequence of a rotation by \u2212\u03b1 about z and a rotation by \u2212\u03b2 about y. \u2022 Rotate by \u03d1 about z. \u2022 Realign with the initial direction of r, which is obtained as the sequence of a rotation by \u03b2 about y and a rotation by \u03b1 about z" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001126_2015-01-2515-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001126_2015-01-2515-Figure2-1.png", + "caption": "Figure 2. A labeled cutaway picture of an injector.", + "texts": [ + " These tube ends slow the wear caused by violent bounce back, and are easy to swap out when they get worn down. After being decelerated, the fastener settles into the guide chute. For headed rivets and bolts, the guide chute geometry indexes the fastener vertically using the head feature. After pausing to ensure that the fastener is fully seated in the guide chute, the pusher is actuated forward, pushing the fastener through the guide chute into the finger assembly, opening the chute in the process. The guide blocks, labeled #5 in Figure 2, are 3D printed. This allows for quick prototyping and iteration, as well as the ability to create complicated geometries easily. They attach with one socket head cap screw each to the guide block holders (4), which are in turn attached to the outputs of the parallel gripper (3). When the pusher (2) is extended, the stopper (1) is lined up at the termination of the feed tube. After the fastener arrives and the pusher retracts, the guide blocks form a receiving hole for the incoming fastener. Then, as the pusher comes forward, the fastener is forced down the guide chute into the finger assembly (6)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001907_00022661111159861-Figure11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001907_00022661111159861-Figure11-1.png", + "caption": "Figure 11 (a) The schematic of the airship geometry and (b) grid for numerical simulation", + "texts": [ + " The numerical simulation of the fluid flow around the 3D airship model (in order of size, 1:13) was carried out in ANSYS FLUENTw 6.3 commercial software for the case with orientation angles of b \u00bc 0 and a \u00bc 0 with reference to upstream flow direction corresponding to Reynolds number of 104 and 106. Since, this was found to be optimum stable position of airship model with respect to upstream flow direction. The numerical simulations were performed to instil confidence in its future usability to inculcate design changes in airship model (McCormick, 1995; Chelli and Dala, 2001) (Figure 11). The flow around the airship is turbulent. We have considered Reynolds averaged Navier-Stokes equations with the Boussinesq hypothesis to model the Reynolds stresses (Launder and Spalding, 1974). The closure problem of the turbulent modeling is solved using k-vmodel with appropriate wall functions. The system of equations for solving 3D, incompressible fluid flow in steady-state regime is as follows. Continuity equation: \u203ay \u203axi Ui \u00bc 0 \u00f07\u00de Where, i \u00bc 1, 2, 3. Navier-Stokes (momentum) equations: \u203a \u203axj rUiUj \u00bc2 \u203a p \u203axi \u00fe \u203a \u203axj \u00f0m\u00femt\u00de \u203a \u203axj Ui \u00fe \u203a \u203axi U j 2 2 3 dijrk \u00f08\u00de Where, Ui\u00f0t\u00de ; Ui \u00fe ui is the component of instantaneous velocity in i-direction (m/s), Ui is the component of time averagedmean velocity in i-direction (m/s), ui is the component of fluctuating velocity in i-direction (m/s), i, j are the direction vectors, r is average fluid density (kg/m3),m is dynamic viscosity of fluid (kg/ms), mt is turbulent viscosity of fluid (kg/ms), p is average pressure, k \u00bc \u00f01=2\u00de ui uj the turbulent kinetic energy per unit mass(m2/s2); dij the Kronecker delta with the condition that, dij \u00bc 1 if i \u00bc j and dij \u00bc 0 if i \u2013 j", + " At the outlet, relative pressure of the fluid is null (outflow). The upper and lower, front and back domain limits are provided with free stream condition. Airship wall has no-slip condition, i.e. Ui,j,k \u00bc 0m/s. The domain is filled with incompressible air with kinematic viscosity n \u00bc 1.789 \u00a3 1025m2 s21. The simulations are based on finite volume method of discretization (Patankar, 1980). In order to limit numerical dissipation, particularly when the geometry is complex inducing an unstructured grid as shown in Figure 11(a) and (b), secondorder discretization schemes are used. Semi-implicit method for pressure linked equations algorithm is used for the solution process. The discretized nonlinear algebraic equations are solved by tri-diagonal method algorithm iterative method. The aerodynamic forces are categorized into various terms based on different physical effects: the added-mass force, the viscous effect on the hull, the force on the fins, the force on the hull due to the fins, and the axial drag.Computationalmethods are provided for each aerodynamic term and incorporated into the nonlinear dynamics model" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003308_12.894412-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003308_12.894412-Figure4-1.png", + "caption": "Figure 4: (a) Chemical structure of [P1,4,4,4][Tos]. (b) Layout of the OECT, indicating the area where the RTIL was confined. (c) Photo- graph of the OECT with a drop of glucose solution added. The balls at the pads are made of silver paste[119]. Reproduced by permission of The Royal Society of Chemistry. (d) Incorporation of the OECT into a flexible material (plaster).", + "texts": [ + " When the solution containing the analyte is added to the device, it mixes with the RTIL. The analyte, the enzyme, and the mediator are then allowed to interact and the OECT transduces this interaction. An important requirement for the RTIL is that it wets the PEDOT:PSS film, thus allowing the enzyme and the mediator to be patterned over the active area of the device. Moreover, the RTIL should be miscible with the aqueous solution that carries the analyte (PBS). The RTIL triisobutyl-(methyl)-phosphonium tosylate ([P1,4,4,4][Tos], Fig. 4a, supplied by Cytec Industries) satisfies these requirements, as the Tos anion gives it a rather hydrophilic character. Previous studies have also shown [P1,4,4,4][Tos] to be a biocompatible medium for glucose consumption by bacteria[120]. The layout of the device is shown in Fig. 4b. Two parallel stripes of PEDOT : PSS, with widths of 100 mm and 1 mm, respectively, were patterned on a glass support using photolithography. Contact pads at the end of the stripes allowed facile electrical connection to the source-measure units. The wide stripe was used as the transistor\u2019s channel and the narrow one as the gate electrode, as it has been shown that for enzymatic sensing the area of the channel must be larger than that of the gate electrode[121]. A monolayer of (tridecafluoro-1,1,2,2-tetrahydrooctyl) trichlorosilane (FOTS) was patterned on the surface of the device leaving uncovered only a small area of the channel and of the gate electrode", + " These areas of PEDOT:PSS which were left uncovered by FOTS served as hydrophilic \u2018\u2018virtual wells\u2019\u2019[122] and were shown to be effective in confining the RTIL (and the glucose solution, when it was added) over the centre of the device. The experiments involved placing a small amount of [P1,4,4,4][Tos] that included the enzyme glucose oxidase and the mediator ferrocene [bis(n5-cyclopentandienyl)iron] on the centre of the device and allowing it to be accommodated in the hydrophilic virtual wells. Subsequently, 50 \u03bcl of glucose solution in PBS were added to the device and allowed to mix with the RTIL solution, as seen in Fig. 4c. Fig 4d shows the incorporation of the OECT / IL electrolyte mixture into a common plaster, illustrating the versatility of the material. Proc. of SPIE Vol. 8118 81180U-9 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 05/13/2015 Terms of Use: http://spiedl.org/terms Fig. 5a shows the transient response of the drain current of an OECT for different concentrations of glucose solution, upon the application of a 0.4 V pulse at the gate electrode with a duration of 3 minutes. The drain voltage was 0" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001298_9783527690312.ch12-Figure12.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001298_9783527690312.ch12-Figure12.2-1.png", + "caption": "Figure 12.2 Schematic (a) and photograph (b) of the dual-chamber MFC.", + "texts": [ + "1 Dual-Chamber and Single-Chamber MFCs Dual-chamber MFC is mainly composed of the anode chamber and the cathode chamber, which are separated by a proton exchange membrane [15, 16]. The proton exchange membrane can realize oxygen and bacteria isolation and prevent oxygen seeping into the anode chamber and affecting the bacteria. Meanwhile, protons can cross the proton membrane to the cathode chamber, keeping the charge and acidity balance. Bacteria, the culture medium, and the electrode materials have a great influence on efficiency of electricity production.TheMFC shown in Figure 12.2 has been designed by our group [17] and is constructed by using two round poly(methylmethacrylate) templates.The anodic or cathodic compartment is an inner cylinder in each template.The two compartments are screwed together and separated by two poly(methyl methacrylate) disk gaskets with a Nafion 112 membrane. Because of the convenient construction and simple operation, dual-chamber MFC is one of the most common configurations under current laboratory level, but the presence of the proton exchange membrane causes high internal resistance, so the power output density is low, thereby reducing its potential application" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002955_amm.633-634.1111-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002955_amm.633-634.1111-Figure1-1.png", + "caption": "Figure 1", + "texts": [ + "199, Purdue University Libraries, West Lafayette, USA-06/06/15,10:27:31) The aim of this article is to extend the application field of rheological models for research and prediction of the development of deformation and destruction of materials under different loading conditions. When the concrete is cyclically compressed from zero to a constant stress [5, 6], the residual deformation per one cycle is gradually reduced. In this case, the stress-strain dependence approaches to linear. The known rheological models cannot be used to explain this feature of the concrete. But it is possible to explain using the new rheological element. The particular case of the proposed rheological element is shown on the Fig. 1 with the following keys: 1 is the rigid matrix; 2 is the elastic puncheon; I, II, III, IV, V are the parts of matrix; ,III \u03b1\u03b1 = ,0III =\u03b1 ,IV\u03b1 V\u03b1 are the angles between the axis and the generating line of matrix on the corresponding parts of matrix; P is the load; R is the reaction force; fRT =L is the friction force when loading, where f is the friction coefficient; UT is the friction force when unloading; \u03b5 is the relative deformation. The projection of forces on the direction of P gives the following expression when loading: \u03b1\u03b1 cos2sin2L fRRP += (1) and the following expression when unloading: ", + " 2 illustrates the \u03c3-\u03b5 relation at the same time, where \u03c3 is the stress and \u03b5 is the relative deformation. If ,tg\u03b1>f the relative deformation does not change during unloading (the curve 5 on the Fig. 2). It means that the element is transformed into a rigid connection. During the repeated loadings the relative deformation begins to increase only when P is higher than maxP of the previous loading. The crack initiation force \u03b1 \u03b1 tg tg max\u0441 + \u2212 = f f PP (6) is necessary for pulling the puncheon 2 (look at the Fig. 1). This force is lower than maxP and opposite in direction. The deformation begins to reduce after the achievement of the force \u0441P . The coefficient of elasticity during unloading is lower than during loading (look at the curves 4 and 6 on the Fig. 2). The new rheological element allows combining the deformation and the strength models of the body. It is possible in the case of assuming that the displacement of the elastic puncheon from the initial position to the open butt of the matrix (i.e. sc\u03b5 on the Fig. 1) corresponds to the start of the main crack. Then further displacement of the puncheon ( )sc\u03b5\u03b5 > will correspond to the development of the main crack, and the condition ( )crsc \u03b5\u03b5 + will correspond to the finish of the destruction process, i.e. the fragmentation of the body. The ultimate stress 0\u03c3 can be determined using nondestructive methods, for example [7]. The durability of material is infinite without the achievement of this stress. The deformation of the rheological element before the start of the crack is determined by correlation: , e 0 sc E \u03c3 \u03b5 = (7) where eE is the coefficient of elasticity of the element" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003092_s106879981101017x-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003092_s106879981101017x-Figure1-1.png", + "caption": "Fig. 1.", + "texts": [ + " In [3], the problem of finding the coordinate of the cross-sectional area variation by the known dependence of the free rod end movement on the disturbing force frequency was solved. The paper [4] addresses the inverse problems on longitudinal traveling waves in a rod of finite length. We consider the natural torsional vibrations of a nonrotating turbocompressor shaft. The shaft has an elastic support at the end with the torsional rigidity 1.c The shaft is connected to the compressor disk with the moment of inertia 1J and the turbine disk with the moment of inertia 2J as shown in Fig. 1. It is assumed that the shaft has a short part (in comparison with the total length of the shaft) with a smaller cross-sectional area. This notch simulates its damage, such as an open crack. We consider the stress-strain state of the shaft within the elasticity limits. Because the crack is a result of a small nucleus propagation, and it is not necessarily appears in the most stressed section, we assume that the notch can be placed anywhere along the shaft length. The task is to determine the coordinates of the notch and its sizes under the plane-sections hypothesis" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002323_amm.440.329-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002323_amm.440.329-Figure1-1.png", + "caption": "Figure 1 is the structure diagram of rugby Scrum trainers as described in this article.", + "texts": [ + " This article analyze the Scrum force characteristics, combing with the movement biology mechanics, mechanical design and manufacturing, computer technology, human body engineering and other professional technology and the close cooperation with the coaches, successfully developed the rugby Scrum special strength training of test equipment. It can realize the real-time observation of the athletes doing Scrum technical action of various forces and achieve real-time monitoring, diagnosis, and correction of Scrum training action, in order to achieve Scrum techniques of quantitative, refined training and to improve the effect of football Scrum As it is shown in figure 1, the football Scrum trainers includes thrust force measuring device 3 which is for measuring the shoulder force signal in the pass practice; the constant push force measuring device 2 which is used to measure and deliver the pedal force signals of the lower limb in training; Control analysis 7 which is used to receive signals from the thrust force measuring device 3 and the constant push force measuring device 2 and to control both of them. The thrust force measuring device 3 includes the side plate fixed to the trainer which is equipped with a head hole placement 1 and a three-directional sensor to measure the force 5" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002868_s12206-015-0143-9-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002868_s12206-015-0143-9-Figure2-1.png", + "caption": "Fig. 2. Configuration of the quad tilt prop PAV.", + "texts": [], + "surrounding_texts": [ + "Since there is no cyclic control in the currently designed PAV, the term \u2018tilt-rotor\u2019 is inappropriate for use, and due to absence of wings, it is also inappropriate to use \u2018tilt-wing\u2019. The uses of four short propellers with high degree of twist lead to the name of quad tilt-prop PAV. If the wings and aerodynamic control surfaces are available, it is then possible to transit to the airplane mode; i.e., tilt angle \u03b3 = 90\u00b0 [11]. However, Fig. 4 shows the flight data that the inboard wing caused a serious uncontrollable state during the VTOL mode flight. It was barely controllable, but the ground effect under the wing caused an uncontrollable state near the ground. The inboard wing had to be modified to a circular support in the initial prototype due to its strong aerodynamic interference that is also prevalent in tilt-rotor [11]. Therefore, unlike a tilt-wing concept, the current prototype has 3 modes: i.e., the \u2018driving\u2019, \u2018VTOL\u2019, and \u2018forward flight\u2019 modes as il- lustrated in Fig. 3. During the forward flight mode, the nacelle tilt angle is limited under \u00b120\u00b0, where the dynamics is similar to a pure quad-rotor, and no additional aerodynamic control surfaces are needed. Fig. 5 shows the frame structure and applied mechanisms. It is mainly constructed with two longitudinal carbon pipes and four aluminum bulkhead frames for a light structure. Two torque tubes are installed on the longitudinal pipes by aluminum housings, where the front torque tube is installed with two nacelles. Each front nacelle is equipped with a propulsion motor, two wheels, a steering mechanism and a steering servo actuator. The rear torque tube is also installed with two nacelles, where each nacelle is equipped with a propulsion motor, two wheels, a driving motor with a reduction gear and a brake drum. The mechanism of tilt actuator is also included in the Fig. 5, where there are total of 8 high-torque servo actuator coupled in a pair that control tilt angle. A stopper pin is also added to release the power of servo actuators during the driving mode. As it was also shown in the Fig. 3 for the orientation of tiltable nacelles during each mode, during the driving mode, nacelles are fixed to 45\u00b0 facing each other to lower the overall height, and locked by the stopper pin to save power consumption. When the VTOL mode is initiated, two pairs of servo actuators deliver control forces to each torque tube without any gear to simplify the overall structure [8]. So far, controls are achieved by manual control during the driving mode, and the manual operator controls the steering angle of front wheels, driving motor and brake drums of rear wheels. During the VTOL mode, nacelles are fixed straight-up, and a pure quad rotor controls are applied. In the forward flight mode, forward velocity is controlled by acceleration created from tilted nacelles, where all nacelles are tilted with the same angle \u03b3. When these two modes are compared, the forward flight mode will minimize the discomfort of flight experience than the VTOL mode that has to tilt its whole body pitch angle to control forward velocity. Small body pitch angle may improve the discomfort during the acceleration and deceleration, however, if the passenger is not sitting at the exact position as the center of gravity, any pitching moment will result to a positive or negative gravity offset, and the passenger will have the feeling that he/she is constantly in a turbulent wind. Therefore, the VTOL mode is only used during initial take-off and at the end of landing, and the tilt angle \u03b3 will constantly be changing from the forward velocity feedback loop whenever it is off the ground. Fig. 6 shows onboard avionics in the PAV. A custom-made flight control computer was built to realize autonomous control using a 150 MHz 32-bit floating-point TMS320F28335 processor from Texas Instrument, Inc. A commercial GPSINS from Microbotics, Inc. was adopted to provide attitude, velocity and position data at 50 Hz rate. A commercial Bluetooth modem was used for wireless data communication that has maximum operating range of 300 ~ 1000 m depending on the line-of-sight condition. It is also equipped with a custommade data recorder that records all flight data in a micro-SD card." + ] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure6.16-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure6.16-1.png", + "caption": "Fig. 6.16 Formal pre-visualized piece", + "texts": [ + " 6 Numerical Simulation of Cranial Distractor Components \u2026 171 any purpose with computational methods, the generated iterations can be applied in other traditional methods and without the use of the numerical equipment. 8. Define the possible materials to be used in manufacturing the object, more than one can be selected, see Fig. 6.14. 9. The program performs a pre-verification where we observe if the necessary information is complete or some adjustments are needed, see Fig. 6.15. 10. A previous visualization of the piece is generated, see Fig. 6.16. 11. Once the parameters have been established a fast test is run to generate the iterations. It is important to understand that the processing is made in the cloud and hence it has to be uploaded. choose which option is optimal, see Fig. 6.17. As we have previously mentioned, the design requirements were plotted in Table 6.1 by Martinez, and so the outcome of that is a tailored device (Fig. 6.18) which is subject of improvements that is why the aim of this paper is to find that the proposed piece (arch-support, proposal 2) would offer a well balance between its weight and structural stiffness" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002342_kem.480-481.974-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002342_kem.480-481.974-Figure3-1.png", + "caption": "Fig. 3 shows convection boundary conditions on the exposed surfaces of the unit components. The heat transfer coefficients can be defined by Eq.7 [10].", + "texts": [], + "surrounding_texts": [ + "The example of a certain type of bearing with shaft in the gyro motor is taken. Its dimension is 30.6mm\u0445\u044413mm, outer ring land riding. Fig.4 Temperature distribution of bearing with shaft assembly 5 10 15 20 25 30 35 40 45 AXIAL FORCE(N) T E M P E R A T U R E (\u00b7 C ) Outer raceway temperature Temperature of C-type spacer Inner raceway temperature Shaft middle temperature Fig.6 Effect of axial load upon temperature rise 0 2 4 6 8 10 30 35 40 45 RADIAL FORCE (N) T E M P E R A T U R E ( \u00b7C ) Outer raceway temperature Temperature of C-type spacer Inner raceway temperature Shaft middle temperature Fig.7 Effect of radial load upon temperature rise The temperature of bearing with shaft varied with rotational speed as shown in Fig. 5 (the outer ring is stationary, axial load is 8N, radial load is 1N, the initialization temperature is 28\u2103).The temperatures of different parts of bearing with shaft increased when the rotational speed changes from 10000r/min to 30000r/min, however the temperature of inner raceway surface greatly changes (increasing by 16\u2103).The middle temperature of shaft increases relatively obviously. The temperature of C-type spacer doesn\u2019t change very obviously, since the most heat has transferred to the bearing housing. The temperatures of bearing with shaft vary with axial loads as shown in Fig. 6(the outer ring is stationary, rotational speeds is 15000r/min, radial load is 1N, the initialization temperature is 28\u2103). The temperature of inner raceway surface increase by 7\u2103.The middle temperature of shaft increase relatively obviously. However the temperature of C-type spacer only increases by 2\u2103. The temperature of bearing doesn\u2019t change obviously when changing radial load as shown in" + ] + }, + { + "image_filename": "designv11_84_0003599_1.c031306-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003599_1.c031306-Figure6-1.png", + "caption": "Fig. 6 Vertical deflection of tire due to vehicle load.", + "texts": [ + " The analysis becomes slightly more complex if it accounts for the finite deflection of the tire surface caused by vehicle loading. This deflection may be expressed in terms of the easily measured footprint length of the tire using Eq. (8) taken from [1] and derived in the Appendix: R R2 L2=4 p (8) If the assumption ismade that the tire surface clear of the runway is undistorted by runway contact, the tire deflection may simply be added to the height h of the contact point on the stone when calculating the speed of tire\u2013stone contact, Fig. 6. Thus, Eqs. (7) and (8) provide a simple, complete, and relatively accurate computation for the downward-directed velocity of a tire tread as it contacts a D ow nl oa de d by W E ST E R N M IC H IG A N U N IV E R SI T Y o n Ja nu ar y 27 , 2 01 5 | h ttp :// ar c. ai aa .o rg | D O I: 1 0. 25 14 /1 .C 03 13 06 stone surface as a function of stone height, tire geometry, and aircraft forward velocity [1]. For an aircraft tire, the vertical deflection is also a function of the aircraft speed, due to greater lift at higher speeds" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003583_detc2013-12991-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003583_detc2013-12991-Figure1-1.png", + "caption": "Figure 1. TWO PHASES OF THE rTPS LIMB", + "texts": [ + " To demonstrate the method, this paper is arranged in the following structure: section 1 introduces variable topologies of the 3(rT)PS metamorphic parallel mechanism based on two phases of the reconfigurable rTPS limb; section 2 solves the unified inverse displacement, velocity and acceleration analysis, following which, unified dynamic modeling in screw theory is proposed in section 3 with an example simulation illustrated in section 4. Conclusions make up section 5. The under analyzed 3(rT)PS metamorphic parallel mechanism consists of three reconfigurable (rT)PS limbs as in Fig.1. In the limb, there is a reconfigurable Hooke (rT) joint [2], a prismatic joint and a spherical joint. The reconfigurability of this limb stems from the configuration change of the rT joint which has two rotational degrees of freedom (DOFs) about two perpendicularly intersecting rotational axes (radial axis and bracket axis) as in Fig. 1. A grooved ring is used to house the radial axis and make it have the ability of altering its direction by rotating and fixing freely along the groove. This allows the radial rotation axis change with respect to the limb, resulting in two typical phases of the rTPS limb as in Fig. 1. While in Fig. 1(a), the radial axis is perpendicular to the limb (prismatic joint) denoted as (rT)1PS, it is collinear with the limb (prismatic joint) passing through the spherical joint center in Fig. 1(b) and the limb phase is (rT)2PS. In Fig. 1, \u03b2 is the angle between the limb and its projection on the plane passing through rT joint center and perpendicular to the bracket axis, \u03b2 =0 in the (rT)2PS. \u03b1 is the angle between that projection and the line passing through rT joint center and perpendicular to the bracket surface. Three reconfigurable (rT)PS limbs are symmetrically arranged in the 3(rT)PS metamorphic parallel mechanism by connecting the platform and the base as in Fig. 2 with 3(rT)2PS topology. Based on the screw systems in Fig. 1, the (rT)1PS limb has six DOFs with no constraint to the platform and the (rT)2PS limb has five DOFs with a constraint force passing through the spherical joint centre and parallel to the bracket axis of the rT joint. By altering the limb phase between these two, the 3(rT)PS metamorphic parallel mechanism has variable topologies with different mobility. Start from the 3(rT)2PS topology as in Fig. 2 and let points Ai and Bi denote the spherical joint center and the rT joint center in limb i (i=1,2,3) respectively", + " UNIFIED KINEMATICS Since the 3(rT)PS metamorphic parallel mechanism has variable topologies with different mobility each of which is an independent parallel mechanism, how to model those mechanisms in a unified form for applications becomes a challenge. The following sections solve this problem. Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 02/05/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use 4 Copyright \u00a9 2013 by ASME 2.1 Unified Inverse Displacement Analysis Considering difference between the two phases of the rTPS limb, it can be found that the key part is the rotation about the radial axis which can be represented by angle \u03b2 as in Fig. 1. It can be taken as that the (rT)1PS limb has variable angle \u03b2 while the (rT)2PS limb has a fixed angle \u03b2=0. Thus, the (rT)2PS limb can be taken as a special configuration of the (rT)1PS limb by locking the actuation at \u03b2=0. This gives an important method to unify the geometric and kinematics modeling of the 3rTPS metamorphic parallel mechanism by covering all its reconfigurable topologies with mobility change. The following is to use the prismatic joint in the (rT)2PS limb as actuation input and the rotation about the radial axis is added for the second actuation when the limb changes to phase (rT)1PS" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003678_1.4024704-Figure13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003678_1.4024704-Figure13-1.png", + "caption": "Fig. 13 y-direction deformation (in meters) of the workpiece. Welding away from fixed face AA0B0B.", + "texts": [], + "surrounding_texts": [ + "The GTA welding process and the various transport processes involved are discussed in detail in Part I [2]. The mathematical model can be divided into two parts: (a) weld pool dynamics modeling and (b) structural analysis modeling. In the weld pool dynamics modeling, the melting/solidification problem is handled using the enthalpy-porosity formulation. The molten metal flow in the weld pool is obtained using the governing equations of continuity, momentum and energy, based on the assumption of incompressible laminar flow. The Navier\u2013Stokes (N\u2013S) momentum equation takes into account the mushy zone through the momentum sink term, and includes the electromagnetic (Lorentz) force as a body force term. The Lorentz force is determined using the current continuity equation in association with the steady state version of the Maxwell\u2019s equation in the domain of the workpiece for the current density and magnetic flux. The structural analysis model is developed based on isotropic material behavior. The elastic response is handled using the isotropic Hooke\u2019s law with temperature dependent Young\u2019s modulus and Poisson\u2019s ratio. For the inelastic response or plasticity, incompressible plastic deformation is assumed with rate-independent plastic flow and vonMises yield criterion. The yield strength is considered as a function of temperature only. Also, the bilinear isotropic hardening model is employed to consider the material strain-hardening behavior. The mathematical models for both weld pool dynamics and structural analysis have been discussed in extensive detail in Part I and hence is not represented here. However, it is to be noted that the analysis in this study ignores the influences from the arc pressure and a flat weld pool surface is assumed. These assumptions are reasonable for the present study and discussed in detail in Part I of the present study. Also, the boundary and initial conditions used in the mathematical model are described in detail in Part I." + ] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure9-1.png", + "caption": "Figure 9 INPUT GEAR AND CUTTER", + "texts": [], + "surrounding_texts": [ + "This example considers a spur bevel gear set for motion transmission between intersecting axes to introduce the developed process. The tooth profile is a standard involute tooth profile. The nominal gear pair data is presented in Table 2 whereas the nominal cutter data is presented in Table 3. Figures 9 and 10 show the gear elements in mesh with the hyperboloidal cutter elements. No geometric, rating, or manufacturing data are generated." + ] + }, + { + "image_filename": "designv11_84_0003600_12.2072609-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003600_12.2072609-Figure8-1.png", + "caption": "Figure 8 Copper cash abacus and collapsible abacus with support adding", + "texts": [ + " According to this theory, the overhang angle can be divided into three regions at different thickness, including stable fabrication zone, critical fabrication zone, and hard fabrication zone as shown in Figure 7. During SLM manufacturing processing\uff0clayer thickness was set as 0.035mm. Following the Figure6, when overhang angle smaller than 45 degree, the overhang could be stably manufactured, but it is hard to be manufactured when overhang angle larger than 63.4 degree. Therefore, in this article, the manufacturing direction of copper cash abacus and collapsible abacus were display at 30 degree of overhang angle, as shown in the Figure 8. When the overhang angle larger than 63.4 degree, the support structures were added. When the overhang angle between 45-63.4 degree, whether the support structure is necessary or not, can be freely adjusted according to the actual situation. Proc. of SPIE Vol. 9295 929510-5 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 05/18/2015 Terms of Use: http://spiedl.org/terms Based on the results of the present research, Copper cash abacus and collapsible abacus with support adding were sliced and directly manufactured by DiMetal-100" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002243_ecc.2014.6862275-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002243_ecc.2014.6862275-Figure1-1.png", + "caption": "Fig. 1. Schematic PVTOL diagram", + "texts": [ + " In this way we have solved Problem 1. 1 A vector function f(\u03b5, t) \u2208 Rn is said to be O(\u03b5) over an interval [t1, t2] if there exist positive constants K and \u03b5\u2217 such that \u2016f(\u03b5, t)\u2016 \u2264 K\u03b5, \u2200\u03b5 \u2208 [0, \u03b5\u2217], and \u2200t \u2208 [t1, t2], see [4]. VI. ILLUSTRATIVE EXAMPLE In this Section, we test our control scheme in the simplest flying object in a 2-dimensional space, namely: the PVTOL. Readers interested in the PVTOL control can see [8], [6] and the references therein. The behavioral equations of the PVTOL are (see Fig. 1): M d2y dt2 = (f1 + f2) cos \u03b8 \u2212Mg, J d2\u03b8 dt2 = L(f1 \u2212 f2), M d2x dt2 = (f1 + f2) sin \u03b8, (36) where x, y and \u03b8 are the horizontal, vertical and angular displacements, respectively. M is the total mass, g is the gravity constant, J is the moment of inertia and L is the distance from the center of gravity to the trusters. f1 and f2 are the actuators (thrusters). We assume that f1, f2 \u2208 [ 0, F ] , where F = Mg, and: M = 500\u00d7 10\u22123 [Kg], J = 20\u00d7 10\u22123 [Kg m2], L = 200\u00d7 10\u22123 [m], F = Mg = 4.905 [Kg m s\u22122] Defining: v1 = (f1 + f2) cos \u03b8\u2212Mg \u2208 [ \u2212Mg, Mg ] and v2 = (f1 \u2212 f2) \u2208 [ \u2212Mg, Mg ] , namely:[ f1 f2 ] = 1 2 [ 1/cos \u03b8 1 1/cos \u03b8 \u22121 ] [ v1 +Mg v2 ] , (38) we get the simplyfied model: d2y/dt2 = 1 M v1, d2\u03b8/dt2 = L J v2 d2x/dt2 = ( g + v1 M ) tan \u03b8" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure4.8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure4.8-1.png", + "caption": "Fig. 4.8. The arm exoskeleton", + "texts": [ + "7: \u03c4 = G\u0302+ JT (Fdes +K(Fdes \u2212 F)) (4.5) whereK is a 6\u00d76 diagonal matrix of constant gains. Closing the loop at the hand level allows a better measurement of F, although the rejection of friction torques cannot be effective and is strongly dependent on the arm configuration q. L-EXOS (Light Exoskeleton) [12, 13] is an exoskeleton based haptic interface for the human arm. The L-EXOS has been designed as a wearable haptic interface, capable of providing a controllable force at the center of user\u2019s right-hand palm (Fig. 4.8), oriented along any direction of the space. It is a 5 dof robotic device with a serial kinematics, isomorphic to the human arm. Two configurations of the device have been conceived: C1 In the configuration C1 (L-EXOS), an handle is mounted on the last link and the system is composed of 5 DOF, of which only 4 actuated. The non actuated DOF is the last one, aligned along the anatomical prono-supination axis of the forearm (Fig. 4.8). 68 M. Bergamasco, A. Frisoli, and C.A. Avizzano C2 In the configuration C2, the non-actuated DOF and the handle are replaced with an hand-exoskeleton that can apply generic forces on two fingertips of the hand, preferably thumb and index fingers, as shown in Fig. 4.9. The configuration C2 is particularly innovative, since it can reach up to three contact points, one located on the palm of the user, and the other two directly on index and thumb. The following section will focus in particular on the mechanical design and the performance of the exoskeleton in its configuration C1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002805_ijma.2013.055611-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002805_ijma.2013.055611-Figure1-1.png", + "caption": "Figure 1 Objects and robot fingers, (a) case for 3D (m = 6) (b) case for 2D (m = 3)", + "texts": [ + " 2 Possible-generated grasp internal force When a fingertip is in contact with an object at a contact point, the contact finger force vector will depend with the other contact forces on the object. Different other-contact-forces correspond to different finger force vectors. Some finger forces vectors are able to stably grasp the object and some are not. This section will consider what kind of finger force can be classified as a possible stable grasp force. At first, for the multiple objects grasped by a multifingered hand as shown in Figure 1, the discussion in this paper will be based on the assumptions as: 1 The objects are convex polyhedra and rigid, so that the contact parts among the objects are polygonal planes in 3D spatial case or straight lines in 2D planar case. 2 The fingertips are in frictional point-contact with the objects. 3 The contacts among the objects are frictional, and the coefficient of static friction inside a contact part is uniform. Thereby, an s-vertex polygonal plane-contact can be represented by a set of s-point contacts and a 2-endpoint line-contact can be represented by a set of 2-point contacts. 4 At each contact point between a fingertip and an object or between two objects, only force not moment can be transmitted. Thereby, a force on a contact part can be represented by force(s) at one or more than one contact points. A moment on a contact part can be represented by forces at two or more than two contact points. kj = 1, , nj (see Figure 1). Let \u03a3O denote an object frame, x and \u03b8 denote the position and orientation of an object by m-dimensional configuration. Then, let * * *[ , ] ]T T T mw f R\u03c4 \u2208 be a combining form of the force f* and moment \u03c4* acting on objects, where m = 6 (f* and \u03c4* are 3-dimensional vectors) for 3D spatial motion, m = 3 (f* is 2-dimensional vector and \u03c4* is 1-dimensional vector) for 2D planar motion. At first, let us see the case that jth object is in contact with nj fingers. The resultant force fej \u2208 R(3or2) and resultant moment \u03c4ej \u2208 R(3or1) of nj finger forces can be represented respectively as 1 2 (3or2) (3or2) (3or2), , , , jej j j jn j f f f f I I I f = + + + \u23a1 \u23a4= \u23a3 \u23a6 (1) 1 1 2 2 1 2, , , j j j F F F ej j j j j jn jn FF F nj jj j r f r f r f r fr r \u03c4 = \u00d7 + \u00d7 + + \u00d7 \u23a1 \u23a4\u23a1 \u23a4\u23a1 \u23a4 \u23a1 \u23a4 \u00d7= \u00d7 \u00d7\u23a3 \u23a6 \u23a3 \u23a6 \u23a3 \u23a6\u23a3 \u23a6 (2) (3or2) 1 2, , , ,j j T nT T T j j jnj f f ff R\u23a1 \u23a4 \u2208\u23a3 \u23a6 (3) where (3or2) (3or1)[ ] j F jkr R \u00d7\u00d7 \u2208 is a skew symmetry matrix defined by 3or2[ ] ; j j j j j F F F jk jk jk jk jkr f r f r R\u00d7 \u00d7 \u2208 denotes a position vector from the origin of \u03a3O to the touching point of kj th finger at jth object; 3or2 jjkf R\u2208 denotes the contact force of kj th finger; I* is a \u2217-dimensional identity matrix" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003729_detc2011-48019-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003729_detc2011-48019-Figure3-1.png", + "caption": "FIGURE 3: FEM MODEL OF THE FRAME.", + "texts": [ + " The geometric data for the model was obtained by measurements using a 3D measurement device in our laboratory. The 3D geometry model of the frame was imported in the commercial FEM software ANSYS to analyze the structural behavior of the kart frame. The CAD geometry data frame was meshed with shell elements that is a four node element with six degree of freedom at each node. Several interface nodes [4] were prepared to define the joints in the process of flexible multibody dynamics simulation described in the later section of this paper. Figure 3 shows the complete FEM model of the kart frame. When the steering operation was given to the racing kart, the vertical load of each wheel varies according to the movement of the contact point of wheels, which brings the deformation of the frame. The actual deformation of the kart frame according to the steering operation was examined by an experimental way to validate the FEM analysis results. The kart with a driver was positioned on a level plane and vertical load transfers at different steering angle from -80 to 80 degree were measured by four load cells located under the wheels, as is shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure15-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure15-1.png", + "caption": "Figure 15 Crankshaft reaction forces (see online version for colours)", + "texts": [ + " Some friction inevitably occurs in gears: the extra gear absorbs this energy. Thus, a trade-off is required between making the bike as agile as possible and delivering the maximum amount of power to the wheel. Applying a positive torque to the crankshaft turns it in the same direction as the wheels. A reaction torque must be applied to the engine housing at the same time. The direction of this reaction torque is opposite to that of the torque delivered to the crank and creates a lifting force on the front wheel. Figure 15 shows this effect. In the Fedem model, these torques were applied just as explained above, i.e., a torque was applied to the crankshaft, and a reaction torque of the same magnitude but in the opposite direction was applied to the engine housing. These torques operated along the same axis that passed through the centre of the crankshaft (see the red arrows in Figures 13 and 16). The chain pull effect has a great impact on the rear suspension stability and the tyre grip both during acceleration and engine braking" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002496_s10846-013-9971-y-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002496_s10846-013-9971-y-Figure1-1.png", + "caption": "Fig. 1 Coordinate system of the helicopter", + "texts": [ + " The inertial reference frame, also known as earth fixed frame, denoted by f I = [xI yI zI ], is located at a specific position over the earth, where the axis xI is in the direction of the magnetic north, the axis yI is in the direction of the east and zI is in the direction of the earth\u2019s center. The body fixed reference frame denoted by f b = [xb yb zb ], has its origin at the helicopter\u2019s center of gravity (C.G.). The direction of the axis xb is in the longitudinal direction and yb is in the lateral direction of the vehicle. These coordinate systems are depicted in Fig. 1. The fuselage is modelled as a rigid body in a three dimensional space. In terms of body-fixed coordinates the translational and rotational models are described by [19] mV\u0307b + m ( b \u00d7 Vb ) = f b (1) I\u0307b + ( b \u00d7 I b ) = \u03c4 b (2) where Vb = [u v w]T is the translational velocity, with u, v, and w being the longitudinal, lateral and vertical velocities respectively, b = [p q r]T is the angular velocity with p, q, and r being the angular velocities around the xb , yb , and zb axis respectively. The external forces, f b = [X Y Z ]T , and external torques, \u03c4 b = [L M N]T , are expressed in the body-fixed reference frame, where X, Y, and Z indicate the components of the main rotor thrust in the longitudinal, lateral and vertical direction respectively, as well as the tail rotor thrust in the lateral direction" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002334_2013-01-1478-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002334_2013-01-1478-Figure2-1.png", + "caption": "Figure 2. Configuration of high-voltage components", + "texts": [ + " This paper presents a detailed explanation of the low-loss technology applied to this vehicle. Figure 1 shows energy losses of the previous model EV. The losses include acceleration resistance, driving resistance, losses from system components including the motor and the charger, and losses of battery charging and discharging. In this development, improvements have been made with the aim of reducing overall losses by 40%. The following sections explain in detail the improvements made to reduce loss of each type. Figure 2 shows the configuration of the high-voltage components of the developed EV. In order to reduce system losses, reducing losses of the driving motor was required. The new motor was developed with the goal of obtaining a greater efficiency than the previous model, whose average driving efficiency in LA4 mode was 90%. Motor iron loss Wi is determined by Eq.1. (Steinmetz equation) (Eq. 1). where Wh is hysteresis loss, We is eddy current loss, f is frequency of maximum magnetic flux density Bm, and k1 and k2 represent constants" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003766_itec-ap.2014.6941223-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003766_itec-ap.2014.6941223-Figure1-1.png", + "caption": "Fig. 1. Equivalent elliptical loop (EEL) having the same area as the original hysteresis loop", + "texts": [ + " In this condition, the magnetic field intensity in the static hysteresis loop can be decomposed into two components[7], reversible component Hr and irreversible component Hir. Hr is related to reactive power of the material, and Hir is associated with hysteresis losses of the material. As a result, instantaneous hysteresis loss can be computed by h ir dBp H dt = (4) Project 51107133 supported by NSFC 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 60 61 Fig. 1 defines an equivalent elliptical loop in which Hir is evaluated by tracing an elliptical loop having the same area as that of the original hysteresis loop[7]. Thus Hir could be obtained 1 cos( )ir h mH k B \u03b8 \u03c0 = \u22c5 (5) The eddy current loss in the time domain can be expressed as 2 2 1( ) 2c c dBp t k dt\u03c0 \u239b \u239e= \u22c5\u239c \u239f \u239d \u23a0 (6) This FEA method model could compute the coreloss for both soft materials and power ferrite materials. In the 2-D case, the scalar model for soft materials is modified as 22 2 ( ) 1( ) 2 yx h x y yx c c dBdBp t H H dt dt dBdBp t k dt dt\u03c0 \u23a7 = +\u23aa \u23aa \u23a8 \u23a7 \u23ab\u239b \u239e\u23aa \u23aa\u239b \u239e\u23aa = \u22c5 +\u23a8 \u23ac\u239c \u239f\u239c \u239f\u23aa \u239d \u23a0 \u239d \u23a0\u23aa \u23aa\u23a9 \u23ad\u23a9 (7) Ansoft Maxwell is used for the Finite Element Method calculation, Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.11-1.png", + "caption": "Fig. 2.11. Representation of a point P in different coordinate frames", + "texts": [ + " It is easy to see that if Q2 = Q\u22121 1 then the quaternion {1,0} is obtained from (2.37) which is the identity element for the product. See also Problem 2.9. As illustrated at the beginning of the chapter, the position of a rigid body in space is expressed in terms of the position of a suitable point on the body with respect to a reference frame (translation), while its orientation is expressed in terms of the components of the unit vectors of a frame attached to the body \u2014 with origin in the above point \u2014 with respect to the same reference frame (rotation). As shown in Fig. 2.11, consider an arbitrary point P in space. Let p0 be the vector of coordinates of P with respect to the reference frame O0\u2013 x0y0z0. Consider then another frame in space O1\u2013x1y1z1. Let o0 1 be the vector describing the origin of Frame 1 with respect to Frame 0, and R0 1 be the rotation matrix of Frame 1 with respect to Frame 0. Let also p1 be the vector of coordinates of P with respect to Frame 1. On the basis of simple geometry, the position of point P with respect to the reference frame can be expressed as p0 = o0 1 +R0 1p 1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002120_j.euromechsol.2013.11.016-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002120_j.euromechsol.2013.11.016-Figure2-1.png", + "caption": "Fig. 2. A moving frame UVW attached to a moving rigid body.", + "texts": [ + " To this end, we define the following unit vectors: uh p2 p1ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi\u00f0p2 p1\u00de$\u00f0p2 p1\u00de p \u00bc p2 p1 l21 (7) mh p3 p1ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi\u00f0p3 p1\u00de$\u00f0p3 p1\u00de p \u00bc p3 p1 l31 (8) vh m \u00f0m$u\u00deu l ; lh ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi fm \u00f0m$u\u00deug$fm \u00f0m$u\u00deug p (9) whu v (10) In this way, the pose of the bodymay be represented by position vector p1 and the unit vectors u, v andw, which constitute amoving frame UVW (a right-handed orthonormal coordinate system), as it is shown in Fig. 2. It is important to remark that unit vectors u, v and w have been completely defined in terms of the position vectors, p1, p2 and p3, of three noncollinear points pertaining to a moving rigid body. Hence the pose of the body under analysis has been fully characterized in a simple and convenient way. 2.2. Rotation matrix The attitude of the body under analysis may be represented by a suitable rotationmatrix that serve to describe the orientation of the moving frameM : UVW with respect to a fixed frame F : XYZ,3 see Fig", + " Furthermore, five different but equivalent representations of angular velocity vector uP2 can be readily obtained if we make a different choice for m from the vectors m2,m3,.,m6 shown in Appendix B. 4.5. Alternative representations of the angular velocity vector Given the particular derivation of the angular velocity vector proposed in this paper, it is possible to extend the results to three useful and important representations of the angular velocity vector that have been reported in literature Fenton and Willgoss (1990), Kane and Levinson (1985) and Wittenburg (2008). 4.5.1. The angular velocity vector and two nonparallel vectors As it is shown in Fig. 2,m and u are two nonparallel vectors fixed on a rigid body which is rotating with respect to a fixed system XYZ. Such a pair of unit vectors must satisfy the following constraint equations: m$u \u00bc cos g; m$m \u00bc 1; u$u \u00bc 1 (59) equations that, after taking their time derivative lead to: _m$u \u00bc m$ _u; _m$m \u00bc 0; _u$u \u00bc 0 (60) From equations (49)e(51) and (60), the following vector products can be readily computed: _u _v \u00bc _u n 1 l _m m$u l _u o _u _v \u00bc 1 l \u00f0 _u _m\u00de (61) and: _u$v \u00bc _u$ n 1 l m m$u l u o _u$v \u00bc 1 l \u00f0 _u$m\u00de (62) On the other hand, resorting now to equations (58), (61) and (62), it is found that: uK1 \u00bc _u _v _u$v \u00bc l l _u _m _u$m uK1 \u00bc _u _m _u$m (63) Moreover, with the aid of equation (60), equation (63) can be transformed into: uK2 \u00bc 1 2 _u _m _u$m \u00fe _u _m _u$m uK2 \u00bc 1 2 _u _m _u$m \u00fe _m _u _m$u uK2 \u00bc 1 2 _u _m _u$m \u00fe _m _u _m$u (64) Grouping equations (63) and (64) it is finally obtained that: uK1 \u00bc uK2 \u00bc _u _m _u$m \u00bc 1 2 _u _m _u$m \u00fe _m _u _m$u (65) a result that is in agreement with Kane and Levinson (1985)" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002517_amm.658.299-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002517_amm.658.299-Figure3-1.png", + "caption": "Fig. 3. The contact angles Fig. 4. Ball loading", + "texts": [ + " The common contact deformation is determined as the difference between the position vectors corresponding to the final state: - the deformation of the inner contact: { } { } { } Oii b i Owd i Oidi lnrr \u2212\u22c5\u2212=\u03b4 (9) - the deformation of the outer contact: { } { } { } Ooo b i Owd i Ooo lnrr \u2212\u22c5\u2212=\u03b4 (10) Contact Angles. The contact angles of the moved bearing are: - for the inner contact: xxOi zzOi i uvl uvl tg +\u2212\u22c5 +\u2212\u22c5 = 0 0 cos sin \u03b1 \u03b1 \u03b1 (11) - for the outer contact : xOo zOo o vl vl tg +\u22c5 +\u22c5 = 0 0 cos sin \u03b1 \u03b1 \u03b1 (12) where the components of the displacement vectors, Fig. 3, are: { } { } [ ] { }b ba T zaxa a vTvvv \u22c5== ,0, (13) { } { } [ ] { }b ba T zaxa a uTuuu \u22c5== ,0, (14) Ball Equilibrium. The values of the elastic deformations provide the resulting contact force oiQ , : n oiopioi kQ ,,, \u03b4\u22c5= (15) where opik , represents the rigidity of the ball-raceway contact and 5.1=n for the point contact. The equations given in Harris [5], have been used to calculate the rigidities values. The condition of the disconnection has been also considered: 0, =oiQ if 0, \u2264oi\u03b4 (16) For imposed values of zx \u03b4\u03b4 , and y\u03b3 , the components of the displacement vector { }v have to satisfy the equilibrium equations, Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003523_s10948-014-2678-x-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003523_s10948-014-2678-x-Figure1-1.png", + "caption": "Fig. 1 a The floating magnet can carry load. b The setup for magnetic levitation: 1 ring magnet, 2 x axis control coils, 3 hall components, 4 y axis control coils, and 5 floating magnet", + "texts": [ + " have achieved spin-stabilized magnetic levitation [2] and Yang et al. have achieved stable levitation of an iron ball [3]. This paper introduces a control model using a ring magnet for the stable levitation of a cylindrical ferromagnetic object. A stable levitation with a 60-mm air gap has been achieved, and the floating ferromagnetic object can carry a 2-kg load. This model was invented by Wang and Li [4] and has been applied for commercial use. The setup of the levitation device is shown in the Fig. 1. An axially magnetized permanent-magnet ring whose north pole is on its upper surface is placed on the base of the setup. Three hall sensors are used for detecting the instantaneous coordinates of the floating magnet at three axes. The horizontal position of the floating magnet is controlled by two groups of coils. The north pole of the floating magnet is also on the upper surface, which is inconsistent with the common sense that unlike poles attract each other, not repelling each other to provide antigravity force for the floating magnet" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003999_s12283-014-0158-y-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003999_s12283-014-0158-y-Figure5-1.png", + "caption": "Fig. 5 A ball that spins about its frontal velocity axis", + "texts": [ + " Now, state of motion at points A, B and D are evaluated and the scheme is continued until the normal relative velocity between the balls, _yA\u2014as in Eq. (5c)\u2014becomes negative, i.e., the termination of the compression phase. The work done up to this iteration is Wy Pc I . Equation (10c) is used to determine Wy P f I . The numerical algorithm is stopped when, W \u00bc Wy P f I : \u00f015\u00de 2.5 Parabolic path subsequent to impact Execution of the numerical scheme that is described in Sect. 2.4.2 shows that, in general, both the cue ball and the object ball will have spins about their frontal velocity axes, as shown in Fig. 5. The ball shown in Fig. 5 has a spin component of x1 about its centroid velocity V. In this case, irrespective of the other two spin components, the ball will move along a curved path. This is called masse\u0301 in billiards [21]. Curved shots can be made by elevating the cue when striking the ball. Curved ball trajectories are also produced due to frictional percussions during the impact between two balls or that between a ball and a cushion. The second type is of interest here. However, the derivations given in this section are, essentially, applicable for any general curved shot" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001981_detc2013-12899-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001981_detc2013-12899-Figure1-1.png", + "caption": "Figure 1 CYLINDRICAL HOB WITH SPUR GEAR.", + "texts": [ + " For example, the machines used to produce hypoid gears cannot be readily used or retrofitted for fabricating spur cylindrical gears. The methods of manufacture associated with bevel and hypoid gears do not allow these gears to be treated with the same type of geometric considerations that currently exist for cylindrical gears. To illustrate, spur cylindrical gears are helical gears with a zero helix angle and both gear types are produced using the same machine. Spur hyperboloidal gears cannot be produced using existing fabrication techniques for spiral hyperboloidal gears. Depicted in Figure 1 is a cylindrical hob in mesh with a spur gear. Depicted in Figure 2 is a cylindrical hob cutter with gashes or cutting flutes. Here, only the motion of the hyperboloidal hob cutter is considered. 1 Copyright \u00a9 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use The majority of hypoid and bevel gear manufacture today is the focus of The Gleason Corporation and KlingelnbergOerlikon. Today, the following three companies provide the machines and machine tools for the production of crossed axes gear pairs: The Gleason Works\u00f1 Klingelnberg-Oerlikon\u00f1 Yutaka Seimitsu Kogyo, LTD" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002937_amc.2014.6823323-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002937_amc.2014.6823323-Figure4-1.png", + "caption": "Fig. 4. Different phases in traveling throughout a slope", + "texts": [ + " In such cases, since the user and the robot are not on the same plane (same slope or same horizontal plane), their respective velocity will be different. If the admittance control as shown in Eq.(2) is directly adopted, the load that the user feels will inevitably have suddenly change. Here, our purpose is to make the user\u2019s load feeling have no big change throughout traveling a slope, i.e, from traveling on a horizontal plane, starting ascending a slop, leaving the slope to a horizontal plane, to descending the slope to horizontal plane as shown in Fig. 4. Therefore, we first investigate the movement when starting to ascend the slope, and propose an approach to suppress the load change of the user during the movement. Then we introduce the iterative learning control into the admittance in order to achieve the same \u201cload feeling\u201d as the horizontal plane. When the user and the robot are not on the same plane, for example, when the robot starts to ascend the slope (as phase B shown in Fig. 4), in which, the user is on a horizontal plane while the front wheels of the robot are already on the slope, the respective velocities of the user and the robot will be different. Now we analyze the movement in phase B of starting to ascend a slope. As shown in Fig. 5, F and R are respectively the centers of the front and rear wheels of the robot; A is the user\u2019s holding point at the handle bar of the robot; H is the height of the handle bar from A to R. Since the front wheels of the robot are on the slope, their velocity VF is along the slope too", + " If the robot starts to ascend a slope with the same traveling velocity as the one on a horizontal plane, then as aforementioned, the user\u2019s necessary velocity to get the same \u201cload feeling\u201d as on the horizontal plane should be less than the robot\u2019s velocity. However, at this moment of time, since the user is still on the horizontal plane and his/her walking velocity is kept unchanged as before the robot starts to ascend the slope, the user\u2019s velocity is larger than the necessary velocity. Therefore, according to Eq.(2) or Eq.(3), the user\u2019s \u201cload feeling\u201d will inevitably get larger than the load on the horizontal plane. As a result, it is necessary to increase the robot\u2019s traveling velocity at phase B as shown in Fig. 4. The new desired velocity Vd1 of the robot should be{ Vd1 = Vd + Vc1 Vc1 = ( 1 s1 \u2212 1) \u00b7 Vd (12) where, Vc1 is the robot\u2019s compensation velocity. So that the horizontal component VH of the velocity VA at the holding point A on the handler bar coincides with the walking velocity of the user on the horizontal plane. Consequently, the same load feeling as walking on the horizontal plane can be expected. Moreover, the above analysis and expressions for phase B can be completely applied to the phase H as shown in Fig. 4, except the robot\u2019s rotation direction is contrary to the direction in phase B. Similarly, for phases D and F as shown in Fig. 4, in which the robot are respectively leaving from a slope to a horizontal plane and leaving a horizontal plane to descend a slope, the velocity VA at the holding point A on the handler bar is larger than the robot\u2019s velocity VR. The desired walking velocity of the user is VH2 = s2 \u00b7 VR (13) Hence, the desired velocity Vd2 of the robot should be{ Vd2 = Vd + Vc2 Vc2 = ( 1 s2 \u2212 1) \u00b7 Vd (14) where, s2 = 1 + H cos(\u03b80 \u2212 \u03b8) sin \u03b8 L cos \u03b8 > 1 (15) With Eq.(14) and Eq.(15), when the robot starts to descend a slope, the user will get the same \u201cload feeling\u201d on the horizontal plane", + "(19), the relationship between E(n) and E(n\u2212 1) is E(n) = F (n)\u2212 Fd = (1 +K \u00b7G) \u00b7 E(n\u2212 1) (20) At this time, K is a known coefficient. By appropriately selecting the learning gain G, it is possible to let |1 + kG| less than 1. This means is |E(n)| < |E(n\u2212 1)| \u21d2 |E(n)| \u2192 0 (as n \u2192 \u221e) (21) Consequently, in theory, a more smooth load feeling should obtained by the parameters D in admittance control adaptively generated with iterative learning control. Here, we confirm our proposed approaches experiments in which our wheelchair typed robot travels through a slope as shown in Fig. 4. According to the Japan\u2019s national standards, the gradient of the public facility slope should be under 1/12, i.e, the angle of the inclination of a slope should be less than 4.76\u25e6. Thus, the inclination angle of the slope used in our experiments is 5.0\u25e6. As aforementioned, during traveling through the whole slope under admittance control, since both the user and the robot are on the same plane and their velocities are same in phase C and D, the load the user feels is same as the load on the horizontal plane as in phase A, I , as well as E", + " Contrarily, in phase D and F since the user\u2019s walking velocity is larger than the robot\u2019s velocity, Eq.(14) is used to speed down the robot in order to get a small load fluctuation. Further, the load fluctuation is further reduced by adaptively altering the damping coefficient D with iterative learning control. The diagram of the control process is illustrated in Fig. 6. To save the space, the experimental results of the robot traveling from a horizontal plane, going up a slope and stopping at a horizontal plane of the slope summit, as phase A to phase E in Fig. 4 are presented in this paper. At first, we conduct an experiment with admittance control as a preliminary experiment. The two parameters in admittance control, the virtual damping D and the time constant \u03c4 are respectively set to be D = 25N \u00b7 s/m and \u03c4 = 0.3s (when the user\u2019s velocity is 0.4m/s, then in theory, his/her load is about 0.4\u00d7 25 = 10N). Fig. 7 shows the preliminary experimental result in ascending a slope from phases A to E as showns Fig. 4. Here, the subject is holding the robot handle to ascend a slope while no person sitting in the seat. The result shows that the user\u2019s load fluctuation in phase B is about 3.5N This is because in phase B, the user\u2019s walking velocity is still same as the velocity on horizontal plane and it is larger than the desired velocity expressed by Eq.(12). After phase B and when both the user and the robot are on the slope in phase C, the load about 9.5N. This is equivalent to load on horizontal plane in steady state" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001195_978-81-322-1656-8_29-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001195_978-81-322-1656-8_29-Figure3-1.png", + "caption": "Fig. 3 Schematic of a foil bearing and detailed configuration of the foil [5]", + "texts": [ + " 4 is simplified as, \u00f0Pi; j\u00de2K1 \u00fe \u00f0Pi; j\u00deK2 \u00fe K3 \u00bc 0 \u00f06\u00de where Pi;j is the pressure at any point \u00f0i; j\u00de and K1, K2 and K3 are given in appendix A Equation 6 is a nonlinear system of the form F\u00f0P\u00de \u00bc 0 \u00f07\u00de Using Newton\u2013Raphson method for its solution, Pn\u00fe1 \u00bc Pn F\u00f0Pn\u00de F0 \u00f0Pn\u00de \u00f08\u00de where, Pn is the pressures obtained after nth iteration and F 0 \u00f0P\u00de is first derivative of F\u00f0P\u00de with respect to P. Newton\u2013Raphson iterative process is repeated until the following convergence criterion is satisfied. \u00f0 P Pi;j\u00den 1 \u00f0 P Pi;j\u00den \u00f0 P Pi;j\u00den 10 6 \u00f09\u00de A simple elastic foundation model which is the original work of Heshmat et al. [5] is considered for this work as shown in Fig. 3. The local deflection of the foil structure (wt) depends on the bump compliance (a) and the average pressure across the bearing length, wt \u00bc a\u00f0 p pa\u00de \u00f010\u00de The compliance of the bump foil, with the geometry specified in Fig. 3 is determined by Heshmat et al. [5] as: a \u00bc 2s Eb lo tb 3 1 t2 \u00f011\u00de where; s = Bump pitch (m) lo = Half of the bump length (m) tb = Bump foil thickness (m) Eb = Young\u2019s modulus (Modulus of elasticity) of bump foil material m = Poisson\u2019s ratio of bump foil material By using following substitutions: P \u00bc p pa , W \u00bc wt C Non-dimensional foil deflection equation is given by where S is so-called the Compliance number. To obtain steady state characteristics of GFB, it is required to obtain pressure distribution by solving Eq" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002763_12.880610-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002763_12.880610-Figure2-1.png", + "caption": "Figure 2. Diagram showing the unmorphed and fully morphed configurations of the aircraft. The morphing degrees of freedom are: (a) wing incidence, (b) tail boom angle, and (c) tail incidence.", + "texts": [ + " Section 3 presents two feedback control methods, and Section 4 describes how these controllers are tuned to the aircraft system. Section 5 presents simulation results and a discussion of the compensators\u2019 performance. Section 6 provides the conclusions of this paper. This work uses the ARES-C aircraft model,5 a redesigned variant of the Aerial Regional Environmental Survey vehicle.6 This variant includes actuated degrees of freedom that allow for large-scale morphing of the aircraft. Morphing in this case refers to the bulk rotation of rigid aircraft structures: wings, tail boom, and tail platform. Figure 2 shows a sketch of the aircraft and identifies the locations of the morphing axes of rotation. The wings and the tail boom rotate about the fuselage, and the tail platform rotates about the tail boom. The aircraft uses traditional control surfaces, such as ailerons and ruddervators, to control its flight. The morphing structures and rotational axes of the ARES-C provide the aircraft with the ability to change its form in a manner that reflects the morphological changes of a bird, especially with regards to the perch maneuver" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002923_gt2013-95585-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002923_gt2013-95585-Figure4-1.png", + "caption": "Figure 4. SKETCH OF LEAF ROTATION", + "texts": [ + " It is thus inferred that hydrodynamic air-riding is unlikely to create gaps, G, in excess of 2.0\u00b5m. To predict the changes in slope height h, it is necessary to consider how the leaf lay angle changes during seal operation. The nominal leaf lay angle is a parameter set during initial seal design, however during operation, especially when the leaf tip is moved radially, the leaf lay angle is altered slightly. This change can be characterised by considering each of the leaves as a beam that hinges at its support, as shown in Fig. 4. Such a crude approximation is permissible, as the bending moments in the leaf are largest at the root, the deflections are small and only the effect on the tip is of interest here. From this approximation it can be shown that \u2206\u03b80 = \u2206\u03b8, (4) and that \u2206R = L sin(\u03b80)\u2212L sin(\u03b80 \u2212\u2206\u03b80) \u2248 L\u2206\u03b80 cos(\u03b80) (5) for small values of \u2206\u03b8 when cos(\u03b80) is approximately invariant. The relationship in Eqn. (5) also holds for leaves that are curved, however the constant L cos(\u03b80) needs to be adjusted. By combining Eqns" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001907_00022661111159861-Figure12-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001907_00022661111159861-Figure12-1.png", + "caption": "Figure 12 The contours of relative pressure overlay with velocity vectors around the exterior surface of airship model from numerical simulation", + "texts": [ + " The discretized nonlinear algebraic equations are solved by tri-diagonal method algorithm iterative method. The aerodynamic forces are categorized into various terms based on different physical effects: the added-mass force, the viscous effect on the hull, the force on the fins, the force on the hull due to the fins, and the axial drag.Computationalmethods are provided for each aerodynamic term and incorporated into the nonlinear dynamics model. The simulation results are then verified by comparing with wind-tunnel data. The isobars of relative pressure are shown in Figure 12, with overlay of velocity vectors over the exterior surface of the airship model. As observed, the low-pressure zone is created on the frontal portion of the airship Alves Ribeiro et al. Volume 83 \u00b7 Number 5 \u00b7 2011 \u00b7 255\u2013265 model with the value of 3.53 \u00a3 103 Pa. The low-pressure area is also foundon frontpart offins andat tail endof the airshipmodel. The contours of turbulent kinetic energy around the exterior surface of airship model from numerical simulation is shown in Figure 13, with the maximum values at the frontal end of airship and also near the frontal part of the fins" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001550_chicc.2015.7259633-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001550_chicc.2015.7259633-Figure1-1.png", + "caption": "Fig 1. Ground coordinate and unfixed coordinate.", + "texts": [ + " Aim at above problem, variable universe fuzzy sliding mode controller is choose in this paper which combined fuzzy controller with sliding mode control, not knowing accurate mathematic model, according fuzzy information of system, it can get better control effect, so the system can achieve optimum control effect. The simulation proved that the control method can produce better effect. In order to describe the motion of submarine and study the control system simulation, we must set up coordinate which can describe the motion of submarine. The coordinate system is shown in Fig. 1, including Ground coordinate and unfix coordinate. The matrix[14] of coordinate conversion from unfixed coordinate system to ground coordinate system is O ER which is shown in (1). cos cos sin cos sin sin cos sin cos cos sin sin sin cos sin sin sin cos cos sin sin cos cos sin sin cos sin cos cos + = +O ER (1) Where, , , is attitude angle of submarine. Using the defined variable in Table.1 we describe the speed and rotational speed in unfixed coordinate, position and attitude in ground coordinate, and external force and external moment" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.8-1.png", + "caption": "Fig. 7.8 Description of the Orthoglide kinematic chain. a Kinematics of one leg. b Equivalent kinematics of one leg. c Fixed base", + "texts": [ + " The reader is referred to Bonev (2002) for further investigations. TheOrthoglide is a parallel robot composed of three identical legs (Fig. 7.7) allowing three translational DOF of its end-effector (parameterized by the variables x , y and z that represent respectively the translation along x0, y0 and z0 of the base frameF0). Each leg is made of one linearly actuated link (parameterized by the variables qi1, i = 1, . . . , 3, i.e. qT a = [q11 q21 q31]) linked at its extremity to a spatial parallelogram (Fig. 7.8a). The parallelogram is also attached to the mobile platform. Kinematically speaking for obtaining the inverse kinematics, and without loss of generality, each parallelogram chain can be replaced by an equivalent chain composed of two Ujoints connected by a rigid link (Fig. 7.7b). The directions of the three linear actuators of the Orthoglide are orthogonal (Fig. 7.7b). The purpose is to create a mechanism with a workspace shape close to a cube and whose behavior is close to the isotropy (Merlet 2006a) wherever it is located in its workspace (Chablat and Wenger 2003)", + " Considering in the present example that the end-effector location is parameterized by the homogeneous transformation, 0Tp = \u23a1 \u23a2 \u23a2 \u23a3 1 0 0 x 0 1 0 y 0 0 1 z 0 0 0 1 \u23a4 \u23a5 \u23a5 \u23a6 it get that, for leg 1: \u23a1 \u23a3 x y z \u23a4 \u23a6 = \u23a1 \u23a3 d4 cos q12 cos q13 + d6 \u2212d4 sin q13 q11 \u2212 d4 cos q13 sin q12 \u23a4 \u23a6 (7.23) for leg 2: \u23a1 \u23a3 x y z \u23a4 \u23a6 = \u23a1 \u23a3 \u2212a + q21 \u2212 d4 cos q23 sin q22 d4 cos q22 cos q23 + d6 a \u2212 d4 sin q23 \u23a4 \u23a6 (7.24) and for leg 3: \u23a1 \u23a3 x y z \u23a4 \u23a6 = \u23a1 \u23a3 d4 cos q32 cos q33 + d6 \u2212a + q31 \u2212 d4 cos q33 sin q32 a + d4 sin q33 \u23a4 \u23a6 (7.25) where a, d4 and d6 are geometric parameters defined in Fig. 7.8a, and, to take into account the parallelogram constraints, we have [ 0 0 ] = [ qi2 + qi5 qi3 + qi4 ] (7.26) Simplifying (7.23)\u2013(7.25), it turns out that: hp = \u23a1 \u23a3 (x \u2212 d6)2 + y2 + (z \u2212 q11)2 \u2212 d2 4 (x + a \u2212 q21)2 + (y \u2212 d6)2 + (z \u2212 a)2 \u2212 d2 4 (x \u2212 d6)2 + (y + a \u2212 q31)2 + (z \u2212 a)2 \u2212 d2 4 \u23a4 \u23a6 = 0. (7.27) Developing, each row of (7.27) leads to a polynomial of the second order in qi1 q2 i1 + ci1qi1 + ci0 = 0, for i = 1, . . . , 3 (7.28) where c11 = \u22122z c10 = (x \u2212 d6) 2 + y2 + z2 \u2212 d2 4 c21 = \u22122(x + a) c20 = (x + a)2 + (y \u2212 d6) 2 + (z \u2212 a)2 \u2212 d2 4 c31 = \u22122(y + a) c30 = (x \u2212 d6) 2 + (y + a)2 + (z \u2212 a)2 \u2212 d2 4 from which we can find: qi1 = \u2212ci1 \u00b1 \u221a c2i1 \u2212 4ci0 2 ", + "133) Finally, we have Ar 0tr + Bq\u0307a = 0. (7.134) In this section, we study only the input/output kinematic relations of the Orthoglide introduced in Sect. 7.1.2.4. The computation of the passive joint velocities is tedious, this is the reason why it is not detailed here but is given in: http://www.irccyn.ec-nantes.fr/~briot/Books.html. Following the method of the Sect. 7.3.1, and by using the results presented in the AppendixC.4 and using the fact that, when the actuator i is blocked, the leg shown in Fig. 7.8b is a UU passive system, we have a matrix A equal to: AT = [ \u03b6 1 \u03b6 2 \u03b6 3 ] (7.135) with \u03b6 T i = 1 d4 [ 0rT Ai3 Ai4 ( 0rP Ai4 \u00d7 0rAi3 Ai4 )T ] (7.136) where the points Aij are described at Fig. 7.8, and d4 = \u2225 \u2225rAi3 Ai4 \u2225 \u2225 is a constant length defined in Table7.3. \u03b6 i is a force directed along \u2212\u2212\u2212\u2212\u2192 Ai3Ai4. Moreover, for any leg of the robot, $i1 = $ia is a twist representing a pure translation along the P joint direction. As a result, 0$T 1a = [ 0 0 1 01\u00d73 ] (7.137) 0$T 2a = [ 1 0 0 01\u00d73 ] (7.138) 0$T 3a = [ 0 1 0 01\u00d73 ] . (7.139) Thus, the matrix B is equal to: B = \u2212 1 d4 \u23a1 \u23a3 \u03b6 T 1 0$1a 0 0 0 \u03b6 T 2 0$2a 0 0 0 \u03b6 T 3 0$3a \u23a4 \u23a6 (7.140) with \u03b6 T 1 0$1a = z A13 A14 (7.141) \u03b6 T 2 0$2a = xA23 A24 (7" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001116_978-3-319-19788-3_7-Figure7.13-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001116_978-3-319-19788-3_7-Figure7.13-1.png", + "caption": "Fig. 7.13 The two assembly modes of the planar five-bar mechanism", + "texts": [ + "6, a list of reference papers dealing with the FGM of other SPM is provided. Let us consider again the five-bar mechanism presented in Sect. 7.1.2.1. Starting from (7.10) and developing the expressions, we get hp(x, qa) = [ x2 + y2 + a1x + b1y + c1 x2 + y2 + a2x + b2y + c2 ] = 0 (7.38) where ai = \u22122(di1+di2 cos qi1), bi = \u22122di2 sin qi1 and ci = (di1+di2 cos qi1) 2+ d2 i2 sin 2 qi1 \u2212 d2 i3. From a geometric point of view, solving the two first equations of (7.38) is equivalent to finding the intersection points of two circles (Fig. 7.13): \u2022 Circle C1 centered in A12 of radius d13, which corresponds to the vertex space of the point A13 when considering that it belongs to the link A12A13, \u2022 Circle C2 centered in A22 of radius d23, which corresponds to the vertex space of the point A13 when considering that it belongs to the link A22A13. Thus, the two robot assembly modes correspond to the intersection points of circles C1 and C2 whose expressions are, if b1 = b2, x = \u2212 f2 \u00b1 \u221a f 22 \u2212 4 f1 f3 2 f1 y = e1x + e2 (7.39) where f1 = 1 + e21 or, if b1 = b2 x = \u2212 c2 \u2212 c1 a2 \u2212 a1 y = \u2212b1 \u00b1 \u221a b21 \u2212 4(x2 + a1x + c1) 2 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001895_amm.288.208-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001895_amm.288.208-Figure4-1.png", + "caption": "Fig. 4 Spiral bevel gear 3D modeling process", + "texts": [ + " The gear tooth surface boundary intersects to four points, A1, A2, A3, and A4. Their coordinates can be calculated based on gear blank parameters. The coordinates and the deduced tooth surface are substituted for the coordinate projection formula to acquire formula (5), to compute the scope of the gear tooth surface parameters. =\u2212 =+\u2212 0),( 0),(),( 22 \u03b8 \u03b8\u03b8 urZ ururR wzi wzwyi (5) Where rwx(u, \u03b8), rwy(u, \u03b8) and rwz(u, \u03b8) are the first three components of rw(u, \u03b8); (Zi, Ri)is the two-dimensional coordinates of A1, A2, A3, and A4 in Fig. 4, and i equals one to four. According to the tooth surface equations rw(u, \u03b8), generally, a parameter is seen as a variable and another is changing its value to gain a series of curves, in which u is the parameter along the tooth profile direction and \u03b8 is the parameter along the tooth width. When u is a constant, the theoretical tooth width curve is received; when \u03b8 is a constant, the theoretical tooth curve is generated. The density of curves is determined by the dispersion degree of u and \u03b8 values", + " With higher accuracy surface, the program runs longer, so the dispersion degree should be appropriately chosen in accordance with actual requirements, here we take n equal to 50 for example. Then parameter values are substituted in the tooth surface equation and the normal vector for n * n discrete points on the tooth surface saved to a file, providing data sources for 3D modeling. Gear specific modeling steps are as follows: (1) The txt file output from MATLAB was changed to ibl format identified by Pro/E, and then was imported into Pro/E, as shown in Fig.4 (a). (2) The discrete points imported were fitted to curves by the curve fitting tool in Pro/E, seen in Fig.4 (b). (3)The free curves above were constructed to a smooth surface by boundary mixing tools, as Fig.5 (c) shows. (4) Repeated the above steps to complete configuration of convex and concave (See Fig.4 (d)). (5) Drew the front tapered surface, back cone, top cone and root cone, as Fig.4 (e) shows. Used the pruning tool to make Boolean operations on convex and concave surfaces established and four cones, and then used the boundary suture command to obtain a curved cogging (Fig.4 (f)). (6) Arrayed the curved cogging, as shown in Fig. 4(g). Made Boolean subtraction on these cogging and the gear blank to generate surfaces, and then to be materialized as a 3D solid model, shown in Fig.4(h). The model established by 3D software was saved as an STL format file [4], and was imported into VGMC (Fig.5). Fig.5 Spiral bevel gear solid model imported into VGMC After clamping virtual workpiece under the virtual environment in Fig.5, the model data involved in computing was loaded. The upper measurement procedures communicated through a dynamic link library with VGMC to simulate the virtual measurement. At the same time, the interface between the dynamic link library and the upper measurement procedures remained unchanged" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003565_j.mechmachtheory.2011.08.004-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003565_j.mechmachtheory.2011.08.004-Figure1-1.png", + "caption": "Fig. 1. Dynamic models of (a) an eight-planet toroidal drive, (b) a worm/planet pair, (c) a stator/planet pair, and (d) a rotor/planet pair.", + "texts": [ + " When number of pole pairs of the worm coils p=2, the stator is divided into four parts. If the tooth number of the stator is increased, pairs of the N pole and S pole should be increased in each part. So, the tooth number increased should be multiple of eight, 8n (here n=0, 1, 2 \u2026) and the tooth number of the stator z0=4+8n=4(1+2n). Hence, the tooth number of the stator should be taken as following z0 \u00bc 2p 1\u00fe 2n\u00f0 \u00de \u00f017\u00de where n is the positive integer. Substituting Eq. (17) into Eq. (12) or Eq. (15), the speed ratio of the drive system for two output ways can be given. Fig. 1 gives the dynamic model of a toroidal drive with eight planets(m=8). The planets are mounted on a rotor through bearings and pins. The main assumptions employed in the dynamic model are identical to the ones in ref. [12]. The difference between them is that stator support is modeled as linear springs acting in the two coordinate axes directions here, and it is modeled as rigid body in ref. [12]. The toroidal drive in Fig. 1(a) is considered as a combination of three sub-systems (i) a worm/planet pair [Fig. 1(b)], (ii) a stator/ planet pair [Fig. 1(c)], and (iii) a rotor/planet pair [Fig. 1(d)]. The dynamic model allows worm, stator, rotor and each planet to rotate about these translational axes and allows each planet to translate in x and y directions. For the sake of convenience, the rotations are replaced by their corresponding translational mesh displacements as uj \u00bc rj\u03b8j; j \u00bc w; s;p; r\u00f0 \u00de where \u03b8j is the rotation of worm, stator, rotor and planet, rj is the rolling circle radius for worm, stator, planet and the radius of the circle passing through planet centers for the rotor. Fig. 1(b) shows a worm/planet pair which represents the worm (subscript w) meshing with planet-i (subscript pi). Here, \u03b3wpi is the lead angle of the helix tooth at reference circle on the worm and \u03b4 is the half cone angle of the planet tooth; tan \u03b3wp=1/[iwp (a/R+1)], a is the center distance between worm and planet, R is the reference circle radius of planet, iwp is the speed ratio between planet and worm in coordinate system attached to the rotor; kwp is the mesh stiffness between the planet and the worm", + " The displacement pspi equals algebraic sum of the displacements in the contacting line direction between planet and stator. Thus, one obtains pspi \u00bc \u2212ui cos\u03b4 cos\u03b3spi\u2212us cos\u03b4 sin\u03b3spi\u2212xi cos\u03b4 sin\u03b3spi\u2212yi sin\u03b4 \u00f022\u00de Substituting Eq. (22) into Eq. (21), yields Ms 0 0 mii us :: qi :: \u00fe kuss kis T kis ks us qi \u00bc 0 0 \u00f023\u00de where kus=kss+kspi cos 2\u03b4 sin 2\u03b3spi kis T \u00bc kspi cos2\u03b4 cos\u03b3spi sin\u03b3spi cos2\u03b4 sin2\u03b3spi sin\u03b4 cos\u03b4 sin\u03b3spi h i ks \u00bc kspi cos2\u03b4 cos2\u03b3spi cos2\u03b4 sin\u03b3spi cos\u03b3spi sin\u03b4 cos\u03b4 cos\u03b3spi cos2\u03b4 sin\u03b3spi cos\u03b3spi cos2\u03b4 sin2\u03b3spi sin\u03b4 cos\u03b4 sin\u03b3spi sin\u03b4 cos\u03b4 cos\u03b3spi sin\u03b4 cos\u03b4 sin\u03b3spi sin2\u03b4 2 64 3 75 Fig. 1(d) shows the model of the rotor/planet pair. Here, Kr denotes the torsion stiffness of the rotor. From force and torque balance conditions of the rotor and the planet, one can give Mru :: r \u00fe krur \u00fe kcxi xi\u2212ur\u00f0 \u00de \u00bc 0 mpi ui :: \u00bc 0 mi xi :: \u00fekcxi xi\u2212ur\u00f0 \u00de \u00bc 0 mi yi :: \u00fekcyiyi \u00bc 0 8>< >: \u00f024\u00de Eq. (24) can be given in matrix form as Mr 0 0 mii u :: r q :: i \u00fe krr krpi kTrpi kc \" # ur qi \u00bc 0 0 \u00f025\u00de where kc \u00bc 0 0 0 0 kcxi 0 0 0 kcyi 2 4 3 5, mii \u00bc mpi 0 0 0 mi 0 0 0 mi 2 4 3 5, krr=kr\u2212kcxi, krpi=[0kcxi0] Eqs. (20), (23) and (25) can be combined systematically to obtain the dynamic equations of the overall drive system which consists of a worm, a stator, a rotor and m planets. The dynamic model of the drive system is given in Fig. 1(a). The undamped dynamics equations of the 3m+3 DOF model of the drive system are given in matrix form as Mq \u22c5\u22c5\u00fe Kq \u00bc 0 \u00f026\u00de The displacement and load vectors, q and F, and the mass and the stiffness matrix, M and K, are given respectively as follows q \u00bc uw q1 q2 q3 \u2026 \u2026 qm ur usf gT M \u00bc diag Mw m11 m22 \u2026 \u2026mii Mr Msf g K \u00bc \u2211 m i\u00bc1 kuw k1 T k2 T \u2026\u2026 ki T :::: km T 0 0 kc \u00fe kw \u00fe ks 0 \u2026\u2026 0 \u2026 0 kTrp1 k1s kc \u00fe kw \u00fe ks 0 :::: 0 \u2026 0 kTrp2 k2s :0 ::::: 0 \u2026 0 : : :0 ::::: 0 \u2026 0 : : kc \u00fe kw \u00fe ks 0 \u2026 0 kTrpi kis :0 \u2026 0 : : :0 \u2026 0 : : kc \u00fe kw \u00fe ks krpm kms krr 0 symmetric \u2211 m i\u00bc1 kus 2 6666666666666666666664 3 7777777777777777777775 Eq" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002494_amr.1028.105-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002494_amr.1028.105-Figure1-1.png", + "caption": "Fig. 1 Structural diagram of the conveyor", + "texts": [ + " Firstly, you need to analyze the static structure, calculated as follows: [ ] { } { }FxK =\u2022 (2) The stress stiffness matrix ( [ ] [ ]S\u21920\u03c3 ) is used in modal analysis, the original modal equation is modified into the following: [ ] [ ]( ) { } 0 2 =\u2022\u2212+ ii MSK \u03c6\u03c9 (3) The ultimate goal of modal analysis is to identify the modal parameters of the system and provide a basis for the analysis of the vibration characteristics of the structure of the system, vibration fault diagnosis, prediction and optimization of dynamic characteristics of the design of the structure [7]. The toothpaste tube conveyor is a part of automation equipment of toothpaste tube shoulder seaming. To fit to flip the tube strictly requires that synchronous transmission belt can stop when the tube reaches the specified position, which demands that the stepping motor is controlled strictly. The different speeds of the stepping motor may cause different vibration amplitudes of the drive shaft. So it is important to control the speed of the stepping motor. As figure 1 shown, the frame structure of the conveyor is created by Solidworks [5]. In the design process, the axial length of the drive shaft is 347mm, the diameter of the end portion is 20mm and the diameter of the central portion is 22mm according to the length of the tube to be transmitted. It is very important to establish accurate and reliable structural FEM model when the finite element method is applied into the pre-stressed modal analysis. The application of ANSYS Workbench is divided into two steps: the establishment of the solid model and the finite element model" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003191_amr.216.539-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003191_amr.216.539-Figure3-1.png", + "caption": "Fig. 3 16-pole 15-slot PMLSM FEM model", + "texts": [ + " Finally, the direct driving high-speed elevator has four security protection measures. In the conditions of system power failure or other emergency, excepting three conventional protection measures, working brake, safety gear over-speed protection and buffer protection, there is the fourth power failure generation protection, which is the inherent character of PMLSM with winding shorted. The driving motor is designed by finite element method (FEM), 16-pole 15-slot as a unit. The FEM model of the unit driving motor is shown in Fig. 3. Fig. 4 shows the waveform of force with different power angle. The cogging force of 16-pole 15-slot motor is shown in Fig.5. Waveform of force with different power angle 0 250 500 750 1000 1250 -10 0 10 20 30 40 50 60 70 80 90 Power angle(deg) V al u e( N ) Cogging force of 16p15s motor -90 -60 -30 0 30 60 90 0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25 Position(mm) C o g g in g f o rc e( N ) Fig. 4 Force with different power angle Fig. 5 Waveform of the cogging force As Fig.4 shown, considering 1.25 time of overload magnification, the best operating force of the unit driving motor with 16-pole 15-slot, namely the rated force, is 1000N" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003877_pedstc.2013.6506716-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003877_pedstc.2013.6506716-Figure1-1.png", + "caption": "Figure 1. SR machine and one phase bifilar winding", + "texts": [ + " The problem of optimal control for accomplishing maximum energy conversation in SRG is investigated in [3]. Phase advancing in a high speed, 6 by 4 switched reluctance generator is perused in [4]. Some researches are on SRM or SRG including bifilar converter. A new bifilar converter which uses dump capacitor during discharge period is presented in [5]. A new bifilar converter which uses resonant circuit during discharge period is proposed in [6]. Switched reluctance motor with higher rotor poles than stator poles is studied in [7]. More infonnation about SRM and SRG can be found in [8]. Fig. 1 shows the cross section of a 6 by 4 switched reluctance machine which can operate in motoring or generating mode of operation depend on the switching firing angles. This machine utilizes bifilar winding. In aligned position, when the rotor and stator poles are in full alignment, the phase inductance has the maximum value. As rotor rotates, the rotor poles reach exactly between the two adjacent stator poles; at this position the phase inductance has the minimum value. So the phase inductance profile is similar to Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002354_carpi.2014.7030038-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002354_carpi.2014.7030038-Figure4-1.png", + "caption": "Figure 4. Rotary base and the electrical winch", + "texts": [ + " the segment between the steel wire and the bucket, is tied by a dielectric material rope. All these procedures assure the electric insulation of the electrician. A hydraulic piston, attached to the column, provides the inclination motion of the column. This includes the motion of setting the column to the working position and resting the column into transportation position. The mechanism is shown in Fig.3. A hydraulic motor rotates the column through a pinion coupled to an internally toothed wheel (Fig.4). A mini hydraulic unit supplies the hydraulic power for both hydraulic actuators (Fig.4). This unit is driven by an electric motor, thus the Elevator can operate using batteries even when the vehicle engine is turned off. The operation of the Elevator IV is monitored by a sort of sensors and controlled by a Programmable Logic Controller (PLC). Fig.5 shows the dashboard of the Elevator, including buttons for commanding basic movements. The Elevator is operated using this panel in an emergency. Under normal conditions, the Elevator is operated using a remote control unit (Fig.6) that the electrician carries inside the bucket" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002293_icinfa.2013.6720420-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002293_icinfa.2013.6720420-Figure1-1.png", + "caption": "Fig. 1 Proposed DFUAV model", + "texts": [ + " Till now, control vanes are usually used in the majority of the ducted-fan UAV to maintain the attitude stabilization. Output torque of control vanes is nonlinear and commonly have effect on the rotary speed of the wing in order to supply enough lift. The variation of the wing's rotary speed will cause disturbance. This work presents a new structure which actuator system consists of four rotary cylinders which are symmetrically installed at bottom of inside duct. The ducted fan UAV is shown in Fig.1. The control torque of the actuator system is based on Magnus effect. The control of a small, autonomous ducted-fan aircraft presents many unique challenges [4]. The ducted fan UAV is unstable and susceptible to wind due to its external structural shape. Also the ducted fan UAV must change the attitude to gain speed, so it has different characteristics with respect to velocity. The literatures [5] presented a fuzzy gain scheduler to stabilize the DFUAV. The aerodynamic characteristics of ducted-fan during the transition maneuvers is highly unstable and therefore the use of conventional control techniques are inefficient", + " In the simulation, the boundary condition is that: the wind speed of the inlet is 2m/s; the two cylinders parallel to the y-axis is moving wall, the absolute rotary speed is 6000rpm and the direction of the angular speed points to the positive y-axis; the two cylinders parallel to the x-axis is stationary wall; the domain surrounds the wing is moving mesh and the rotary speed is 6000rpm; the wing is moving wall and he speed of relative to adjacent cell zone is 0; the angle of attack of the ducted-fan UAV is 10 degree. From the fig.5, it can be seen from the picture that the direction of the flow and plane of the four cylinders within the ducted-fan are orthogonal. Thus, the generated Magnus force can supply the control torque to stabilize the DFUAV. Fig.5 Streamline of the DFUAV B. Aerodynamic models of cylinder and propeller As shown in Fig. 1, the control device consists of four lightweight hollow cylinders and has been installed within the duct-A. Therefore, the adverse impacts coming from external wind gust can be avoided. Similarly, power system including the engine and the propeller are completely imbedded in duct B. The sectoral slot of duct-B acts as the airflow inducement and counterbalances the reaction of propeller. Here, Magnus-Forces 41 ~ ff satisfy the following equation: GVf j \u03c1= , 4~1=j (1) where \u03c1 , G , and V stand for the air density, the vertex strength, and the speed of downwash at duct exit plane respectively" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002240_sii.2011.6147563-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002240_sii.2011.6147563-Figure4-1.png", + "caption": "Fig. 4. Appearance of computer graphics to display simulation output", + "texts": [ + "4971 X 10-3, and {3z = -0.36624. Meanwhile, the details about the PID gains tuned experimentally are described in section V. D. Visualization using 3D computer graphics The numerical robot motion corresponding to the above mentioned dynamics calculation is visualized through 3D computer graphics. The software of the 3D computer graph ics is created using a useful c++ graphics library \"Open Inventor [14].\" An appearance of the whole NMLD system visualized through 3D computer graphics is shown in Fig. 4 (1). In addition, a view around the levitated object visualized in the same way is shown in Fig. 4 (2). In order to show the motion of the levitated object in an easy-to-understand way, the Z axis of the coordinate system described in section II is depicted as a red line, and an initial stable region for the microrobot is colored in light pink. Additionally an SI Intemational 2011 appearance of initial state and the appearances of typical 3- D motions (x-axis motion, y-axis motion and z-axis motion) are shown in Fig. 5 (1) '\" (4), respectively. E. Network system This simulator can be connected to network using socket communication, it aims at having the functions to confirm the manipulation performance by connecting directly to actual robot controller, remote operation device, and so on" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001914_s11740-014-0582-7-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001914_s11740-014-0582-7-Figure2-1.png", + "caption": "Fig. 2 Design of the planar drive", + "texts": [ + " the windings number differs. In this case, an induced voltage will arise, but it is low and will be compensated by current control. 3.1 Primary and secondary part To validate the functional principle of the cross winding technology a small demonstrator was designed. For manufacturing reasons, this experimental motor is based on a massive iron yoke instead of a lamination. The concentrated windings for each axis are stacked in two layers. The windings of the z-axis are closer located to the air gap (Fig. 2). Between both winding layers, a water cooling device is placed. The secondary part consist of a chessboard magnet array with 20 9 20 9 8 mm NdFeB magnets. The alternating flux direction for the magnets is shown with arrows in Fig. 2. To absorb lateral forces, the permanent magnets are engaged in pockets of 3.5 mm depth. The air gap between primary and secondary part is defined to 1 mm. The used materials are listed in Table 1. In Fig. 3 a sectional drawing of the planar drive is shown. The drive has six teeth and slots. The distance between two teeth is named as slot distance and is S = 19 mm. The active length of the primary part correlates to the number of slots multiplied with the slot distance. In this prototype, the active length is 114 mm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002480_2015-01-0680-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002480_2015-01-0680-Figure4-1.png", + "caption": "Figure 4. Mechanical structure of crank mechanism", + "texts": [ + " The compression force generated by the actuator is applied on the small end of the connecting rod; hence on each side of the piston wrist pin and pin bore there is only half of the compression load. The response time of the PZT (Piezoelectric Transducer) actuator is 100 ms (Figure. 3); thus the actuator needs 100 ms to increase to its maximum range and also needs 100 ms to return from the maximum range to the original position. The maximum force depends on the stiffness of the entire structure. In the bench test rig, the maximum force obtainable is 1.5-1.6 kN. A crank mechanism (Figure. 4 and 5) was designed in the bench test rig to simulate the swing motion of the connecting rod. The rotational motion is driven by an AC motor and the crank mechanism transfers the motor rotation into the swing motion of the connecting rod. The actuator response time is 100 ms and there is one peak load every two connecting rod rotation cycles, hence the rotation speed of the wheel is limited to 300 rpm to ensure the actuator runs at its maximum load. The angular velocity and angular position of the connecting rod are calculated versus the rotation angle of the wheel as shown in Figure 6" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003530_detc2011-48476-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003530_detc2011-48476-Figure3-1.png", + "caption": "Fig. 3. Leg i of the 3-CUP (3-PRUP) manipulator", + "texts": [ + " The legs of the parallel mechanism are equally distributed by an angular position given by the angle \u03c8i, which is a rotation about the Z-axis (i.e.\u03c81 = 0\u00b0, \u03c82 = 120\u00b0, and \u03c83 = 240\u00b0), and located in the XY plane by the vector ai (where a is the magnitude of ai and represents the distance from O to point Ai). Let the magnitude di of vector di represent the stroke of joint ji1, while its joint axis si1 is perpendicular to the fixed base. Joint axes si3, si4, and si5 are chosen to intersect at point Bi (forming a non-canonical S joint). As shown in Fig. 3, the location point Bi relative to Ai is denoted by the vector di.. Further, the P joints located on the platform are equally distributed by an angular position given by the angle \u03c7i which is a rotation about the W-axis (\u03c71 = 0\u00b0, \u03c72 =120\u00b0, \u03c73 = 240\u00b0), and their strokes are denoted by vectors bi with magnitude bi. Note that the axes of joints ji5 intersect at point C. The joint axis si4 is parallel to the V-axis, while the joint axis si3 is parallel to the XY plane and is perpendicular to si4. The 3-CUP parallel manipulator can be considered a noncanonical 3-PSP parallel mechanism, according to [19], where is proved that both architectures have the same mobility. This paper considers that P joints ji1 are active joints of this particular architecture. The following section presents a mobility analysis the 3-CUP parallel manipulator also called 3- PRUP mechanism. 3. MOBILITY ANALYSIS VIA SCREW THEORY A schematic drawing of the kinematic chain of the parallel mechanism under study is shown in Fig. 3. Note that the motion-screw set is made of five independent joint screws, corresponding to each joint axis sij all expressed in the frame G and relative to O 1 1 i i 0 $ s , 2 2 2 ( ) i i i i i s $ a d s , 3 3 3 ( ) i i i i i s $ a d s , 4 4 4 ( ) i i i i i s $ a d s , and 5 5 i i 0 $ s (1) where T c , s , 0 i ii a a a , T 0, 0, i i dd , T 1 0, 0, 1 i s , and T 5 , , x y zi r r r s (2) where rx, ry, and rz represent the components of the unit vector bi on the X, Y, and Z-axes" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003190_ijvsmt.2015.067522-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003190_ijvsmt.2015.067522-Figure5-1.png", + "caption": "Figure 5 Final Ilmor X3 model (see online version for colours)", + "texts": [ + " A rigid body formulation can only represent one revolute joint, whereas two joints impose ten constraints on one rigid body part with six DOFs. The refined geometry of the final model matched the dimensions of the Ilmor X3. The suspension properties of the model were tuned using the measurements of the real bike (LMS, 2014). The tyre models matched the new tyres that are used in the MotoGP. The dimensions and the radial stiffness were correct, and the tyre models could tolerate large camber angles. Figure 5 shows the Ilmor model with transparent fairing to display the engine components and more of the modelled geometry. Some components are shown as lines to visualise the suspension units and the tyre models. The gear connections and loads are not shown. The coefficients of the springs and dampers in the simulation model were determined to match those of the real bike suspension. The fork assembly is quite complicated and consists of several structures and gas and oil-based springs and dampers. The model captured the essential properties of the spring and damper in the fork as well as the overall length and end effects that limit the maximum stroke lengths" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001803_1.4031066-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001803_1.4031066-Figure2-1.png", + "caption": "Fig. 2 Definition of frames e; et ; ep ; and ef", + "texts": [ + " 137 / 061010-1Copyright VC 2015 by ASME Downloaded From: http://vibrationacoustics.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/journals/jvacek/934234/ on 02/08/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use added to the track around the axis of the base vector e2 in the inertia reference frame e h\u00f0t\u00de \u00bc H sin Xt (1) where H and X are the amplitude and the angular frequency of the input motion angle h, respectively. The floating reference frame ef which moves with the rotor except its spin motion is considered. Figure 2 represents the coordinate transformation from the inertia reference frame e to the floating reference frame ef . The rotor\u2019s orientation is determined by the input motion angle h, precession angle a, nutation angle d, and the spin angle c relative to the floating reference frame ef . The base vectors ef 1; ef 2; and ef 3 of the floating reference frame ef are used, and the angular velocity vector of the floating reference frame ef is represented as xf \u00bc xf 1ef 1 \u00fe xf 2ef 2 \u00fe xf 3ef 3. The angular velocity components xf 1, xf 2, and xf 3 around each axis and their derivatives are represented as xf 1 \u00bc _h sin a cos d _a sin d xf 2 \u00bc _h cos a\u00fe _d xf 3 \u00bc _a cos d\u00fe _h sin a sin d 9= ; (2) _xf 1 \u00bc \u20ach sin a cos d\u00fe _h _a cos a cos d _h _d sin a sin d \u20aca sin d _a _d cos d _xf 2 \u00bc \u20ach cos a _h _a sin a\u00fe \u20acd _xf 3 \u00bc \u20aca cos d _d _a sin d\u00fe \u20ach sin a sin d\u00fe _h _a cos a sin d \u00fe _h _d sin a cos d 9>>>>= >>>>; (3) By substituting them into the Euler\u2019s equation represented in the floating reference frame ef [7], the following set of equations of motion is derived [5]: I1 _xf 1 \u00fe \u20acc \u00bc 2 d dj jRaFb \u00fe caR2 t _a sin d ccR 2 a _c I2 _xf 2 \u00fe xf 3 I1 xf 1 \u00fe _c I2xf 1 \u00bc 2 d dj jRa sin d Rt cos d Fa I2 _xf 3 \u00fe xf 2 I2xf 1 I1 xf 1 \u00fe _c \u00bc 2Rt\u00f0Fb \u00fe Fc\u00de caR2 t _a cos d 9>>>= >>>; (4) where I1 and I2 are the polar and diametrical moment of inertia of the rotor, respectively", + " Figure 3 shows the contact model using equivalent spring and damper. Z \u00bc Rt tan d is the displacement of the rotor\u2019s tip to ep3 direction, and it is the value at the positive side of ef 1 direction. Z0 is the radius gap between the rotor and the track. The velocity vc of the rotor relative to the track at the contact point P in the tangential direction is represented as vc \u00bc d dj jRa _c\u00fe Rt _a (5) When the rotor is in the sliding condition on the track, velocity vc at the contact point P relative to the frame et, which is fixed to the track (refer Fig. 2), is not zero. Normal contact force between the rotor and the track Fa, tangential force of friction between the rotor and the track Fb, frictional force between the rotor\u2019s case and the track Fc, and the relative velocity vc at the contact point P change depending on whether the contact occurs at the upper side or lower side [5]. The displacement of the rotor\u2019s tip, Z, at the positive side of the direction of base vector ef 1 is used for the definition of the side of contact. When the contact occurs and Z< 0 (nutation angle is positive, d > 0), the rotor contacts with the lower side of the track and it is referred to as \u201clower side contact" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002309_icems.2011.6073725-Figure3-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002309_icems.2011.6073725-Figure3-1.png", + "caption": "Fig. 3. Diagram of the rotation of stator flux.", + "texts": [ + " TABLE I VOLTAGE SPACE VECTORS SV 0U 1U 2U 3U 4U 5U 6U 7U sasbsc 000 100 110 010 011 001 101 111 I IV III V II )010(3U (110)2U VI (100)1U)011(4U )001(5U )101(6U )000(0U )111(7U Fig. 2. Distribution diagram of both voltage space vectors and sectors. (6) can be rewrote as \u03b4\u03c8 sin 2 3 ssne ipT = (7) where \u03b4 is the angle between s\u03c8 and si . Because the stator flux magnitude s\u03c8 is kept nearly constant and the stator current magnitude si can\u2019t be suddenly changed, quick response of the electromagnetic torque only can be implemented by changing \u03b4 . In fact, the change of \u03b4 can be achieved by rotating the stator flux, as is shown in Fig. 3. From Fig. 3, the variation of the stator flux s\u03c8\u0394 can be derived by \u23aa\u23a9 \u23aa \u23a8 \u23a7 \u2212\u0394+=\u0394 \u2212\u0394+=\u0394 \u03b8\u03c8\u03b4\u03b8\u03c8\u03c8 \u03b8\u03c8\u03b4\u03b8\u03c8\u03c8 \u03b2 \u03b1 sin)sin( cos)cos( * * sss sss (8) where * s\u03c8 ,\u03b8 and \u03b4\u0394 is the reference of the stator flux magnitude, the phase angle of s\u03c8 and the variation of \u03b4 , respectively; sss j \u03b2\u03b1 \u03c8\u03c8\u03c8 \u0394+\u0394=\u0394 . By (4) and (8), the reference of the stator voltage refU is calculated by \u23a9 \u23a8 \u23a7 +\u0394= +\u0394= sssss sssss TiRu TiRu /)( /)( \u03b2\u03b2\u03b2 \u03b1\u03b1\u03b1 \u03c8 \u03c8 (9) \u23aa\u23a9 \u23aa \u23a8 \u23a7 \u2220= += ref ssref U uuU \u03b3 \u03b2\u03b1 22 )()( (10) where Ts is the switching period, \u03b3 is the phase angle of refU " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002772_amm.658.495-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002772_amm.658.495-Figure1-1.png", + "caption": "Fig. 1. Body position before throwing Fig. 2. Body position after throwing", + "texts": [], + "surrounding_texts": [ + "The analyses of the movement of the kinematic chain arm-forearm-hand was realized with the aid of the direct kinematic, by expressing the coordinates of the contact point between hand and ball, towards the axes system (x\u2019O\u2019y\u2019), attached to the shoulder joint (Figs. 1 and 2) The coordinates of the contact points between hand and ball, expressed as function of the flexion angle of the arm, are: \u03d5\u22c5+\u03d5\u22c5+\u03d5\u22c5=\u2032 \u03d5\u22c5+\u03d5\u22c5+\u03d5\u22c5=\u2032 332211 332211 sinqsinqsinqy cosqcosqcosqx , (1) where: q1, q2, q3 \u2013 are the length of the arm, the forearm and the hand, respectively, 1\u03d5 , 2\u03d5 , 3\u03d5 \u2013 are the flexion angles of the arm, forearm and hand respectively, When reporting the contact point coordinates between hand and ball towards the fix coordinated axes (xOy), represented in Fig. 3, one could write the relations: \u03d5\u22c5+\u03d5\u22c5+\u03d5\u22c5+= \u03d5\u22c5+\u03d5\u22c5+\u03d5\u22c5+\u2212= \u21d2 \u2032+= \u2032+\u2212= 3322111 332211 1 sinqsinqsinqhy cosqcosqcosqLx yhy xLx , (2) where: L \u2013 represents the throwing distance, By derivation with time the Eq. (2) equation system, one could obtain the initial speed of the ball: \u03d5\u22c5\u22c5\u03c9+\u03d5\u22c5\u22c5\u03c9+\u03d5\u22c5\u22c5\u03c9= \u03d5\u22c5\u22c5\u03c9\u2212\u03d5\u22c5\u22c5\u03c9\u2212\u03d5\u22c5\u22c5\u03c9\u2212= 333222111y0 333222111x0 cosqcosqcosqv sinqsinqsinqv (3) where: 1\u03c9 , 2\u03c9 , 3\u03c9 \u2013 are the angular speed of the arm rotation, the forearm and the hand, respectively. Using these components of the initial angular speed of the ball, the initial throwing angle could be determined, as follows: =\u03b1 \u2212 x0 y01 0 v v tg . (4) For the throwing successful, a series of conditions must be fulfilled [4, 16]: 1h > g2 v H 2 y0 \u2212 , (5) 1h > \u2212\u22c5\u22c5 \u22c5\u2212 \u2212\u22c5 \u2212 2 c 2 m 2 c 2 m2 ox2 c 2 m2 y0 D D 1g2 D D v D D 1v H , (6) a\u2212 < 2 DD pc \u2212 , when a < 0, (7) a\u2212 > 2 DD cp \u2212 , when a > 0, (8) where: Dm \u2013 ball diameter, Dc \u2013 basket diameter, Dp \u2013 the projection of the ball diameter on the basket diameter, having the incidence angle F\u03b1 , \u2212 + \u2212 \u2212\u2208 2 DD , 2 DD a pcpc , in correlation with Fig. 4 and Eqs. (7) and (8). The flexion-extension angles of the segments of upper limb may be determined using Eqs. (3), by imposing one of them and calculating the other two, through inverse kinematics, solving the system of two equations. At the same time, using the Eqs. (5), (6), (7) and (8), the initial data used for Eqs. (3) are checked, so that the success of throwing to be achieved." + ] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure3.2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure3.2-1.png", + "caption": "Fig. 3.2. Geometrical model", + "texts": [ + " The kinematic analysis of the proposed 6-URS platform is based on a modification of the algorithms of the 6-UPS platform, and has been presented on [21], so here we make a brief resume. 50 J.M. Sabater et al. The kinematic analysis of the platform has been carried out considering only 13 bodies, because the adding of the transmission links of the pantographs does not give additional information to the kinematic model and extends the vectorial equations. This assumption must be reviewed for a dynamical modelling. Fig. 3.2 shows the used model. Using the Euler parameters to represent the orientation, each body needs seven generalized coordinates, leading to 91 generalized coordinates for the 13 bodies to completely define the display. The sum of the constraints imposed by the spherical, the rotational and the universal joints, and the constraints imposed by the normalization of the Euler parameters, give a total of 85 constraints. The difference are the degrees of freedom of the mechanism. The motion of the device is defined by the time variation of angle C\u0302 between the even and odd links", + " Given the position and orientation of the \u201clink 1\u201d joystick, by a vector q1 = [r1,p1]T where r1 = [x1, y1, z1]T is the cartesian position and p1 = [e0, e1, e2, e3]T are the euler parameters, or q1 = [r1, \u03b11, \u03b21, \u03b31]T if the orientation is given with the 313 euler angles. The distance between the universal and spherical joints can easily be obtained by: rAnBn = r1 + A1s1 \u2032Bn \u2212 sBn 0 (3.4) where A1 is the rotation matrix given the orientation of \u201clink 1\u201d. Getting the norm of rAnBn, the solution is shown by (3.5) (see Fig. 3.2). C\u0302n = arccos (( BC )2 + ( AC )2 \u2212 norm (rAnBn)2 2 ( BC ) ( AC ) ) (3.5) Forward Kinematics To solve the forward kinematics (FKP) of a URS platform is to establish the relations between the command variables of the angles C\u0302 and the final position of the end effector. A solution to the FKP problem is obtained using multibody formulation in [21], but it is not needed here. In order to obtain the forces rendered to the operator, the relation between the joint forces exerted by the actuators and the cartesian force rendered at the end-effector (\u201clink1\u201d) must be known" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003420_j.mspro.2014.06.203-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003420_j.mspro.2014.06.203-Figure1-1.png", + "caption": "Fig. 1. Calcuation scheme of the model.", + "texts": [ + " It is realized for the cases of contact interaction of rolling and fretting fatigue. In particular peculiarities of formation of such typical defects as pits and spalls in rolling bodies and growth of edge cracks in the elements of fretting couples under conditions of sliding/sticking between them are investigated. Examples of assessing the life time by damages formation in the contact region are presented. Consider the case of contact interaction of two bodies one of which is damaged by cracks. This body is modeled by a half plane weakened by a system of cracks (Fig. 1). The second body (counterbody) is modeled in the form of normal p(x, , t) and tangential q(x, f, , t) forces distributed over the edge of the half plane (t is a time). The friction forces are taken into account in the form of tangential forces q(x, f, , t). The simplest possible representation of the distribution of tangential forces is given by their relationship with the normal forces specified by the AmontonCoulomb law q(x) = fp(x) under the condition of complete sliding of bodies, where f is the coefficient of friction between the bodies", + " Within the frames of linear fracture mechanics the SIFs KI and KII are determined by the known relations of -criterion and -criterion in terms of SIF KI and KII and also the angle * of crack inclination angle at its tip from the tangent. Later, in the case of rolling contact interaction, the counterbody action is modeled by a unidirected repeated movement of the Hertzian normal forces with tangent component along the semi-plane edge and the main parameter that forms a contact cycle is parameter = x0/a (Fig. 1). Under contact interaction of the fretting fatigue a cycle is formed by the same distributed and oppositely directed tangent forces q+(x, f, ) and q\u2013(x, f, ). In the conditions of complete counterbody sliding xfpfxq ,, , = const. (2) Following the widely-accepted principles of fatigue fracture mechanics let us assume that fatigue macrocrack propagates (by this or that mechanism) under such conditions: thmm KlK \u0430\u0431\u043e fcmm KlK (m = I, II). (3) When the first condition (3) is realized (where KIth and KIIth are ranges of fatigue crack growth fracture thresholds by mode I and mode II mechanisms) the crack grows stably; under the second condition (3) (where KIfc and KIIfc are critical SIF ranges KI and KII) the crack grow spontaneously", + " Let us note here that when relations (3) are equalities we can find from the first one the initial lengths of marocrack l0 and l0 and from the second one \u2013 its critical lengths lc and lc for relations (1). The path of the macrocrack growth is constructed by using a step-by-step procedure based on the algorithms proposed in (Panasyuk et al (1995)). To construct the path, we introduce two types of steps, namely, main steps connected with crack growth and auxiliary steps connected with variations of the load in contact cycles. In each stage of the path construction, the increment of crack path h is plotted from the crack tip in the direction specified by the angle = * (Fig. 1). The auxiliary step is used to find the extrema and to determine the range of the parameter Km in the contact cycles. In each stage of the path construction, the SIF KI and KII are determined from the solution of the singular integral equations of the static (in the general case, contact) problem of the theory of elasticity for a half plane containing curvilinear crack of different configurations. Each increment of the path is approximated by a third-degree polynomial. In constructing the crack growth paths for a system of cracks, i", + " In general contact life time together with two stages of crack growth which forms spalling/notching or starts to grow spontaneously deep into the material is calculated for many configurations of service parameters (p0, f, fc, ) with account of fatigue crack growth resistance characteristics of rail and roll steels. Within the framework of the model proposed in the present work, the process of crack propagation in bodies operating in contact under the conditions of fretting fatigue is studied. The body in contact weakened by cracks is modeled by an elastic half plane with cracks (see Fig. 1) and the action of the counterbody is modeled by forces distributed over the contact region. As the normal component of the contact load p(x, , t) = p(x, ), the Hertz pressure (relation (4)) is used. Tangent component in case of complete sliding of the fretting couple is modeled by forces (2). To model the action of the counterbody in the case when a stick region is in the contact zone of the fretting couple elements (Fig. 5a), the distribution of contact pressure is used established independently by Cattaneo and Mindlin (see Datsyshyn (2005)): ,, ;,, 0 2 0 22 0 2 0 00 2 0 2 0 cxxxxcxxaapf axxccxxaxxaapf xq y = 0, (6) where the relative length of the region of stick is specified by the ratio c/a; Q and \u0420 are the tangential and the normal components of the external load vector" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000833_978-3-540-77974-2_13-Figure13.11-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000833_978-3-540-77974-2_13-Figure13.11-1.png", + "caption": "Figure 13.11 Enlarging the robot", + "texts": [ + " Therefore we use the following trick. We make the robot slightly larger, and use the method described above on the enlarged robot R\u2032. This is done in such a way that although R\u2032 can collide during rotations, the original robot R302 Section 13.6 NOTES AND COMMENTS cannot. To achieve this, the robot is enlarged as follows. Rotate R clockwise and counterclockwise over an angle of (180/z)\u25e6. During this rotation R sweeps a part of the plane. We use for the enlarged robot R\u2032 a convex polygon that contains the sweep area\u2014see Figure 13.11. We now compute the trapezoidal maps and the road map for R\u2032 instead of R. It is not difficult to prove that R cannot collide with an obstacle during a purely rotational motion between two adjacent slices, even though R\u2032 can. By enlarging the robot we have introduced another way to incorrectly decide that there is no path. Again, this becomes less likely when the number of slices increases. So with a large number of slices, the method probably performs reasonably well in practical situations. The motion-planning problem has received a lot of attention over the years, both from people working in computational geometry and from people working in robotics, and this chapter only scratches the surface of all the research" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003963_detc2013-12427-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003963_detc2013-12427-Figure6-1.png", + "caption": "FIGURE 6. REGIONS OF HARDENING (SHADED, ABOVE THE CURVES) AND SOFTENING BEHAVIOR (BELOW THE CURVES) FOR m = 2,3,4,5,6.", + "texts": [ + " For nonlinearly elastic rings, the nonlinearity of the modes dictated by \u03931 may switch from softening to hardening depending on the constitutive parameters. The switching from softening to hardening behavior takes place when \u03931(M o 2 ,M o 3 ) changes sign. The locus of constitutive parameters for which \u03931(M o 2 ,M o 3 ) = 0 signals the switching condition and separates regions where the ring exhibits a softening behavior from those where the ring exhibits a hardening behavior. These loci are represented in Fig. 6 for various modes (m= 2,3,4,5,6) by the solid curves in the plane of the constitutive nondimensional parameters M o 2 and M o 3 . According to these curves, the sets of constitutive parame- 6 Copyright c\u20dd 2013 by ASME Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use ters for which the behavior is hardening can be determined. As an example, Fig. 7 shows the hardening backbone surface of the lowest mode when M o 2 = 0 and M o 3 = 2\u00d7102" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001446_icuas.2015.7152381-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001446_icuas.2015.7152381-Figure2-1.png", + "caption": "Fig. 2. Flat turn", + "texts": [ + " Hence, the equations set in (4) can be rewritten as follows \u03b2\u0307 = g V\u0304 \u03b3r + T\u0304 s\u03b2c\u03b1\u0304 mV\u0304 + p ( s\u03b1\u0304 + q0 mV\u0304 CYp ) + r ( q0 mV\u0304 CYr \u2212 c\u03b1\u0304 ) + CY\u03b2 q0 mV\u0304 \u03b2 + CY\u03b4r q0 mV\u0304 \u03b4r p\u0307 = Izzq0b ( CL\u0304\u03b2 \u03b2 + CL\u0304p c1p+ CL\u0304r c1r + CL\u0304\u03b4a \u03b4a ) IxxIzz \u2212 I2xz + Ixzq0b ( CN\u03b2 \u03b2 + CNp c1p+ CNr c1r + CN\u03b4r \u03b4r ) IxxIzz \u2212 I2xz + Ixzq0bCN\u03b4a \u03b4a IxxIzz \u2212 I2xz r\u0307 = Ixzq0b ( CL\u0304\u03b2 \u03b2 + CL\u0304p c1p+ CL\u0304r c1r + CL\u0304\u03b4a \u03b4a ) IxxIzz \u2212 I2xz + Ixxq0b ( CN\u03b2 \u03b2 + CNp c1p+ CNr c1r + CN\u03b4r \u03b4r ) IxxIzz \u2212 I2xz + Ixxq0bCN\u03b4a \u03b4a IxxIzz \u2212 I2xz (6) where q0 = 1 2 \u03c1V 2S The first step is to design the lateral-directional attitude control, thus we defined the attitude signal errors as follow \u03c6e = \u03c6d \u2212 \u03c6 pe = pd \u2212 p \u03c8e = \u03c8d \u2212 \u03c8 re1 = rd \u2212 r in order to drive the rotational dynamics to a desired point a PD controller is established \u03b4a1 = k\u03c6\u03c6e + kppe \u03b4r1 = kr1\u03c8e + kr2re1 (7) with k\u03c6, kp as gains for roll controller and kr1 , kr2 as gain of yaw controller. The equations (7) are the PD controllers for \u03c6 and \u03c8 lateral-directional attitude angles and they are controlling by ailerons and rudder respectively. For a fixed wing aircraft, there are three ways to perform a turn on the Cartesian plane (x,y): flat turn, bank turn, and coordinated turn [11]. A. Flat Turn A flat turn maneuver is preformed by the action of rudder input like in a boat. For instance, to describe a circle of radius R, (see Fig. 2), it is not possible to assume that the instantaneous direction of flight coincides with the longitudinal axis of the airplane, thus an sideslip angle \u03b2 exists [11]. This way of turning is not commonly used since the sideslip angle is not desirable. In addition, the turning radius R is big, and it takes a lot of time to make the turn. B. Banked Turn One of the many discoveries made by the Wright brothers is that a fixed wing aircraft is best turned by rolling. Rolling means to perform the turn with ailerons (banked turn) instead of rudder [1]" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001985_1350650114540625-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001985_1350650114540625-Figure2-1.png", + "caption": "Figure 2. Contact model.", + "texts": [ + " As many engineering surfaces possess fractal characteristics, the rough surface used in this paper is generated based on the two-variable Weierstrass\u2013 Mandelbrot (W-M) function.11,12 Figure 1 shows the solid model with a fractal rough surface. In this model, the fractal dimension of the surface profile (D) is equal to 2.5, the fractal roughness parameter (G) is equal to 3.0 10 6m, the density of frequencies in the surface profile ( ) is 1.5,12 the sample length is 456 10 6m, and the nominal area (A0) is equal to 8.12 10 8m2. Figure 2 gives the elasto-plastic sliding model between a fractal rough solid (A) and a flat solid (B). In Figure 2, A1, B1 represent the friction surfaces of solid A and B, respectively, and A2, B2 are the back surface of the friction surface of solid A and B, respectively, A3, A4, A5, A6, B3, B4, B5, B6 are the side surfaces of solid A and B, respectively, e1, e2 are the thickness of solid A and B, respectively, l, L are the length of solid A and B, respectively. To solve the frictional heat flux coupling, it is assumed that on the micro-contact point the local instantaneous temperatures of solid A (TA) and solid B (TB) are equal and the local micro-contact interfaces are a series of very small ideal planes", + " Considering the complicated geometry of the fractal surface, the tetrahedral element C3D4T is chosen to mesh the fractal solid and has 24,992 elements. Comparatively, the plane is easy to mesh, so hexahedral element C3D8T is used and has 10,731 elements. A finer mesh is built near the fractal surface of solid A and the working surface of solid B to obtain accurate solutions for stress/temperature fields, as shown in Figure 3. These two elements are displacement and temperature coupling elements. For the boundary conditions, as shown in Figure 2, the applied pressure Pz exerts uniformly on the back surface (A2) of solid A along z direction. The displacements of two side surfaces (A3 and A4) of solid A along x direction are equal to zero and the displacements of other two side surfaces (A5 and A6) of solid A along y direction are equal to zero. The whole solid B moves along the negative x direction, and the displacement of solid B along y direction is equal to zero. The displacement of the surface B2 of solid B is immobile along z direction" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001866_ecce.2015.7310233-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001866_ecce.2015.7310233-Figure9-1.png", + "caption": "Fig. 9. Space vectors generated by the Conventional 2.", + "texts": [ + " This occurs because the phase currents of the both proposed configurations are sinusoidal, while that the conventional 1 generates a waveform of current phase are asymmetric, as illustrated in Fig. 8. When confronted to the conventional 2, the configurations 1 and 2 have a THD reduction of approximately 42%. This decrease occurs due to the proposed topologies synthesize the reference voltage V \u2217 using voltage vectors closer than the conventional 2, since this structure generates only four voltage vectors that are 90 degrees lagged from each other, as illustrated in Fig. 9. Consequently, there is a reduction in the ripple of the input current in the proposed topologies, which results in improved THD levels. In this section, a comparative analysis is made between the proposed and conventional topologies according to the conduction and switching losses in the semiconductor devices. Table VI presents the values of power loss by conduction (Pcond), switching (Pswit) and total losses (Ptot = Pcond + Pswit). The values of power loss were obtained to the following conditions: input RMS phase voltage equal to 120 V, a twophase PM machine of 1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001489_iecon.2014.7049305-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001489_iecon.2014.7049305-Figure2-1.png", + "caption": "Fig. 2. Model of two-wheel wheelchair", + "texts": [ + " Finally the proposed method is valid through experiments. Contents of this paper are as follows. In section II, the dynamic model of the two-wheel wheelchair is introduced. In section III, the control system for the wheelchair is represented. In section IV, the experimental results are described. Finally, in section V, conclusion and future works are described. Kinematics and dynamics of the two-wheel wheelchair are described in this section. The modeling of the two-wheel wheelchair is shown in Fig. 2(a) and (b). Fig. 2(a) shows top view and Fig. 2(b) shows side view. P which is the middle of the axle, is the reference point of position. G is CoG point of position. In this model, inclination angle of the body about the wheel-axis (pitch angle) is represented as \u03b8p. Parameters are shown in Table I. Vectors are determined as eq.(1) and eq.(2). \u03b8 = [ \u03b8p \u03b8r \u03b8l ]T (1) x = [ \u03b8p d \u03c6 ]T (2) \u03b8 and x denote the variables of joint space and translational/rotational space respectively. Two-wheel wheelchair cannot move lateral direction. The constraint condition is represented as eq" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003082_cca.2013.6662890-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003082_cca.2013.6662890-Figure2-1.png", + "caption": "Fig. 2. Unique trajectory for a given exit state", + "texts": [], + "surrounding_texts": [ + "In this section we find the longest path for a single UAV inside the ellipse. First, we fix the exit point, and find the longest trajectory for a single UAV with a given exit point. Then, we set the exit point free, and obtain the longest path for a UAV for all possible exit points. In this process, we consider only the exit points in the first quadrant. Since the common surveillance region has two axes of symmetry, the results for any other point can be obtained by correlating with the analysis of these points." + ] + }, + { + "image_filename": "designv11_84_0003721_detc2014-34297-Figure10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003721_detc2014-34297-Figure10-1.png", + "caption": "FIGURE 10. Parallel platform where the mobile platform generates a Scho\u0308nflies subalgebra, xu\u03021 , with respect to the fixed platform.", + "texts": [ + " Leg 3 generates the Scho\u0308nflies subalgebra in the direction of u\u03021, corresponding to case 10, Table 2, where the first three kinematic pairs generate a subalgebra t\u22a5\u2217u\u03024 , while the last two kinematic pairs generate just a subspace, S1,u\u03021 , where S1,u\u03021 < gu\u03021 . Leg 4 generates the Scho\u0308nflies subalgebra in the direction of u\u03021, corresponding to case 1, Table 2. Moreover, the condition for forming platform 3 is also used, that is, the directions of the unit vectors u\u03021 coincide, in legs 2 and 4, except when the subspace S1,u\u03021 generates a subalgebra t\u22a5u\u03021 or t\u22a5\u2217u\u03021 , in which cases, the directions of the unit vectors u\u03022, must coincide with the directions of the unit vectors u\u03021, in legs 1 and 3. 2. Consider the parallel platform shown in Figure 10. Leg 1 generates the Scho\u0308nflies subalgebra in the direction of u\u03021, corresponding to case 4, Table 2, where the first two kinematic pairs generate just one subspace, S1,u\u03021 , such that S1,u\u03021 < gu\u03021 , while the last three kinematic pairs generate the subalgebra t\u22a5u\u03022 . Leg 2 generates the Schnflies subalgebra in the direction of u\u03021, corresponding to case 1, Table 2. Leg 3 generates the Scho\u0308nflies subalgebra in the direction of u\u03021, corresponding to case 10, Table 2, where the kinematic pairs one, two and four generate the subalgebra t\u22a5\u2217u\u03023 , while the kinematic pairs three and five generate just one subspace, S2,u\u03021 , such that S2,u\u03021 < gu\u03021 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003265_isccsp.2014.6877889-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003265_isccsp.2014.6877889-Figure2-1.png", + "caption": "Fig. 2 shows an electrohydraulic active suspension of a quarter-car model. The suspension is composed of the hydraulic system that plays the active part of it, while the spring-damper system constitutes the classic passive suspension.", + "texts": [], + "surrounding_texts": [ + "The servo-valve used in this application, is a double stage one with matched and symmetric orifices. Its dynamic equation, referring to [10], is given by: KuAA vvv =+\u03c4 (7) where u is the control input, K is the servo-valve constant, v\u03c4 is its time constant and Av is the orifice opening area. The flow rate from and to the servo-valve, assuming small leakage, is given as: \u03c1 LS vd PP ACQQ \u2212 == 21 (8) where PL is the differential of pressure due to load, 21 PPPL \u2212= in the positive direction of motion and 12 PPPL \u2212= in the negative direction of motion. Finally, 21 PPPS += is the source pressure, Cd is the flow discharge coefficient and \u03c1 is the fluid oil density. Thus we give the compressibility equation: \u2212\u2212\u2212 \u2212 \u2212\u2212 = )( )( )( 2 222 0 0 xxALP AsigmPP AC xxAV V P sL vLs vd s L \u03c1 \u03b2 (9) V0 is the oil volume in one chamber of the actuator when the piston is at the center position, )(0 xxAV s \u2212\u2212 is the volume of the actuator chamber at every piston position, L is the flow leakage coefficient, and \u03b2 is the oil bulk modulus. Note that in equation (9), as in [11], the sigmoid function \u201csigm(Av)\u201d is an approximation of the non-differentiable sign function \u201csign(Av)\u201d. The sigmoid function has the following properties: 0 ; 1 1 )( > + \u2212 = \u2212 \u2212 a e e Asigm v v aA aA v (10) This is a continuously differentiable function with: \u221e\u2192\u2212 \u2192 \u221e\u2192 = - ax if 1 0 ax if 0 + ax if 1 )(xsigm (11) and: 2)1( 2)( ax ax e ae dx xdsigm \u2212 \u2212 + = (12) Now, considering the hydraulic actuator equation of motion given by Newton\u2019s second law and referring to Figs. 2 and 3, we can write: )( )()()( L r r r r r r r s r p s r p P M A xx M B xx M K xx M B xx M K x \u2212\u2212\u2212 \u2212\u2212\u2212+\u2212= (13)" + ] + }, + { + "image_filename": "designv11_84_0003499_ehb.2013.6707308-FigureI-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003499_ehb.2013.6707308-FigureI-1.png", + "caption": "Fig. I. The scheme of the electro-magnetic actuated vibrating platform", + "texts": [], + "surrounding_texts": [ + "The service will be delivered from a central location in a private cloud model. The access to the platform will be done after the user registration and authentication via a module that will grant access through a username and a password from any device. These credentials can exist either temporary or penn anent, depending on the user's role. The access system is based on roles so, user credentials decides which applications are accessible and where following and adding content, comments or notes are allowed. Accounts expire when the user leaves the assigned project group or on explicit request of the project manager. The allowed actions will include content creation, post sharing and editing, forums access, video and audio collaboration, shared workspaces. The user management system will be centrally managed to ensure a common policy. A valid user account will be usable from any workstation and will grant the same rights to its owner. Based on the considerable costs of such a project, the decision of the interdisciplinary team was to split it into a short-term and a mid-term project. The short term part will be covered by the local hardware infrastructure and the server and the IT-administration of one of the universities (most likely the Polytechnic), by extending the activities of the Center of Innovation and Technology Transfer for Mechatronics, in form of a pilot project with a duration of one year and features and functional requirements are going to be stepwise implemented. Meanwhile, some fundraising actions are planned, especially the trial of another concept of the open source society: crowd funding. Although the research community is skeptical about venture financing generally and of crowdfunding for scientific project particularly and as in Romania the first crowdfunding platform (Multifinantare) was established, it is one of the ways the Center is ready to take into consideration, among the other funding applications and sources. For sustainability reasons, on medium-term, the platform will be remotely implemented on a server of an outsource provider. III. PROJECT EXAMPLES As mentioned, there has been already a collaborative research among the engineering and the medical team. It is a work in progress to implement parts of the project management communication on the VRC: one project is the electro-magnetic actuated vibrating platform, the other one is the customized acetabular cage. 1. Electro-magnetic actuated vibrating platform In vibration therapy, mechanical vibrations are induced in the body while the person sits upright on a high frequency vibrating platform. Many meta-analyzes have shown growing interest in this type of therapy, not only to fortify muscles, for the prevention and treatment of bone decalcification (osteoporosis), but also for patients suffering from certain chronic diseases [5]. As in [6], the vibrations of the platfonn are specific for each patient because the frequency and amplitude of the oscillations are directly related to the person's weight. 2. Customized acetabular cage Hip replacement is today a common and highly successful surgical procedure. However, in the case of revision arthroplasty the procedure is more complicated because standard devices have to be adapted to different shapes of acetabular defects, which are particular to the patient. Such problems as severe acetabular defects, caused by tumours, protrusion and pelvic discontinuity can be solved with a Muller Ring or as in Fig. 3, a Burch-Schneider Anti protrusio Cage (APC). As in [7], the purpose of the acetabular components is providing a painless and durable construct maintaining the bone stock and restoring the hip mechanics by restoring the true centre of the hip. The proposed patient specific acetabular cage design improves the standard cage solution by solving problems that standard cages sometimes do not fix, e.g. insufficient fixation due to the fact that the inferior flange did not engage the ischial bone, problems of bone graft resorption through a better adaptation to patient and tries to remain an affordable solution. As in Fig. 3, the cage adapted to patient for specific acetabular defects is similar to a normal Burch-Schneider APC in the sense that the iliac flange and the ischial flange are present; the cage cup is also present. As in the case of the Burch-Schneider cage, the cup insert is not geometrically fixed to the metal structure and allows the surgeon to reorient the PE inlay if necessary. Furthermore the rim of the cage is designed so that it allows maximum movement even if the PE inlay is reoriented. For this reason the cage design tries to keep all the advantages of the Burch-Schneider APC and add some new benefits: improved fit to the healthy bone due to particularization of the contact surface ensuring good initial fixation, screws can be easily and correctly positioned thanks to the drill guide, which ensures no accidents with screws that exit outside of the bone volume and a correct orientation. Being custom fit to the bone allows a good restoration of the anatomical centre of rotation. Where necessary, the structure of the cage can be thicker than standard acetabular cages to support the bone weakness. As in [10], the subsequent steps for a customized acetabular cage are: using computed tomography (CT) to obtain patient specific information; building the 3D pelvis model using the scanned CT data of the prosthesis, agreeing on design medical technical parameters with the surgeon regarding established objective: hip centre of rotation, desired anteversion and inclination angle, surface to be covered by the cage; fabricating the cage through rapid prototyping in titanium and drilling guides in plastic, testing the prosthesis on a plastic model, sterilizing it and finally inserting it into the patient. IV. CONCLUSION Great achievements consist of small successful steps, of a significant time investment for research and gathering of ideas. The human touch is as necessary as funds, because only in a collaborative environment, exchange of ideas get permanent and consistent. Well-structured knowledge clusters have better chances to progress and aggregate some results. Inputs from the potential producing company are also taken into consideration during research and tests on the functional model. On a small scale, the value chain is simplified by the virtual platform, because it saves time and improves the information and experience exchange among specialists, engineers, physicians and marketing managers, which have otherwise no or very few opportunities to come together and discuss all the aspects of innovation for biomechanic and biomechatronic products. Another advantage of the virtual platform is that it stimulates the involvement of the PhD. students both from polytechnic and medical universities along all the other channels of technology transfer." + ] + }, + { + "image_filename": "designv11_84_0002754_icelmach.2014.6960214-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002754_icelmach.2014.6960214-Figure1-1.png", + "caption": "Fig. 1. Constitution of the hub dynamo.", + "texts": [ + " In order to increase the efficiency and decrease the size of the dynamo, this paper indicated magnetic-geared generator (MGG) [8-16], which is reported to have high generation efficiency and be effective for downsizing [17]. This paper proposes a new generator, which we call the claw pole type magnetic-geared generator (CP-MGG), which combines the magnetic gear with the claw pole type generator. The proposed CP-MGG realized aim performance to satisfy JIS despite smaller size of conventional hub dynamo. Its performance is evaluated using 3-D finite element analysis. The hub dynamo is attached to the axis of the front wheel and also the tire spokes which are connected to the tire as shown in Fig. 1. When the tire turns, the hub case turns along with the tire at the same speed. H. Ukaji, K. Hirata, and N. Niguchi are with the Department of Adaptive Machine Systems, Graduate School of Engineering, Osaka University, Yamadaoka, Suita, Osaka, 565-0871 Japan (e-mail:hajime.ukaji@ams.eng.osaka-u.ac.jp). The distance between the spoke of right and left is given at 53.6mm. B. Voltage regulation The hub dynamo is a single-phase alternating current generator. JIS standards require that hub dynamos satisfy the requirements shown in Table I, where a 15\u03a9 resistor is connected to the coil of the hub dynamos as an electrical load, and 120 rpm is equivalent to a speed of 15 km/h for 26- inch bicycles" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure1.42-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure1.42-1.png", + "caption": "Fig. 1.42 Implementation of the chamfer and rounding tools on our part", + "texts": [ + " The support board was made in a PRT file, generated by Creo Parametric, with different tools such as sketching, extrusion, rounding, plane, and chamfer. The plate was started with a regular, rectangular shape, as shown in the outline of the plate in Fig. 1.40. 1 Comparative Study of Interferometry and Finite Element Analysis \u2026 27 Once the sketch was created, an extrusion of 4 mm was performed, to generate a solid part as shown in Fig. 1.41. Next, the chamfer tools were used twice and round once over the part for more ergonomics and anatomical functionality. Figure 1.42 shows the strokes. The next step was the creation of the holes shown in Fig. 1.43, these are useful as the anchor areas to both the bones and muscles. For the creation of these holes, the sketching and extrude tools were used. By the top line, the circles are at 6.25 mm; while on the bottom line are separated 6mmbetween each center, andwith a diameter 28 J. A. Beltr\u00e1n-Fern\u00e1ndez et al. of 2.5 mm all circles, as shown in Fig. 1.43. Finally, these circles were extruded for 4 mm. Next, it was to create an elliptic shape to the top of the part with an elliptical extrusion, as shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003787_j.proeng.2015.01.465-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003787_j.proeng.2015.01.465-Figure2-1.png", + "caption": "Fig. 2. Longitudinal section of the valve.", + "texts": [ + " In the first part we will consider the influence on transfer coefficient kv, in the second \u2013 on parameters of transfer function of a mechanical part. Also there will be an assumption that processes in electromechanical part are so fast that could be neglect. If it is necessary to consider processes take place in electromagnetic part, this could be made using well-known electric theory. As an example was taken proportional spool valve Festo MYPE\u20135\u20131/8LF\u2013010\u2013B. Schematic circuit of the valve is in fig. 1, on fig. 2 \u2013 longitudinal section, where 1 \u2013 body, 2 \u2013 spool, 3 \u2013 electronic block. In this part is considered an influence of gas-dynamic forces on transfer coefficient of volume flow rate, which one can write: Gas volume in pneumosystems strongly depend on pressure and temperature. For this reason it is more correct to obtain relations for mass flow rate, and then return to volume flow rate with following formula: (2) where G \u2013 mass flow rate; \u03c1 \u2013 density, R \u2013 universal gas constant, \u03b8 \u2013 absolute temperature, p \u2013 pressure" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001133_978-1-4471-5102-9_181-1-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001133_978-1-4471-5102-9_181-1-Figure1-1.png", + "caption": "Fig. 1 SARbot (Courtesy of SeaBotix Inc.) http://www.seabotix.com/products/sarbot.htm", + "texts": [ + " Underwater robots (ROV, remotely operated vehicle) are deployed to responder organizations in preparation for water damage such as tsunami, flood, cataract, and accidents in the sea and rivers. They are equipped with cameras and sonars and remotely controlled by crews via tether from land or shipboard within several tens of meters area for victim search and damage investigation. At Great Eastern Japan Earthquake in 2011, Self Defense Force and volunteers of International Rescue System Institute (IRS) and Center for Robot-Assisted Search and Rescue (CRASAR) used various types of ROVs such as SARbot shown in Fig. 1 for victim search and debris investigation in port. E-mail: tadokoro@rm.is.tohoku.ac.jp Page 1 of 7 In order to reduce risk of monitoring and recovery construction at volcano eruption, application of robotics and remote systems is highly expected. Various types of UAVs (unmanned aerial vehicles) such as small-sized robot helicopters and airplanes have been used for this purpose. Unmanned construction system consists of teleoperated robot backhoes, trucks, and bulldozers with wireless relaying cars and camera vehicles as shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003538_amm.325-326.870-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003538_amm.325-326.870-Figure5-1.png", + "caption": "Fig. 5. Determining the local coordinate system", + "texts": [ + " The parametrical equations of the face gear\u2019s flanks are: = \u22c5\u22c5\u2212\u22c5\u2212= \u22c5\u22c5\u2212\u22c5= \u03a3 vvuz tgvuvuy tgvuvux aa aa st ),( cossin),( sincos),( :)( 1 1 1 \u03c8\u03b1\u03c8 \u03c8\u03b1\u03c8 (1) = \u22c5\u22c5+\u22c5= \u22c5\u22c5\u2212\u22c5= \u03a3 vvuz tgvuvuy tgvuvux aa aa dr ),( cossin),( sincos),( :)( 1 1 1 \u03c8\u03b1\u03c8 \u03c8\u03b1\u03c8 (2) The used parameters are: \u03c8a the suitable angle of the quarter of the pitch, \u03b1 the pressure angle, u and v are liear parameters. In figure 3 is presented the relative movements between the face gear and the bevel gear, [6]. After the transformations, according to figure 5, the transfer matrix is obtained: \u22c5+\u22c5\u22c5\u2212= \u22c5+\u22c5\u22c5= \u22c5+\u22c5\u22c5= \u22c5+\u22c5\u22c5\u2212= \u22c5 \u22c5\u2212 \u2212\u22c5\u2212\u22c5 = 12122,3 12121,3 12122,2 12121,2 22,31,3 22,21,2 11 cossinsinsincos sinsincossincos coscossinsinsin sincoscossinsin coscos cossin sinsincoscoscos \u03d5\u03d5\u03d5\u03b4\u03d5 \u03d5\u03d5\u03d5\u03b4\u03d5 \u03d5\u03d5\u03d5\u03b4\u03d5 \u03d5\u03d5\u03d5\u03b4\u03d5 \u03b4\u03d5 \u03b4\u03d5 \u03b4\u03d5\u03b4\u03d5\u03b4 m m m m mm mmM T (3) Therefore the parametrical equations are: \u22c5= 1 ),( ),( ),( 1 ),,,( ),,,( ),,,( 1 1 1 212 212 212 vuZ vuY vuX M vuZ vuY vuX T\u03d5\u03d5 \u03d5\u03d5 \u03d5\u03d5 (4) Expression (4) represents the equations of the surface family in the bevel gears system, which envelope curve represents the tooth flanks of the generated gear", + " The bevel gear has the following initial parameters: \u2022 Modul: m = 4 mm; \u2022 Half angle of the pitch cone: \u03b4 = 37,875 o ; \u2022 Number of teeth: z = 28; \u2022 Pressure angle: \u03b1 = 20 o ; \u2022 Specific high of the addendum: ho = 1 mm; \u2022 Specific pasttime of the dedendum: co = 0,25 mm; \u2022 Teeth length: b = 20 mm. Using the coordinate measuring machine, type MITUTOYO Crista Plus, the experimental data was determined. Coordinate measuring machines have their own coordinate system. To simplify the post-processing of the experimental data, a local coordinate system was determined, fixed on the work piece (Figure 5). The coordinate points measured on the bevel gear\u2019s flanks were detected with a RENISHAW touching system. The measurement settings are: \u2022 A = 0 o , B = 0 o (the touching probe orientation); \u2022 The scanning direction: from right to the left; \u2022 Setting the plane in which the touching probe is moving: the OLXLYL plane; \u2022 The start and stop points of the scanning ; \u2022 Determining the approaching direction by setting the angle between the touching probe and the axes of the coordinate system: XL = 180 o , YL = 90 o , ZL = 90 o ; \u2022 The step of the scan: p = 0,3 mm; \u2022 The security distance: ds = 0,5 mm" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001113_gt2015-43561-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001113_gt2015-43561-Figure2-1.png", + "caption": "Figure 2: Gear and shroud geometry modelled in earlier CFD study (from [1]).", + "texts": [ + " The flow in this system is very complex and interactive; for example changes in the air flow in the system can negatively affect the sump which can cause oil to be ingested by the gear. This in turn leads to lower IGB efficiency. A computational model of this system would be a powerful tool for the aeroengine designer. Using commercial CFD, single phase (air only) models have been created of a shrouded crown spiral bevel gear [5, 6, 1 & 7] with a number of geometrical and operational parameters investigated. These CFD models are of the shrouded gear only with the computational domain extending from the shroud inlet to the rear cavity. For example, Figure 2 shows the geometry of Webb [1] where it can be seen that the upstream and downstream volumes are not connected and are in no way representative of chamber geometry. In particular, the chamber behind the gear is far larger than on test rig or engine and the outlet is not through shroud slots but through an idealised CFD outlet boundary as indicated in the figure. Webb\u2019s methodology for modelling these spiral bevel gears used a domain of one tooth valley. Comparison to a model of a five tooth passage sector yielded results no different than for the single valley model. Although Webb\u2019s pressure data showed a good correlation with Johnson et als\u2019 experimental results [4] he was unable to accurately match the windage losses, over-predicting gear torque by around 20%. The shroud covering a Trent-style IGB crown gear contains slots to allow oil and air to exit the gear rear chamber and travel into the front chamber from which it can be scavenged. Comparison of Figure 1 and Figure 2 shows that an aeroengine rear chamber is quite constrained in size compared to that utilised in the CFD models where attention was focussed on the gear-shroud system. Also, and perhaps more importantly, the CFD modelling has no coupling between outlet flow and inlet flow. Consequently, any swirl induced in the chamber in front of the gear by the gear rotation is not propagated through to the shroud inlet. Figure 3 shows a stylised representation of an IGB crown gear illustrating how it is likely that swirl at shroud exit slots would affect the condition of the air drawn into the gear at shroud inlet", + "44) where a maximum value for mass flow and gear torque are achieved. The amount of swirl introduced by flow exiting the shroud is, to some extent, controllable via the local shroud exit geometry and there may be value in conducting a design study on this feature. Having established how inlet conditions affect shroud performance, it is also important to understand how the shroud outlet affects the upstream flow through the shrouded gear system. The shroud geometry Webb modelled [7] is the geometry illustrated in Figure 2, characterised by the \u201cnose\u201d extension at the bottom of the gear and the gutter region near the top of the gear and the shroud fits closely to the gear. Figure 10c and d show this \u201cUTC design\u201d shroud geometry. Data for this shroud geometry, proposed and investigated by the UTC, is in [4]. The shroud geometry of the current CFD studies is somewhat more engine relevant and is illustrated in Figure 4; it has no \u201cnose\u201d or gutter. Figure 10a and b show this \u201cengine-relevant\u201d geometry alongside the \u201cUTC\u201d design facilitating direct comparison" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002480_2015-01-0680-Figure5-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002480_2015-01-0680-Figure5-1.png", + "caption": "Figure 5. Components of crank mechanism", + "texts": [], + "surrounding_texts": [ + "In a fired engine, there is one peak load every two connecting rod cycles. Therefore, the dynamic loading frequency is half of the connecting rod swing frequency. In the bench test rig, a proximity switch is utilized to detect the loading position. A flip-flop and AND gate are applied to produce a signal which is half the rod swing frequency. The proximity switch provides noncontact object detection for very close distances. On the rotation wheel, every position is synchronized with a rotation angle of the small end of the connecting rod. The load is applied to the small end of the connecting rod when the connecting rod rotates to the desired angle. This is accomplished by a metal target which is attached to the rotation wheel. When it is close to the proximity switch, the proximity switch will be triggered to a high voltage output. When the metal moves away from the proximity switch, the proximity switch will return to low voltage. The duration of high voltage will depend on the width of the metal target. From the output signal (low voltage triggered to high voltage), the rotation angle of the connecting rod at which the normal load is applied is detected, as seen in Figure 7. With a flip-flop, the frequency of the modified position signal is decreased by half of the original position signal. The modified position signal is combined through an AND logic calculation with the original signal. This generates the load signal satisfying the frequency requirement (Fig. 8). The experimental position and loading signal is shown in Figure 9. Because the PZT (Piezoelectric Transducer) actuator needs 100 ms to react from its original position to maximum displacement, the width of the high level loading signal which triggers the actuator should be at least 100 ms. If the width of loading occupies half of the entire loading period, the loading period is 400 ms when the rotation wheel is rotating at 300 rpm, at which the test rig rotates. Figure 10 shows the load on the piston assembly and rotation angle of the small end of the connecting rod versus the angle of the crank shaft. connecting rod versus the angle of the crank shaft" + ] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure13.10-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure13.10-1.png", + "caption": "Fig. 13.10. Experimental setup", + "texts": [ + " In this case, the standard RLS method is applied, being the linear filter a classical ARX system, with the input \u02c6F (e)(h) and the output F (h)4. Note, that in this case the two estimators work in a cascade configuration (see Fig. 13.9), and convergence problems do not arise. The only interaction is that the the input signal of the linear system is computed at each iteration on the basis of the the updated values of \u03b3i. In order to verify the above estimation algorithm, a laboratory setup has been implemented. The setup consists of a linear electric motor equipped with a position sensor and a load cell, Fig. 13.10. The measures required by the estimation algorithm have been obtained by imposing a motion profile to a linear motor in contact with different materials. The contact force F is measured by means of the load cell, the penetration \u03b4 is measured by comparing the current motor position with the position measured at the time of impact, and finally the penetration velocity \u03b4\u0307 is obtained from x by means of a state variable filter. The linear motor and the load cell have been connected, by means of a AD/DA board, to a standard PC running control and estimation algorithms in a mixed MATLAB/RTLinux environment" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001946_amm.698.552-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001946_amm.698.552-Figure9-1.png", + "caption": "Fig. 9. Mechanism (-100-80-70-110-120-60) and its direction vectors", + "texts": [ + " 1): \u03b11= -90\u00b0 \u03b12= -90\u00b0 \u03b13= -90\u00b0 \u03b14= -90\u00b0 \u03b15= -90\u00b0 \u03b16= -90\u00b0 So: , where k=\u00b11 (2) System (2) is correct for k=-1 so Bricard\u2019s linkage is mobile. For the mechanism shown in Fig. 6: \u03b11= -90\u00b0 \u03b12= -80\u00b0 \u03b13= -70\u00b0 \u03b14= -100\u00b0 \u03b15= -120\u00b0 \u03b16= -80\u00b0 So: , where k=\u00b11 (3) System (3) is not correct for k=\u00b11 so the mechanism is stationary. Let us change link parameters to provide mobility of the mechanism: \u03b11= -100\u00b0 \u03b12= -80\u00b0 \u03b13= - 70\u00b0 \u03b14= -110\u00b0 \u03b15= -120\u00b0 \u03b16= -60\u00b0 The assembly of mechanism links in a different order can not affect the assemblability, but it may affect the mobility. Changing the order of links of the mechanism shown in Fig. 9 to (-100-70- 60-80-110-120) will not affect its assemblability, but will lead to the loss of mobility as condition 1 will be violated. Opening Brikard\u2019s linkage for further analysis of its link chain and its direction vectors made it possible to reveal the condition of assemblability and mobility of Brikard\u2019s linkage modification: The modification of Brikard\u2019s linkage (\u03b11 \u03b12 \u03b13 \u03b14 \u03b15 \u03b16) is assemblable, if direction vectors of the first and last hinges in a chain are collinear. The modification of Brikard\u2019s linkage is mobile if all links rotate the direction vector in one direction (clockwise or counterclockwise), and each of the three link pairs rotates the direction vector through 180 degrees" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001067_978-3-540-71364-7_4-Figure26.8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001067_978-3-540-71364-7_4-Figure26.8-1.png", + "caption": "Fig. 26.8. Peg-in-hole scheme", + "texts": [ + " Although this is a simple task, it was chosen amongst others because it reflects in a straightforward manner the benefts of Geometric Constraints Haptic Guidance, namely the performance improvement due to the added precision and repeatability. Such benefits can be greater when dealing with more complex tasks that involve restricted motion along higher-order curves and surfaces, as well as sequences of multiple restricted movements, like assembly tasks. The peg-in-hole insertion has the following characteristics: \u2022 Since both peg and hole have a circular cross-section, a line-line coincidence constraint is defined between their revolution axes (Fig. 26.8). By setting this constraint 454 E. Nun\u0303o, A. Rodr\u0131\u0301guez, and L. Basan\u0303ez the motion of the peg is restricted to the hole\u2019s centerline, providing a natural guide towards the task goal. \u2022 The forces coming from the robot sensor provide information about the peg contact status, and since the peg is chamfered, these forces can help its guidance if contacts with the hole\u2019s edge occur. \u2022 The information carried by the guidance signal corresponds to the velocity of the haptic device (gs = x\u0307hd), according to Scheme 1, in Sec" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002696_icolim.2014.6934354-Figure8-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002696_icolim.2014.6934354-Figure8-1.png", + "caption": "Fig. 8. Special connector for passing the polymer pendulums", + "texts": [ + "7 is a conceptual diagram of the method applied to the strain tower arm. As seen in this figure, if the distance between the tower post and the jumper device or conductors is L, two polymer strings are installed on Ll3 and L position separately. In this case, the swing angle and the rope load can decrease, and then the cabin can move easily and softly. Related tools As a next problem, how to pass between polymer insulator strings is important. For exchanging the pendulums easily in the air, a special connector is needed as shown in Fig. 8. This is specially designed for the stepping pendulum method. As shown in figure, it has two coupling parts to be connected with polymer insulator strings and two holes for small sized blocks, one hole in the center point for a cabin. The assembling diagram of the connector with a cabin and strings is the same as Fig. 9. When using the above connector, a lineman can easily transfer from the pendulum No.1 to the pendulum No.2. IV. MOCKUP TEST FOR FIELD ApPLICABILITY After making related tools and polymer insulators, mockup test of the stepping pendulum method was carried out in 765kV single circuit test line for examining the applicable feasibility" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002157_detc2013-12233-Figure9-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002157_detc2013-12233-Figure9-1.png", + "caption": "Fig. 9. COORDINATE SYSTEMS OF A CRADLE-TYPE BEVEL GEAR GENERATOR WITHOUT A CUTTER TILT.", + "texts": [ + " Substituting Eq. (13) into Eq. (11) yields a tooth surface of generating gear in terms of only two variables, u and \u03b2 , which allow the tooth surface to be determined using the two boundary equations of the gear blank. 0 1 1=c c g z R z \u03c6 \u03b2 \u03b2= (13) THE MECHANISM OF FACE-HOBBED SBGS The imaginary generating gear is fixed on the cradle and their rotation axes coincide, and the work gear is placed on the cutting position. Then let the generating gear roll with the work gear to produce gear tooth surfaces. Figure 9 illustrates coordinate systems of a cradle-type bevel gear generator. Coordinate systems cS and 1S are rigidly connected to the imaginary generating gear and the work gear. dS , eS , and fS denote auxiliary coordinate systems that describe the relative motion between the generating gear and the work gear. 1\u03c6 is the rotation angle of the work gear. Using a coordinate transformation from coordinate system cS to 1S , the locus of tooth surface of the generating gear is represented as following: 1 2 1 1 1 2( , , , )= ( ) ( ) ( , )c f fe ed dc c cu u\u03b2 \u03c6 \u03c6 \u03c6 \u03c6 \u03b2r M M M M r (14) here 1 1 1 1 1 2 2 2 2 1 0 0 0 cos 0 sin 0 cos sin 0 0 1 0 0 = 0 sin cos 0 sin 0 cos 0 0 0 0 1 0 0 0 1 1 0 0 0 cos sin 0 0 0 1 0 sin cos 0 0 0 0 1 0 0 1 0 0 0 0 1 0 0 0 1 m m c m m c c m c c A E B \u03b3 \u03b3 \u03c6 \u03c6 \u03c6 \u03c6 \u03b3 \u03b3 \u03c6 \u03c6 \u03c6 \u03c6 \u0394\u23a1 \u23a4 \u23a1 \u23a4 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a3 \u23a6 \u23a3 \u23a6 \u23a1 \u23a4 \u23a1 \u23a4 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5 \u23a2 \u23a5\u0394 \u23a2 \u23a5 \u23a2 \u23a5 \u23a3 \u23a6 \u23a3 \u23a6 M - - - - - where mE is the vertical offset, B\u0394 is the sliding base feed setting, m\u03b3 is the machine root angle, A\u0394 is the increment of machine center to back" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002105_amr.308-310.1571-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002105_amr.308-310.1571-Figure2-1.png", + "caption": "Fig. 2. Micro-contact condition of sealing surface Fig. 3. Contact between rough surface and smooth surface", + "texts": [ + " 2)2(2 2 2 )( +\u2212\u2212 = DD l l r aa a AD an (5) According to the study of Wang and Komvopoulos[7], the size distribution function of the cross-sectional area of micro touch contact is deduced. 2)2(22)2( 2 )( +\u2212\u2212= DD l D aa D an \u03c8 (6) where \u03c8 is expansion coefficient of the size distribution range (\u03c8>1). Through the dichotomy to solve the Eq.7, \u03c8 value can be obtained. 1 /)2( )1( /)2(2/2/)2( = \u2212 +\u2212 \u2212\u2212\u2212 DD DDDD \u03c8\u03c8 (7) Fig.1 Qualitative description of statistical self-affinity for a surface profile Relationship between true contact area and compressive stress The machined sealing surfaces of the flange and the metal gasket are rough, as shown in fig.2. The surface hardness of the flange is usually HB40 above than that of the gasket, it can be thought that the surface topography of which under different compressive stresses is constant. Therefore the contact between the flange and metallic gasket sealing surfaces can be modeled as the contact of a smooth rigid flat surface with a rough fractal surface, as shown in Fig.3. According to Ref. [7], when the loads are imposed on the flange, the touch contacts of the small curvature radius in the sealing surface of the metallic gasket are forged plastic deformation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003369_detc2011-48166-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003369_detc2011-48166-Figure6-1.png", + "caption": "Figure 6. POSITION OF THE PROXIMITY PROBES.", + "texts": [ + " EACH PAD IS PROVIDED BY A TEMPERATURE SENSOR AND BY A PRESSURE PROBE. The bearing is a five shoe rocker-backed tilting pad journal type with a nominal diameter of 100 mm and a length-todiameter ratio of 0.7. As shown in Figure 5, all the pads of one of the two bearings installed on the test rig are instrumented with a temperature sensor and a pressure probe. The pressure probe measures the pressure in the middle of the pad by means of a hole in the babbitt metal. The position of the two proximity probes of each bearing is reported in Figure 6, whereas all the installed sensors are listed in Table 1. Table 1. LIST OF INSTALLED SENSORS. No. Description Code 1 Proximity Probe - Bearing 1 (NDE) - Dir. A X1A 2 Proximity Probe - Bearing 1 (NDE) - Dir. B X1B 3 Proximity Probe - Bearing 2 (DE) - Dir. A X2A 4 Proximity Probe - Bearing 2 (DE) - Dir. B X2B 5 Proximity Probe - Key Phasor XF 6 Proximity Probe - Bearing 1 (NDE) - Pad 4 X1P4 7 Proximity Probe - Bearing 1 (NDE) - Pad 5 X1P5 8 Load Cell - Bearing 1 (NDE) - Dir. H L1H 9 Load Cell - Bearing 1 (NDE) - Dir" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000831_978-1-84628-642-1_2-Figure2.34-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000831_978-1-84628-642-1_2-Figure2.34-1.png", + "caption": "Fig. 2.34. Four-link closed-chain planar arm with prismatic joint", + "texts": [ + " 2.7. Prove that the unit quaternion is invariant with respect to the rotation matrix and its transpose, i.e., R(\u03b7, \u03b5)\u03b5 = RT (\u03b7, \u03b5)\u03b5 = \u03b5. 2.8. Prove that the unit quaternion corresponding to a rotation matrix is given by (2.34), (2.35). 2.9. Prove that the quaternion product is expressed by (2.37). 2.10. By applying the rules for inverting a block-partitioned matrix, prove that matrix A1 0 is given by (2.45). 2.11. Find the direct kinematics equation of the four-link closed-chain planar arm in Fig. 2.34, where the two links connected by the prismatic joint are orthogonal to each other. 2.12. Find the direct kinematics equation for the cylindrical arm in Fig. 2.35. 2.13. Find the direct kinematics equation for the SCARA manipulator in Fig. 2.36. 2.14. Find the complete direct kinematics equation for the humanoid manipulator in Fig. 2.28. 2.15. For the set of minimal representations of orientation \u03c6, define the sum operation in terms of the composition of rotations. By means of an example, show that the commutative property does not hold for that operation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002962_j.ijmecsci.2014.03.030-Figure6-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002962_j.ijmecsci.2014.03.030-Figure6-1.png", + "caption": "Fig. 6. Internal bending moments.", + "texts": [ + " The Gauss\u2013 Codazzi equations make it possible to find three functions \u03c611;\u03c622, \u03c8 , if the coordinates of the surface \u2202D boundary are known. Thus, the shell represents a continual mechanism (there is no need to apply external forces to change its shape) if the plates are thin and the hinges are ideal. If the hinges are not ideal, the geometrically invariable shell will move only under load. Let us study the behavior of the described shell when the cylindrical hinges connecting the trapezoids are rigid plastic (Fig. 6). We will assume that if the angles \u03c81;\u03c82 are changed, internal bending moments M1;M2 occur in the hinges. Let us identify the densities of the internal moments m1 \u00bcM1=abh, m2 \u00bcM2=abh and assume as long as a21m 2 1\u00fea22m 2 2r2m2 y \u00f022\u00de the cell does not change its shape. In the formula (22) my is a constant of the material and a1; a2 are non-dimensional values. When f 2\u00f0mi\u00de \u00bc a21m 2 1\u00fea22m 2 2 \u00bc 2m2 y \u00f023\u00de the change rates of the angles \u03c81;\u03c82 can appear so the equality (24) takes place in case of the constant \u03bb40: \u03c81;t \u00bc _\u03c81 \u00bc \u03bbm1; \u03c82;t \u00bc _\u03c82 \u00bc \u03bbm2 \u00f024\u00de The expression a21m 2 1\u00fea22m 2 2 \u00bc 2m2 y is similar to the von Mises criterion that the dissipative potential corresponds to per (9) \u03a0\u00f0 _\u03c81; _\u03c82\u00de \u00bcmy ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 _\u03c82 1 a21 \u00fe _\u03c82 2 a22 !vuut \u00f025\u00de Let all the cells of the lattice shell be under the same excess internal gas pressure that is equal to p. The angles \u03c81;\u03c82 are limited by (3)\u2014\u0393\u00f0\u03c81;\u03c82\u00de \u00bc 0 [9] that depends on c\u00bc a=b and the angle \u03b2 by the base of the trapezoid (Fig. 6). In case of \u03b2-\u03c0=2 we have 2\u03c81\u00fe4\u03c82 \u00bc 4\u03c0. That is why j _\u03c81j \u00bc j2 _\u03c82j \u00f026\u00de The rate of the volume change will be _V=abh\u00bc \u00f0c cos \u03c81\u00fe cos \u00f0\u03c81=2\u00de\u00de _\u03c81 Rigid plastic motion of the shell is equal to the problem of the functional minimum L\u00f0 _\u03c81; _\u03c82\u00de \u00bc Z D \u03a0\u00f0 _\u03c81; _\u03c82\u00de d\u03b11 d\u03b12 Z D p\u00f0c cos \u03c81\u00fe cos \u03c81 2 \u00de _\u03c81d\u03b11 d\u03b12 \u00f027\u00de and the principle of the virtual powerZ D \u00f0m1\u03b4 _\u03c81\u00fem2\u03b4 _\u03c8 \u00de d\u03b11 d\u03b12 \u00bc Z D p \u03b4 _V d\u03b11 d\u03b12 The ultimate load theorem results in the following estimate of the kinematic coefficient pkr my p\u00f0c cos \u03c81\u00fe cos \u00f0\u03c81=2\u00de\u00de ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 1 a21 \u00fe 1 4a22 " + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0000938_978-3-030-65983-7-Figure3.17-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0000938_978-3-030-65983-7-Figure3.17-1.png", + "caption": "Fig. 3.17 VIVO tool simulation camera", + "texts": [ + " 3 Biomechanical Evaluation of Sharped Fractures \u2026 85 The instrumentation of the fracture was carried out by cutting with dental Dremel along the line shown in Fig. 3.15 (Table 3.11). 86 J. A. Beltr\u00e1n-Fern\u00e1ndez et al. For the physical testing of the lambda plate, it was proposed to perform a joint motion simulation with the help of the LIVE system\u2122 of AMTI which is a specialized tool for the simulation of this type of testing. This system focuses on the kinetic realism of the simulation to provide the closest possible approach to real conditions (Fig. 3.16). This system was selected for its six degrees of freedom at full speed and full load (Fig. 3.17). For physical analysis, the parameters obtained from the mandibular kinematics analysis for moderate jaw opening were used (see Tables 3.3 and 3.5). For the applied forcewith the system, a resulting axial forcewas applied to incisors of a value of 145 and 250N. The proposal of mounting the cadaveric jaw to the VIVO system is shown in Fig. 3.18. To carry out the simulation on the bony jaws, it was necessary to develop a base to hold the base of the jaw (by the chin bulge) and at the same time can be assembled to the machine guideline (Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003482_s207510871501006x-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003482_s207510871501006x-Figure4-1.png", + "caption": "Fig. 4. The layout of the radiation detectors on the rolling missile.", + "texts": [ + " Thus, the theoretical analysis has shown that the control in only one coordinate\u2014the polar radius of deviation from the beam axis\u2014provides for missile positioning in the beam center, but it causes low amplitude oscillations, the magnitude of which gradu ally decreases to zero. MEASURING THE ROLL ANGLE The proposed method of missile control involves measurement of the missile roll relative to the current radius vector of missile deviation from the beam cen ter. We call it a polar roll angle \u03b3p (Fig. 4). The task can be solved by placing one or two radiation detectors (R1, R2) on the end of the outer wing at a certain distance from the longitudinal axis of rotation. When two radiation detectors are used, each con nected to its own CDE, the radius of missile deflection c 2 1\u03c9 = \u03c0\u00d7 2 2 1 1 1( ) ( ) ,i i i i i iS y y z z S \u2212 \u2212 \u2212 = \u2212 + \u2212 + GYROSCOPY AND NAVIGATION Vol. 6 No. 1 2015 MISSILE CONTROL IN THE POLAR COORDINATE SYSTEM 69 \u03c1 and the current value sin\u03b3p or cos\u03b3p, needed to trans form the control command (4) to the rotating coordi nate system Y1OZ1, fixed to the missile body, can be determined by a simple procedure shown in Fig" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001095_978-94-007-7485-8_49-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001095_978-94-007-7485-8_49-Figure1-1.png", + "caption": "Fig. 1 Jeffcott rotor with an eccentric mass", + "texts": [ + " As mentioned before, the model chosen for this study is the classical Jeffcott rotor widely used in rotordynamics [11, 8, 9, 10, 5, 6, 7]. This simple but useful model consists in a massless shaft simply supported at the ends, with a concentrated mass (a disc). The crack is situated at the mid span of the shaft having a straight front, for sake of simplicity, oriented on a plane normal to the axis of the shaft. The eccentric mass has been placed on the disc of the Jeffcott rotor as an additional mass as can be seen in Figure 1.The round bar total length is equal to 900mm, whereas the diameter is 20mm. The material of the shaft is aluminium with the following mechanical properties: Youngs Modulus E=72GPa, Poisson ratio \u03bc=0.33 and density \u03c1=2800Kg/m3. The rotation of the shaft has been simulated considering eight different angular positions, one for every eighth of a rotation, called angle of rotation \u03c6, see Figure 2. At each angular position given, we analyze the static behavior of the shaft (considering the gravity effect), variables such as displacements, open portion of open crack and SIF, among others" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001567_icma.2015.7237783-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001567_icma.2015.7237783-Figure4-1.png", + "caption": "Fig. 4 Eyelid mechanism", + "texts": [ + "00 \u00a92015 IEEE Proceedings of 2015 IEEE International Conference on Mechatronics and Automation August 2 - 5, Beijing, China Fig. 3 is CAD drawing of eye mechanism. Eyeball mechanism has 4 DOFs to pitch and yaw eyeballs. Each motor and eyeball are connected by RSSR mechanism, which is a space linkage mechanism. The two eyeballs have separate mechanism, so they can make independent movement. The eyeball is 30mm diameter. They are manufactured by 3D printer. Each eyeball is equipped with a CCD camera of 13.5mm*13.5mm size to get visual information. Eyelid mechanism is showed in Fig. 4. The mechanism, including upper eyelid and lower eyelid, has 4 DOFs to open and close eyelids. The motors and eyelids also are connected by RSSR mechanism. As Fig. 5 shows, lid mechanism has 4 DOFs to change mouth shape. There are 2 towing points on the upper lid and 1 towing points on each corner of the mouth. RSSR mechanism is also adopted. Jaw mechanism has 3 DOFs. It is a three tier structure, which is showed in Fig. 6. The top tier is upper jaw, connected with the head skeleton. The bottom tier is lower jaw, connected with the robot's chin" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0002315_j.proeng.2011.05.102-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0002315_j.proeng.2011.05.102-Figure1-1.png", + "caption": "Fig. 1. Schematic CAD model of bare cylinder in RMIT Industrial Wind Tunnel [4]", + "texts": [ + " Nomenclature CD Drag Coefficient FD Drag Force, N Density of Air, kg/m3 Re Reynolds Number V Velocity of Air, m/s A Projected Frontal Area of Cylinder, m2 d Diameter of Cylinder, m Dynamic Viscosity, N.s/m2 With a view to obtain aerodynamic properties experimentally for a range of commercially available swimsuits made of various materials composition, a 110 mm diameter cylinder was manufactured. The cylinder was made of PVC material and used some filler to make it structurally rigid. The cylinder was vertically supported on a six components transducer (type JR-3) had a sensitivity of 0.05% over a range of 0 to 200 N as shown in Figure 1. The aerodynamic forces and their moments were measured for a range of Reynolds numbers based on cylinder diameter and varied wind tunnel air speeds (from 10 km/h to 130 km/h with an increment of 10 km/h). Each test was conducted as a function of swimsuit\u2019s seam positions (see Figure 2). As mentioned earlier, the RMIT Industrial Wind Tunnel was used to measure the aerodynamic properties of swimsuit fabrics. The tunnel is a closed return circuit wind tunnel with a turntable to simulate the cross wind effects" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003203_detc2014-34143-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003203_detc2014-34143-Figure1-1.png", + "caption": "Fig. 1. Rotational lumped parameter model of planetary gears.", + "texts": [ + " Finally, the influences of various parameters on instabilities are investigated, such as the amplitude of mesh stiffness variation, the amplitude of speed fluctuations and the frequency ratio. Downloaded From: http://proceedings.asmedigitalcollection.asme.org/pdfaccess.ashx?url=/data/conferences/asmep/82128/ on 04/06/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use 2 Copyright \u00a9 2014 by ASME The analysis deals with rotational vibration of planetary gears with equally spaced planets submitted to engine speed fluctuations. A rotational lumped-parameter model of planetary gears is shown in Fig. 1. All components are modeled as rigid bodies with moments of inertia Ic, Ir, Is, Ip. Gear mesh deformations are represented by linear springs acting along the line of action. Because of engine speed fluctuation, the input speed for rigid-body conditions can be introduced via a Fourier series as [17] 0 0 0 (t) sin t (1) where \u03b1 is the amplitude of speed fluctuations. Based on the kinematic relationship, the rotation angle of the input component can be expressed as 0 0 cos cos t p t dt t t (2) where \u03c9=p\u03a90 is the nominal mesh frequency without considering the speed fluctuation" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003606_amm.339.510-Figure2-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003606_amm.339.510-Figure2-1.png", + "caption": "Figure 2 Finite element model of gear thermal analysis", + "texts": [ + " Build finite element model of thermal analysis. Aim for simulating real actual working condition of gear, ensuring rigidity of teeth and body of gear, and simplifying finite element analysis scale, three teeth on the gear are built in the process of modeling, and the rest part is showed with cylindrical which critical dimension is pitch circle; Choose the condition when gears are meshing on the middle pitch line, and use finite element model of gear thermal analysis based on SOLID 70 element, as is shown in figure 2. Loading and solving thermal analysis model. For a single tooth of gear, according to the differences of heat transfer style of each part, there are 5 calculation areas. As is shown in figure 3, the calculation of boundary condition is shown in reference [5] . In the figure 3, GM area is the meshing surface;GT1, GT2, GT3 are separately top, root and no meshing surface of tooth; GD area is end surface; Gw area is bottom surface; Gj, GJ are sections surfaces of tooth. Besides, load average heat flux density and convection heat transfer coefficient of meshing surface on the pitch cylindrical surface" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0003982_saci.2014.6840063-Figure4-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0003982_saci.2014.6840063-Figure4-1.png", + "caption": "Figure 4. The manipulator moving unit", + "texts": [ + "2) Effect: H1 cylinder will move to the position 1 with the help of solenoid Y1, and after, H1 immediately returning to the position 0, and Y1 will be inverted. 8. Transportation During the transportation the sensors A\u00c91, A\u00c92, A\u00c93 are determining the materials (steel/copper/plastic), the control mechanism evaluating the sensory data and base on evaluation sending the position command to the slip-way unit. (see Fig.3) 9. Gripper lowing End of transport 10. Gripping It is mean that H3 cylinder is in lower position. The gripper is closed. 11. Gripper lifting Gripper is closed (see Fig. 4) 12. Manipulator rotating Cylinder H3 is in the upper position. The gripper is above slipway. 13. Gripping It is mean that H3 cylinder is in lower position. The gripper is closed. 14. Specimen releasing The gripper is above the slip-way. The gripper is releasing the specimen and the specimen is slipping into the correspondent storing unit. This cycle will continue, until specimen is in the feeder tube, or any mistake is occurring. VI. SENSORS FOR THE SORTING In mechatronic system there are used three types of sensors: 1" + ], + "surrounding_texts": [] + }, + { + "image_filename": "designv11_84_0001578_iccse.2015.7250304-Figure1-1.png", + "original_path": "designv11-84/openalex_figure/designv11_84_0001578_iccse.2015.7250304-Figure1-1.png", + "caption": "Figure 1. Schematic for path yaw angle", + "texts": [ + " The fitness equation is as follows: )()()()( 332211 * PfCPfCPfCCPF (1) Where 21,, CCC and 3C are constants implying weight coefficients which are set in line with the requirements and environment conditions, while )(1 Pf , )(2 Pf and )(3 Pf denote the length, smooth and safe of the path in the multirobot fish system. Below are their definitions: Path length n i iPLengthPf 1 1 )}({)( (2) Where l j jiiji ppPLength 1 )1()( is the path length of robotic fish i. Path smooth n i iPTurningPf 1 2 )}({)( (3) Where l j ijiPTurning 2 )( ),...,2( ljij is the absolute value of the yaw angle between vector ijji PP )1( and vector )1( jiijPP as shown in Fig.1, with 21ij Path safe n i iPSafePf 1 3 )()( (4) Where otherwise dyyxx PSafe ii i ,0 )()(,1 )( 2 0 2 0 , with centers of robotic fish and obstacles as their coordinates, coordinates ),( ii yx and ),( 00 yx implying the positions of robotic fish i and the obstacle or another robotic fish nearest to it respectively, d=8r as safe distance ( supposing a robotic fish as ellipse with 4r-long major axis and r-long minor axis and an obstacle as a circle with 4r-long radius ). III. ALGORITHM DESCRIPTION Using n subsets to show the paths of n robotic fish, with each chromosome representing a path of robotic fish" + ], + "surrounding_texts": [] + } +] \ No newline at end of file