[ { "text": "Numerical Simulation of Acoustic Resonance and Streaming in Droplet\n Acoustofluidics: This paper reports the phenomenon of resonance weakening and streaming onset\nin two phase acoustofluidics by performing numerical simulations of a capillary\ndroplet suspended in a microfluidic chamber. The simulations show that\ndepending on the relative acoustic properties of the two phases, it is possible\nto observe (i)~the decrease in the total acoustic energy as the oscillation\namplitude at the wall increases, and (ii)~the onset of acoustic streaming. The\nimpact of these findings in terms of acoustic focusing inside droplets is also\ndiscussed.", "category": "physics_flu-dyn" }, { "text": "Capillary rise of water in hydrophilic nanopores: We report on the capillary rise of water in three-dimensional networks of\nhydrophilic silica pores with 3.5nm and 5nm mean radii, respectively (porous\nVycor monoliths). We find classical square root of time Lucas-Washburn laws for\nthe imbibition dynamics over the entire capillary rise times of up to 16h\ninvestigated. Provided we assume two preadsorbed strongly bound layers of water\nmolecules resting at the silica walls, which corresponds to a negative velocity\nslip length of -0.5nm for water flow in silica nanopores, we can describe the\nfilling process by a retained fluidity and capillarity of water in the pore\ncenter. This anticipated partitioning in two dynamic components reflects the\nstructural-thermodynamic partitioning in strongly silica bound water layers and\ncapillary condensed water in the pore center which is documented by sorption\nisotherm measurements.", "category": "physics_flu-dyn" }, { "text": "Mesoscale Elucidation of Biofilm Shear Behavior: Formation of bacterial colonies as biofilm on the surface/interface of\nvarious objects has the potential to impact not only human health and disease\nbut also energy and environmental considerations. Biofilms can be regarded as\nsoft materials, and comprehension of their shear response to external forces is\na key element to the fundamental understanding. A mesoscale model has been\npresented in this article based on digitization of a biofilm microstructure.\nIts response under externally applied shear load is analyzed. Strain stiffening\ntype behavior is readily observed under high strain loads due to the unfolding\nof chains within soft polymeric substrate. Sustained shear loading of the\nbiofilm network results in strain localization along the diagonal direction.\nRupture of the soft polymeric matrix can potentially reduce the intercellular\ninteraction between the bacterial cells. Evolution of stiffness within the\nbiofilm network under shear reveals two regions: a) initial increase in\nstiffness due to strain stiffening of polymer matrix, and b) eventual reduction\nin stiffness because of tear in polymeric substrate.", "category": "physics_flu-dyn" }, { "text": "Variational approach for the flow of Ree-Eyring and Casson fluids in\n pipes: The flow of Ree-Eyring and Casson non-Newtonian fluids is investigated using\na variational principle to optimize the total stress. The\nvariationally-obtained solutions are compared to the analytical solutions\nderived from the Weissenberg-Rabinowitsch-Mooney equation and the results are\nfound to be identical within acceptable numerical errors and modeling\napproximations.", "category": "physics_flu-dyn" }, { "text": "A weakly compressible hybridizable discontinuous Galerkin formulation\n for fluid-structure interaction problems: A scheme for the solution of fluid-structure interaction (FSI) problems with\nweakly compressible flows is proposed in this work. A novel hybridizable\ndiscontinuous Galerkin (HDG) method is derived for the discretization of the\nfluid equations, while the standard continuous Galerkin (CG) approach is\nadopted for the structural problem. The chosen HDG solver combines robustness\nof discontinuous Galerkin (DG) approaches in advection-dominated flows with\nhigher order accuracy and efficient implementations. Two coupling strategies\nare examined in this contribution, namely a partitioned Dirichlet-Neumann\nscheme in the context of hybrid HDG-CG discretizations and a monolithic\napproach based on Nitsche's method, exploiting the definition of the numerical\nflux and the trace of the solution to impose the coupling conditions. Numerical\nexperiments show optimal convergence of the HDG and CG primal and mixed\nvariables and superconvergence of the postprocessed fluid velocity. The\nrobustness and the efficiency of the proposed weakly compressible formulation,\nin comparison to a fully incompressible one, are also highlighted on a\nselection of two and three dimensional FSI benchmark problems.", "category": "physics_flu-dyn" }, { "text": "Nonlinear Wave Transformation over Steep Breakwaters: Wave shoaling of water waves over mild bottom slopes is well described by\nlinearized theories. However, the analytical treatment of nonlinear wave\nshoaling subject to rapidly varying bottoms has proven to be elusive in the\npast decades. As the spatial evolution of the exceedance probability of\nirregular waves is affected by second-order effects in steepness, the nonlinear\nshoaling coefficient throughout a symmetrical and steep breakwater is\ninvestigated through a stochastic framework. By inverting the effect of slope\non normalized wave height distribution, it is possible to obtain a closed-form\nslope dependence of the nonlinear shoaling coefficient compatible with\nexperiments over steep breakwaters.", "category": "physics_flu-dyn" }, { "text": "A critical point for bifurcation cascades and featureless turbulence: In this Letter we show that a bifurcation cascade and fully sustained\nturbulence can share the phase space of a fluid flow system, resulting in the\npresence of competing stable attractors. We analyse the toroidal pipe flow,\nwhich undergoes subcritical transition to turbulence at low pipe curvatures and\nsupercritical transition at high curvatures, as was previously documented. We\nprovide decisive evidence that the nature of the supercritical transition is of\nRuelle--Takens type and that, in a narrow range of intermediate curvatures, its\ndynamics competes with that of sustained turbulence emerging through\nsubcritical transition mechanisms.", "category": "physics_flu-dyn" }, { "text": "Stochastic representation of the Reynolds transport theorem: revisiting\n large-scale modeling: We explore the potential of a formulation of the Navier-Stokes equations\nincorporating a random description of the small-scale velocity component. This\nmodel, established from a version of the Reynolds transport theorem adapted to\na stochastic representation of the flow, gives rise to a large-scale\ndescription of the flow dynamics in which emerges an anisotropic subgrid\ntensor, reminiscent to the Reynolds stress tensor, together with a drift\ncorrection due to an inhomogeneous turbulence. The corresponding subgrid model,\nwhich depends on the small scales velocity variance, generalizes the Boussinesq\neddy viscosity assumption. However, it is not anymore obtained from an analogy\nwith molecular dissipation but ensues rigorously from the random modeling of\nthe flow. This principle allows us to propose several subgrid models defined\ndirectly on the resolved flow component. We assess and compare numerically\nthose models on a standard Green-Taylor vortex flow at Reynolds 1600. The\nnumerical simulations, carried out with an accurate divergence-free scheme,\noutperform classical large-eddies formulations and provides a simple\ndemonstration of the pertinence of the proposed large-scale modeling.", "category": "physics_flu-dyn" }, { "text": "The Clapping Book: A steady horizontal air stream flows across a book clamped at its downstream\nend. Pages lift off to form a growing bent stack whose shape is determined by\nthe torques associated with aerodynamic forces, weight and elastic resistance\nto bending. As more pages lift off to join the bent stack, the increasing\nimportance of bending rigidity to dynamic pressure eventually causes the book\nto clap shut. The process restarts, and self-sustained oscillations emerge.\n[Fluid dynamics video]", "category": "physics_flu-dyn" }, { "text": "Sparse Convolution-based Markov Models for Nonlinear Fluid Flows: Data-driven modeling for nonlinear fluid flows using sparse convolution-based\nmapping into a feature space where the dynamics are Markov linear is explored\nin this article. The underlying principle of low-order models for fluid systems\nis identifying convolutions to a feature space where the system evolution (a)\nis simpler and efficient to model and (b) the predictions can be reconstructed\naccurately through deconvolution. Such methods are useful when real-time models\nfrom sensor data are needed for online decision making. The Markov linear\napproximation is popular as it allows us to leverage the vast linear systems\nmachinery. Examples include the Koopman operator approximation techniques and\nevolutionary kernel methods in machine learning. The success of these models in\napproximating nonlinear dynamical systems is tied to the effectiveness of the\nconvolution map in accomplishing both (a) and (b) mentioned above. To assess\nthis, we perform in-depth study of two classes of sparse convolution operators:\n(i) a pure data-driven POD-convolution that uses left singular vectors of the\ndata snapshots - a staple of Koopman approximation methods and (ii) a sparse\nGaussian Process (sGP) convolution that combines sparse sampling with a\nGaussian kernel embedding an implicit feature map to an inner product\nreproducing kernel Hilbert space.", "category": "physics_flu-dyn" }, { "text": "Turbulent fluxes of entropy and internal energy in temperature\n stratified flows: We derive equations for the mean entropy and the mean internal energy in the\nlow-Mach-number temperature stratified turbulence (i.e., for turbulent\nconvection or stably stratified turbulence), and show that turbulent flux of\nentropy is given by ${\\bf F}_s=\\overline{\\rho} \\, \\overline{{\\bf u} s}$, where\n$\\overline{\\rho}$ is the mean fluid density, $s$ are fluctuations of entropy\nand overbars denote averaging over an ensemble of turbulent velocity field,\n${\\bf u}$. We demonstrate that the turbulent flux of entropy is different from\nthe turbulent convective flux, ${\\bf F}_c=\\overline{T} \\, \\overline{\\rho} \\,\n\\overline{{\\bf u} s}$, of the fluid internal energy, where $\\overline{T}$ is\nthe mean fluid temperature. This turbulent convective flux is well-known in the\nastrophysical and geophysical literature, and it cannot be used as a turbulent\nflux in the equation for the mean entropy. This result is exact for\nlow-Mach-number temperature stratified turbulence and is independent of the\nmodel used. We also derive equations for the velocity-entropy correlation,\n$\\overline{{\\bf u} s}$, in the limits of small and large Peclet numbers, using\nthe quasi-linear approach and the spectral tau approximation, respectively.\nThis study is important in view of different applications to the astrophysical\nand geophysical temperature stratified turbulence.", "category": "physics_flu-dyn" }, { "text": "Mass and moment of inertia govern the transition in the dynamics and\n wakes of freely rising and falling cylinders: In this Letter, we study the motion and wake-patterns of freely rising and\nfalling cylinders in quiescent fluid. We show that the amplitude of oscillation\nand the overall system-dynamics are intricately linked to two parameters: the\nparticle's mass-density relative to the fluid $m^* \\equiv \\rho_p/\\rho_f$ and\nits relative moment-of-inertia $I^* \\equiv {I}_p/{I}_f$. This supersedes the\ncurrent understanding that a critical mass density ($m^*\\approx$ 0.54) alone\ntriggers the sudden onset of vigorous vibrations. Using over 144 combinations\nof ${m}^*$ and $I^*$, we comprehensively map out the parameter space covering\nvery heavy ($m^* > 10$) to very buoyant ($m^* < 0.1$) particles. The entire\ndata collapses into two scaling regimes demarcated by a transitional Strouhal\nnumber, $St_t \\approx 0.17$. $St_t$ separates a mass-dominated regime from a\nregime dominated by the particle's moment of inertia. A shift from one regime\nto the other also marks a gradual transition in the wake-shedding pattern: from\nthe classical $2S$~(2-Single) vortex mode to a $2P$~(2-Pairs) vortex mode.\nThus, auto-rotation can have a significant influence on the trajectories and\nwakes of freely rising isotropic bodies.", "category": "physics_flu-dyn" }, { "text": "Stationary solution for quasi-homogeneous small-scale magnetic field\n advected by non-Gaussian turbulent flow: We consider fluctuations of magnetic field excited by external force and\nadvected by isotropic turbulent flow. It appears that non-Gaussian velocity\ngradient statistics and finite region of pumping force provide the existence of\nstationary solution. The mean-square magnetic field is calculated for arbitrary\nvelocity gradient statistics. An estimate for possible feedback of magnetic\nfield on velocity shows that, for wide range of parameters, stationarity\nwithout feedback would take place even in the case of intensive pumping of\nmagnetic field.", "category": "physics_flu-dyn" }, { "text": "Numerical evidence of anomalous energy dissipation in incompressible\n Euler flows: Towards grid-converged results for the inviscid Taylor-Green\n problem: Providing evidence of finite-time singularities of the incompressible Euler\nequations in three space dimensions is still an unsolved problem. Likewise, the\nzeroth law of turbulence has not been proven to date by numerical experiments.\nWe address this issue by high-resolution numerical simulations of the inviscid\nthree-dimensional Taylor-Green vortex problem using a novel high-order\ndiscontinuous Galerkin discretization approach. Our main finding is that the\nkinetic energy evolution does not tend towards exact energy conservation for\nincreasing spatial resolution of the numerical scheme, but instead converges to\na solution with nonzero kinetic energy dissipation rate. This implies an energy\ndissipation anomaly in the absense of viscous dissipation according to\nOnsager's conjecture, and serves as an indication of finite-time singularities\nin incompressible inviscid flows. We demonstrate convergence to a dissipative\nsolution for the three-dimensional inviscid Taylor-Green problem with a\nmeasured relative $L_2$-error of $0.27 \\%$ for the temporal evolution of the\nkinetic energy and $3.52 \\%$ for the kinetic energy dissipation rate.", "category": "physics_flu-dyn" }, { "text": "Energy Spectrum of Buoyancy-Driven Turbulence: Using high-resolution direct numerical simulation and arguments based on the\nkinetic energy flux $\\Pi_u$, we demonstrate that for stably stratified flows,\nthe kinetic energy spectrum $E_u(k) \\sim k^{-11/5}$, the entropy spectrum\n$E_\\theta(k) \\sim k^{-7/5}$, and $\\Pi_u(k) \\sim k^{-4/5}$, consistent with the\nBolgiano-Obukhov scaling. This scaling arises due to the conversion of kinetic\nenergy to the potential energy by buoyancy. For weaker buoyancy, this\nconversion is weak, hence $E_u(k)$ follows Kolmogorov's spectrum with a\nconstant energy flux. For Rayleigh B\\'{e}nard convection, we show that the\nenergy supply rate by buoyancy is positive, which leads to an increasing\n$\\Pi_u(k)$ with $k$, thus ruling out Bolgiano-Obukhov scaling for the\nconvective turbulence. Our numerical results show that convective turbulence\nfor unit Prandt number exhibits a constant $\\Pi_u(k)$ and $E_u(k) \\sim\nk^{-5/3}$ for a narrow band of wavenumbers.", "category": "physics_flu-dyn" }, { "text": "Molecular Dynamics Simulations of Janus Particle Dynamics in Uniform\n Flow: We use molecular dynamics simulations to study the dynamics of Janus\nparticles, micro- or nanoparticles which are not spherically symmetric, in the\nuniform flow of a simple liquid. In particular we consider spheres with an\nasymmetry in the solid-liquid interaction over their surfaces and calculate the\nforces and torques experienced by the particles as a function of their\norientation with respect to the flow. We also examine particles that are\ndeformed slightly from a spherical shape. We compare the simulation results to\nthe predictions of a previously introduced theoretical approach, which computes\nthe forces and torques on particles with variable slip lengths or aspherical\ndeformations that are much smaller than the particle radius. We find that there\nis good agreement between the forces and torques computed from our simulations\nand the theoretical predictions, when the slip condition is applied to the\nfirst layer of liquid molecules adjacent to the surface.", "category": "physics_flu-dyn" }, { "text": "Stability of the Couette-Poiseuille flow by the Reynolds-Orr energy\n equation: The normal-mode analysis of the Reynolds-Orr energy equation governing the\nstability of viscous motion for general three-dimensional disturbances has been\nrevisited. The energy equation has been solved as an unconstrained minimization\nproblem for the Couette-Poiseuille flow. The minimum Reynolds number for every\nCouette-Poiseuille velocity profile has been computed and compared with those\navailable in the literature. For fully three-dimensional disturbances, it is\nshown that the minimum Reynolds number is in general smaller than the\ncorresponding two-dimensional counterpart for all the Couette-Poiseuille\nprofiles except plane Couette flow.", "category": "physics_flu-dyn" }, { "text": "Leidenfrost explosions: We present a fluid dynamics video showing the behavior of Leidenfrost\ndroplets composed by a mixture of water and surfactant (SDS, Sodium Dodecyl\nsulfate).\n When a droplet is released on a plate heated above a given temperature a thin\nlayer of vapor isolates the droplet from the plate. The droplet levitates over\nthe plate. This is called the Leidenfrost effect.\n In this work we study the influence of the addition of a surfactant on the\nLeidenfrost phenomenon. As the droplet evaporates the concentration of SDS\nrises up to two orders of magnitude over the Critical Micelle Concentration\n(CMC). An unexpected and violent explosive behavior is observed. The video\npresents several explosions taken with a high speed camera (IDT-N4 at 30000\nfps). All the presented experiments were performed on a plate heated at\n300{\\deg}C. On the other hand, the initial quantity of SDS was tuned in two\nways: (i) by varying the initial concentration of SDS and (ii) by varying the\ninitial size of the droplet. By measuring the volume of the droplet just before\nthe explosion, we were able to estimate the final concentration of SDS. We\nfound that the explosion always occurs around a critical concentration, about\n100 times the CMC.\n The droplets have also been studied just before the explosion. By isolating\nthe droplet on a cold plate just before the explosion, we evidenced the\npresence of a shell surrounding a liquid core.\n We conclude that above a critical concentration a solid shell is formed. This\nleads to an increase of pressure into the droplet until the shell breaks. The\nrelease of the pressure is accompanied by a violent explosion, and in some\ncases foaming.", "category": "physics_flu-dyn" }, { "text": "Effects of interfacial curvature on Rayleigh-Taylor instability: In this work a non-trivial effect of the interfacial curvature on the\nstability of accelerated interfaces, such as liquid rims, is uncovered. The new\nstability analysis, based on operator and boundary perturbation theories,\nreveals and quantifies influence of the interfacial curvature on the growth\nrate and on the wavenumber selection of the Rayleigh-Taylor instability. The\nsystematic approach developed here also provides a rigorous generalization of\nthe widely used \\textit{ad hoc} idea, due to Layzer [Astrophys. J.\n\\textbf{122}, 1-12 (1955)], of approximating the potential velocity field near\nthe interface.", "category": "physics_flu-dyn" }, { "text": "Flow onset for a single bubble in yield-stress fluids: We use computational methods to determine the minimal yield-stress required\nin order to hold static a buoyant bubble in a yield-stress liquid. The static\nlimit is governed by the bubble shape, the dimensionless surface tension\n($\\gamma$) and the ratio of the yield-stress to the buoyancy stress ($Y$). For\na given geometry, bubbles are static for $Y > Y_c$, which we determine for a\nrange of shapes. Given that surface tension is negligible, long prolate bubbles\nrequire larger yield-stress to hold static compared to oblate bubbles. Non-zero\n$\\gamma$ increases $Y_c$ and for large $\\gamma$ the yield-capillary number\n($Y/\\gamma$) determines the static boundary. In this limit, although bubble\nshape is important, bubble orientation is not. 2D planar and axisymmetric\nbubbles are studied.", "category": "physics_flu-dyn" }, { "text": "Transition scenario of the round jet in crossflow topology at low\n velocity ratios: We study experimentally a round Jet In CrossFlow (JICF) at low values of the\njet to-crossflow velocity ratio R using instantaneous and time-averaged\nthree-dimensions three-components (3D3C) velocimetry. The difference between\ninstantaneous and time-averaged swirling structures of the JICF is emphasized.\nThrough the analysis of spatial distribution of instantaneous transverse and\nlongitudinal vortices the main transitions of the JICF are characterized for\n0.15 < R < 2.2. A new transition at very low velocity ratio is found (R < 0.3).\nWhen R is large enough (R > 1.25), the classic JICF topology is recovered. In\nbetween, a deformation of the classical JICF topology is observed consisting in\na progressive disappearance of the leading-edge vortices, a bending of the jet\ntrajectory and thus a strengthened interaction with the boundary layer. toward\nthe wall. Thanks to a state-of-the-art review on the JICF topology and using\nvisualizations of the flow structures extracted from our experimental\nvolumetric velocimetry measurements, this article provides a complete\ntransition scenario of the JICF topology from the high velocity ratios to the\nlowest ones, and gives the topological transition threshold associated with\neach kind of vortex.", "category": "physics_flu-dyn" }, { "text": "Estimation of the Characteristic Wavelength Parameter in 1D\n Leray-Burgers Equation with PINN: In this paper, we employ the Physics-Informed Neural Network (PINN) to\nestimate the practical range of the characteristic wavelength\nparameter(referred to as the smoothing parameter) $\\alpha$ in the Leray-Burgers\nequation. The Leray-Burgers equation, a regularization of the inviscid Burgers\nequation, incorporates a Helmholtz filter with a characteristic wavelength\n$\\alpha$ to replace the usual convective velocity, inducing a regularized\nconvective velocity. The filter bends the equation's characteristics slightly\nand makes them not intersect each other, leading to a global solution in time.\nBy conducting computational experiments with various initial conditions, we\ndetermine the practical range of $\\alpha >0$ that closely approximates the\nsolutions of the inviscid Burgers equation. Our findings indicate that the\nvalue of $\\alpha$ depends on the initial data, with the practical range of\n$\\alpha$ being between 0.01 and 0.05 for continuous initial profiles and\nbetween 0.01 and 0.03 for discontinuous initial profiles. The Leray-Burgers\nequation captures shock and rarefaction waves within the temporal domain for\nwhich training data exists. However, as the temporal domain extends beyond the\ntraining interval, data-driven forward computation demonstrates that the\npredictions generated by the PINN start to deviate from the exact solutions.\nThis study also highlights the effectiveness and efficiency of the\nLeray-Burgers equation in real practical problems, specifically Traffic State\nEstimation.", "category": "physics_flu-dyn" }, { "text": "Disjoining Potential and Spreading of Thin Liquid Layers in the Diffuse\n Interface Model Coupled to Hydrodynamics: The hydrodynamic phase field model is applied to the problem of film\nspreading on a solid surface. The disjoining potential, responsible for\nmodification of the fluid properties near a three-phase contact line, is\ncomputed from the solvability conditions of the density field equation with\nappropriate boundary conditions imposed on the solid support. The equation\ndescribing the motion of a spreading film are derived in the lubrication\napproximation. In the case of quasi-equilibrium spreading, is shown that the\ncorrect sharp-interface limit is obtained, and sample solutions are obtained by\nnumerical integration. It is further shown that evaporation or condensation may\nstrongly affect the dynamics near the contact line, and accounting for kinetic\nretardation of the interphase transport is necessary to build up a consistent\ntheory.", "category": "physics_flu-dyn" }, { "text": "From the simple reacting sphere kinetic model to the reaction-diffusion\n system of Maxwell-Stefan type: In this paper we perform a formal asymptotic analysis on a kinetic model for\nreactive mixtures in order to derive a reaction-diffusion system of\nMaxwell-Stefan type. More specifically, we start from the kinetic model of\nsimple reacting spheres for a quaternary mixture of monatomic ideal gases that\nundergoes a reversible chemical reaction of bimolecular type. Then, we consider\na scaling describing a physical situation in which mechanical collisions play a\ndominant role in the evolution process, while chemical reactions are slow, and\ncompute explicitly the production terms associated to the concentration and\nmomentum balance equations for each species in the reactive mixture. Finally,\nwe prove that, under isothermal assumptions, the limit equations for the scaled\nkinetic model is the reaction diffusion system of Maxwell-Stefan type.", "category": "physics_flu-dyn" }, { "text": "Turbulence and heat transfer on a rotating, heated half soap bubble: We use Direct Numerical Simulations to study the two-dimensional flow of a\nrotating, half soap bubble that is heated at its equator. The heating produces\nbuoyancy and rotation generates a Coriolis forces in the fluid. However, due to\nthe curved surface of the bubble, the buoyancy and Coriolis forces vary with\nlatitude on the bubble, giving rise to rich flow behavior. We first explore the\nsingle-point properties of the flow, including the Reynolds and Nusselt\nnumbers, mean fields, and Reynolds stresses, all as a function of latitude. For\na given Rayleigh number, we observe a non-monotonic dependence on the Rossby\nnumber Ro, and large scale mean circulations that are strongly influenced by\nrotation. We then consider quantities that reveal the multiscale nature of the\nflow, including spectrums and spectral fluxes of kinetic and thermal energy,\nand enstrophy, and structure functions of velocity and temperature. The fluxes\nshow that just a for nonbuoyant two-dimensional turbulence on a flat surface,\nthere is an upscale flux of kinetic energy at larger scales (fed by buoyancy\ninjection of turbulent kinetic energy at smaller scales), and a downscale flux\nof enstrophy at smaller scales. The kinetic energy spectrum and velocity\nstructure functions are well described by Bolgiano-Obukhov (BO) scaling at\nscales where the effects of rotation are weak. The temperature structure\nfunctions do not, however, satisfy BO scaling in general, due to strong\nintermittency in the temperature field.", "category": "physics_flu-dyn" }, { "text": "Still water: dead zones and collimated ejecta from the impact of\n granular jets: When a dense granular jet hits a target, it forms a large dead zone and\nejects a highly collimated conical sheet with a well-defined opening angle.\nUsing experiments, simulations, and continuum modeling, we find that this\nopening angle is insensitive to the precise target shape and the dissipation\nmechanisms in the flow. We show that this surprising insensitivity arises\nbecause dense granular jet impact, though highly dissipative, is nonetheless\ncontrolled by the limit of perfect fluid flow.", "category": "physics_flu-dyn" }, { "text": "Data-Driven Filtered Reduced Order Modeling Of Fluid Flows: We propose a data-driven filtered reduced order model (DDF-ROM) framework for\nthe numerical simulation of fluid flows. The novel DDF-ROM framework consists\nof two steps: (i) In the first step, we use explicit ROM spatial filtering of\nthe nonlinear PDE to construct a filtered ROM. This filtered ROM is\nlow-dimensional, but is not closed (because of the nonlinearity in the given\nPDE). (ii) In the second step, we use data-driven modeling to close the\nfiltered ROM, i.e., to model the interaction between the resolved and\nunresolved modes. To this end, we use a quadratic ansatz to model this\ninteraction and close the filtered ROM. To find the new coefficients in the\nclosed filtered ROM, we solve an optimization problem that minimizes the\ndifference between the full order model data and our ansatz. We emphasize that\nthe new DDF-ROM is built on general ideas of spatial filtering and optimization\nand is independent of (restrictive) phenomenological arguments.\n We investigate the DDF-ROM in the numerical simulation of a 2D channel flow\npast a circular cylinder at Reynolds number $Re=100$. The DDF-ROM is\nsignificantly more accurate than the standard projection ROM. Furthermore, the\ncomputational costs of the DDF-ROM and the standard projection ROM are similar,\nboth costs being orders of magnitude lower than the computational cost of the\nfull order model. We also compare the new DDF-ROM with modern ROM closure\nmodels in the numerical simulation of the 1D Burgers equation. The DDF-ROM is\nmore accurate and significantly more efficient than these ROM closure models.", "category": "physics_flu-dyn" }, { "text": "Head-on collisions of dense granular jets: When a dense stream of dry, non-cohesive grains hits a fixed target, a\ncollimated sheet is ejected from the impact region, very similar to what\nhappens for a stream of water. In this study, as a continuation of the\ninvestigation why such remarkably different incident fluids produce such\nsimilar ejecta, we use discrete particle simulations to collide two\nunequal-width granular jets head-on in two dimensions. In addition to the\nfamiliar coherent ejecta, we observe that the impact produces a far less\nfamiliar quasi-steady-state corresponding to a uniformly translating free\nsurface and flow field. Upon repeating such impacts with multiple continuum\nfluid simulations, we show that this translational speed is controlled only by\nthe total energy dissipation rate to the power $1.5$, and is independent of the\ndetails of the jet composition. Our findings, together with those from impacts\nagainst fixed targets, challenge the principle of scattering in which material\ncomposition is inferred from observing the ejecta produced during impact.", "category": "physics_flu-dyn" }, { "text": "Contactless pressure measurement of an underwater shock wave in a\n microtube using a high-resolution background-oriented schlieren technique: A high-resolution background-oriented schlieren (BOS) technique, which\nutilizes a high-resolution camera and a microdot background pattern, is\nproposed and used to measure the pressure field of an underwater shock wave in\na microtube. The propagation of the shock wave subsequently reaches a concave\nwater-air interface set in the microtube resulting in the ejection of a focused\nmicrojet. This high spatial-resolution BOS technique can measure the pressure\nfield of a shock front with a width as narrow as the order of only 101 {\\mu}m\nwith a peak pressure as large as almost 3 MPa, which is significantly narrower\nand larger, respectively, than a previous study [1]. This significant\nbreakthrough has enabled the simultaneous measurement of the pressure impulse\nof the shock front and the velocity of the microjet tip. As a result, we have\nexperimentally observed the linear relation between the velocity of the\nmicrojet tip and the pressure impulse of the shock front for the cases without\nsecondary cavitation in the liquid bulk. Such relation was\ntheorectically/numerically predicted by Peters [2]. This study demonstrated the\ncapability of the proposed high-resolution BOS technique as a microscale\ncontactless pressure measurement tool for underwater shock waves and\npotentially other micro- and nano-fluids.", "category": "physics_flu-dyn" }, { "text": "The minimal seed for transition to convective turbulence in heated pipe\n flow: It is well known that buoyancy suppresses, and can even laminarise turbulence\nin upward heated pipe flow. Heat transfer seriously deteriorates in this case.\nThrough a new DNS model, we confirm that the deteriorated heat transfer within\nconvective turbulence is related to a lack of near-wall rolls, which leads to a\nweak mixing between the flow near the wall and centre of pipe. Having surveyed\nthe fundamental properties of the system, we perform a nonlinear nonmodal\nstability analysis. it is found that, the minimal seed becomes thinner and\ncloser to the wall, with increase of buoyancy number C. Most importantly, we\nshow that the critical initial energy required to trigger shear-driven\nturbulence keeps increasing, implying that attempts to artificially trigger it\nmay not be an efficient means to improve heat transfer at larger C. The new\nminimal seed, found at C=6, is localised in streamwise direction and is active\nin the centre of pipe. To find this branch of optimal, we took advantage of a\nwindow of linear stability. While the nonlinear optimal causes transition to\nconvective turbulence directly at this and larger C, transition via the linear\ninstability passes via a travelling wave or periodic orbit solutions. Detailed\nanalysis of the periodic solution reveals three stages: growth of the unstable\neigenfunction, the formation of streaks, and the decay of streaks due to\nsuppression of the instability. Flow visualization at C up to 10 also show\nsimilar features, suggesting that convective turbulence is sustained by these\nthree typical processes.", "category": "physics_flu-dyn" }, { "text": "Modal decomposition of nonlinear interactions in wall turbulence: Coherent structures are found in many different turbulent flows and are known\nto drive self-sustaining processes in wall turbulence. Identifying the triadic\ninteractions which generate coherent structures can provide insights beyond\nwhat is possible in the framework of linearized models. There are infinite\npossible interactions that may generate a given structure. Thus a method to\nsystematically study those, ranking them in terms of their contribution to the\nstructure of interest, is essential. We here use the resolvent-based extended\nspectral proper orthogonal decomposition (RESPOD) approach (Karban, U. et al.\n2022 Self-similar mechanisms in wall turbulence studied using resolvent\nanalysis. Journal of Fluid Mechanics 969, A36) to rank the triadic interactions\nwhich give rise to wall-attached structures in a minimal Couette flow at\nReynolds number 400. Our analysis identifies that six triadic interactions\ndominate the most-energetic wall-attached structure, revealing the capability\nof the methodology to identify and rank nonlinear interactions responsible for\na given coherent structure. The approach can be used to analyse the energy\nexchange in turbulent flows and may guide the construction of reduced-order\nmodels based on the interplay between different flow modes.", "category": "physics_flu-dyn" }, { "text": "Global receptivity analysis: physically realizable input-output analysis: In the context of transition analysis, linear input-output analysis\ndetermines worst-case disturbances to a laminar base flow based on a generic\nright-hand-side volumetric/boundary forcing term. The worst-case forcing is not\nphysically realizable, and, to our knowledge, a generic framework for posing\nphysically-realizable worst-case disturbance problems is lacking. In natural\nreceptivity analysis, disturbances are forced by matching (typically local)\nsolutions within the boundary layer to outer solutions consisting of\nfree-stream vortical, entropic, and acoustic disturbances. We pose a scattering\nformalism to restrict the input forcing to a set of realizable disturbances\nassociated with plane-wave solutions of the outer problem. The formulation is\nvalidated by comparing with direct numerical simulations (DNS) for a Mach 4.5\nflat-plate boundary layer. We show that the method provides insight into\ntransition mechanisms by identifying those linear combinations of plane-wave\ndisturbances that maximize energy amplification over a range of frequencies. We\nalso discuss how the framework can be extended to accommodate scattering from\nshocks and in shock layers for supersonic flow.", "category": "physics_flu-dyn" }, { "text": "Initial evolution of 3D turbulence en route to the Kolmogorov state:\n emergence and transformations of coherent structures, self-similarity and\n instabilities: In this work, we study numerically the temporal evolution of an initially\nrandom large-scale velocity field under governed by the hyperviscous\nincompressible Navier-Stoke equations. Three stages are clearly observed during\nthe evolution. First, the initial condition development is characterized by a\nspectrum evolving in a self-similar way with a wave number front $k_*(t)$\npropagating toward high values exponentially in time. This evolution\ncorresponds to the formation and shrinking of think vortex pancakes\nexponentially in time, as it has been previously reported in simulations of the\nincompressible Euler equations. At the second stage, pancakes become unstable,\nrolling up on the edges and breaking up in the middle, leading to the emergence\nof vortex ribs -- quasi-periodic arrangements of vortex filaments. Those\nfilaments then twist creating structures akin to ropes. At the last stage, a\nfully developed turbulent state is observed, characterized by a Kolmogorov\nenergy spectrum and exhibiting a decay law compatible with standard turbulent\npredictions in the case of an integral scale saturated at the scale of the box.", "category": "physics_flu-dyn" }, { "text": "Central recirculation zone in a V-shaped premixed swirling flame: This paper presents an experimental study on the emergence of the central\nrecirculation zone (CRZ) in a V-shaped premixed swirling flame, using\nsimultaneous measurement of particle image velocimetry (PIV) and CH*\nchemiluminescence. The results show that either increasing the Reynolds number\n(Re) or decreasing the equivalence ratio ({\\Phi}) would facilitate the\nemergence of CRZ. Further analysis demonstrates that the CRZ characteristics\nand its emergence are strongly influenced by the inner shear layer (ISL)\nsurrounding the CRZ, while the swirl intensity remains unchanged. Dimensional\nanalysis is performed to understand the underlying mechanism, suggesting the\nCRZ emergence is controlled by a non-dimensional parameter, Re_s=|{\\gamma}|_max\nD/{\\nu}_s, defined based on the maximum ISL intensity (|{\\gamma}|_max), the\nexit diameter (D), and the kinematic viscosity ({\\nu}_s) of the burnt gas. By\nestimating the temperature and viscosity with a simple heat-loss model, we show\nin the |{\\gamma}|_max D-{\\nu}_s regime diagram that the cases with and without\nCRZ are separated by a single boundary line, corresponding to a critical Re_s\nof about 424. This verifies the applicability of the proposed Re_s criterion to\nlean-premixed V-shaped swirling flames under various conditions. Unlike most\nprevious works that attribute the CRZ of swirling flames to vortex breakdown,\nthe present work reveals the non-negligible effect of the ISL, especially the\nCRZ suppression when the ISL is weakened by flame heating.", "category": "physics_flu-dyn" }, { "text": "Generalized Einstein relation for the mutual diffusion coefficient of a\n binary fluid mixture: The method employed by Einstein to derive his famous relation between the\ndiffusion coefficient and the friction coefficient of a Brownian particle is\nused to derive a generalized Einstein relation for the mutual diffusion\ncoefficient of a binary fluid mixture. The expression is compared with the one\nderived by de Groot and Mazur from irreversible thermodynamics, and later by\nBatchelor for a Brownian suspension. A different result was derived by several\nother workers in irreversible thermodynamics. For a nearly incompressible\nsolution the generalized Einstein relation agrees with the expression derived\nby de Groot and Mazur. The two expressions also agree to first order in solute\ndensity. For a Brownian suspension the result derived from the generalized\nSmoluchowski equation agrees with both expressions.", "category": "physics_flu-dyn" }, { "text": "Steering undulatory micro-swimmers in a fluid flow through reinforcement\n learning: This work aims at finding optimal navigation policies for thin, deformable\nmicroswimmers that progress in a viscous fluid by propagating a sinusoidal\nundulation along their slender body. These active filaments are embedded in a\nprescribed, non-homogeneous flow, in which their swimming undulations have to\ncompete with the drifts, strains, and deformations inflicted by the outer\nvelocity field. Such an intricate situation, where swimming and navigation are\ntightly bonded, is addressed using various methods of reinforcement learning.\nEach swimmer has only access to restricted information on its configuration and\nhas to select accordingly an action among a limited set. The optimisation\nproblem then consists in finding the policy leading to the most efficient\ndisplacement in a given direction. It is found that usual methods do not\nconverge and this pitfall is interpreted as a combined consequence of the\nnon-Markovianity of the decision process, together with the highly chaotic\nnature of the dynamics, which is responsible for high variability in learning\nefficiencies. Still, we provide an alternative method to construct efficient\npolicies, which is based on running several independent realisations of\nQ-learning. This allows the construction of a set of admissible policies whose\nproperties can be studied in detail and compared to assess their efficiency and\nrobustness.", "category": "physics_flu-dyn" }, { "text": "Particle resuspension: challenges and perspectives for future models: The purpose of this review is to analyze the physics at play in particle\nresuspension in order to bring insights into the rich complexity of this common\nbut challenging concern. Following the more-is-different vision, this is\nperformed by starting from a range of practical observations and experimental\ndata. We then work our way through the investigation of the key mechanisms\nwhich play a role in the overall process. In turn, these mechanisms reveal an\narray of fundamental interactions, such as particle-fluid, particle-particle\nand particle-surface, whose combined effects create the tapestry of current\napplications. At the core of this analysis are descriptions of these physical\nphenomena and the different ways through which they are intertwined to build up\nvarious models used to provide quantitative assessment of particle\nresuspension. The physics of particle resuspension implies to hold together\nprocesses occurring at extremely different space and time scales and models are\nkey in providing a single vehicle to lead us through such multiscale journeys.\nThis raises questions on what makes up a model and one objective of the present\nwork is to clarify the essence of a modeling approach. In spite of its\nubiquitous nature, particle resuspension is still at the early stages of\ndevelopments. Many extensions need to be worked out and revisiting the art of\nmodeling is not a moot point. The need to consider more complex objects than\nsmall and spherical particles and, moreover, to come up with unified\ndescriptions of mono- and multilayer resuspension put the emphasis on solid\nmodel foundations if we are to go beyond current limits. This is very much\nmodeling in the making and new ideas are proposed to stimulate interest into\nthis everyday but challenging issue in physics.", "category": "physics_flu-dyn" }, { "text": "Stick-slip-to-stick transition of liquid oscillation in a U-tube: The nonlinear decay of oscillations of a liquid column in a U-shaped tube is\ninvestigated within the theoretical framework of the projection method\nformalized by Bongarzone et al. (2021) [1]. Starting from the full hydrodynamic\nsystem supplemented by a phenomenological contact line model, this\nphysics-inspired method uses successive linear eigenmode projections to\nsimulate the relaxation dynamics of liquid oscillations in the presence of\nsliding triple lines. Each projection is shown to eventually induce a rapid\nloss of total energy in the liquid motion, thus contributing to its nonlinear\ndamping. A thorough quantitative comparison with experiments by Dollet et al.\n(2020) [2] demonstrates that, in contradistinction with their simplistic\none-degree-of-freedom model, the present approach not only describes well the\ntransient stick-slip dynamics, but it also correctly captures the global\nstick-slip to stick transition, as well as the secondary bulk motion following\nthe arrest of the contact line, which has been so far overlooked by existing\ntheoretical analyses. This study offers a further contribution to rationalizing\nthe impact of contact angle hysteresis and its associated solidlike friction on\nthe decay of liquid oscillations in the presence of sliding triple lines.", "category": "physics_flu-dyn" }, { "text": "Hydrodynamic Singularities: We give a brief overview of the physical significance of singularities in\nfluid mechanics.", "category": "physics_flu-dyn" }, { "text": "Field correlations and the ultimate regime of turbulent convection: Using direct numerical simulations of Rayleigh-B\\'{e}nard convection (RBC)\nunder free-slip boundary condition, we show that the normalized correlation\nfunction between the vertical velocity field and the temperature field, as well\nas the normalized viscous dissipation rate, scales as $Ra^{-0.22}$ for\nmoderately large Rayleigh number $Ra$. This scaling accounts for the Nusselt\nnumber ($Nu$) exponent to be around 0.3 observed in experiments. Numerical\nsimulations also reveal that the above normalized correlation functions are\nconstants for the convection simulation under periodic boundary conditions.", "category": "physics_flu-dyn" }, { "text": "The influence of near-wall density and viscosity gradients on turbulence\n in channel flows: The influence of near-wall density and viscosity gradients on near-wall\nturbulence in a channel are studied by means of Direct Numerical Simulation\n(DNS) of the low-Mach number approximation of the Navier--Stokes equations.\nDifferent constitutive relations for density and viscosity as a function of\ntemperature are used in order to mimic a wide range of fluid behaviours and to\ndevelop a generalised framework for studying turbulence modulations in variable\nproperty flows. Instead of scaling the velocity solely based on local density,\nas done for the van Driest transformation, we derive an extension of the\nscaling that is based on gradients of the semi-local Reynolds number\n$Re_\\tau^*$. This extension of the van Driest transformation is able to\ncollapse velocity profiles for flows with near-wall property gradients as a\nfunction of the semi-local wall coordinate. However, flow quantities like\nmixing length, turbulence anisotropy and turbulent vorticity fluctuations do\nnot show a universal scaling very close to the wall. This is attributed to\nturbulence modulations, which play a crucial role on the evolution of turbulent\nstructures and turbulence energy transfer. We therefore investigate the\ncharacteristics of streamwise velocity streaks and quasi-streamwise vortices\nand found that, similar to turbulent statistics, the turbulent structures are\nalso strongly governed by $Re_\\tau^*$ profiles and that their dependence on\nindividual density and viscosity profiles is minor. Flows with near-wall\ngradients in $Re_\\tau^*$ ($d {Re_\\tau^*}/dy \\neq 0$) showed significant changes\nin the inclination and tilting angles of quasi-streamwise vortices. These\nstructural changes are responsible for the observed modulation of the Reynolds\nstress generation mechanism and the inter-component energy transfer in flows\nwith strong near-wall $Re_\\tau^*$ gradients.", "category": "physics_flu-dyn" }, { "text": "Direct numerical simulation of a moist cough flow using Eulerian\n approximation for liquid droplets: The COVID-19 pandemic has inspired several studies on the fluid dynamics of\nrespiratory events. Here, we propose a computational approach in which\nrespiratory droplets are coarse-grained into an Eulerian liquid field advected\nby the fluid streamlines. A direct numerical simulation is carried out for a\nmoist cough using a closure model for space-time dependence of the evaporation\ntime scale. Estimates of the Stokes number are provided, for the initial\ndroplet size of $10 \\mu$m, which are found to be <<1 thereby justifying the\nneglect of droplet inertia. Several of the important features of the\nmoist-cough flow reported in the literature using Lagrangian tracking methods\nhave been accurately captured using our scheme. Some new results are presented,\nincluding the evaporation time for a \"mild\" cough, a saturation-temperature\ndiagram and a favourable correlation between the vorticity and liquid fields.\nThe present approach is particularly useful for studying the long-range\ntransmission of virus-laden droplets.", "category": "physics_flu-dyn" }, { "text": "Pressure fluctuations of liquids under short-time acceleration: This study experimentally investigates the pressure fluctuations of liquids\nin a column under short-time acceleration and demonstrates that the Strouhal\nnumber $St$ [$=L/(c\\Delta t)$, where $L$, $c$, and $\\Delta t$ are the liquid\ncolumn length, speed of sound, and acceleration duration, respectively]\nprovides a measure of the pressure fluctuations both for limiting cases (i.e.\n$St\\ll1$ or $St = \\infty$) and for intermediate $St$ values. Incompressible\nfluid theory and water hammer theory respectively imply that the magnitude of\nthe averaged pressure fluctuation $\\overline{P}$ becomes negligible for\n$St\\ll1$ (i.e., in the condition where the duration of acceleration $\\Delta t$\nis large enough compared to the acoustic timescale) and tends to $\\rho cu_0$\n(where $u_0$ is the change in the liquid velocity) for $St\\geq O(1)$ (i.e., in\nthe condition where $\\Delta t$ is small enough). For intermediate $St$ values,\nthere is no consensus on the value of $\\overline{P}$. In our experiments, $L$,\n$c$, and $\\Delta t$ are varied so that $0.02 \\leq St \\leq 2.2$. The results\nsuggest that the incompressible fluid theory holds only up to $St\\sim0.2$ and\nthat $St$ governs the pressure fluctuations under different experimental\nconditions for higher $St$ values. The data relating to a hydrogel also tend to\ncollapse to a unified trend. The inception of cavitation in the liquid starts\nat $St\\sim 0.2$ for various $\\Delta t$, indicating that the liquid pressure\nbecomes negative. To understand this mechanism, we employ a one-dimensional\nwave propagation model with a pressure wavefront of finite thickness that\nscales with $\\Delta t$. The model provides a reasonable description of the\nexperimental results as a function of $St$. The slight discrepancy between the\nmodel and experimental results reveals additional contributing factors such as\nthe container motion and the profile of the pressure wavefront.", "category": "physics_flu-dyn" }, { "text": "Two-point velocity average of turbulence: statistics and their\n implications: For turbulence, although the two-point velocity difference u(x+r)-u(x) at\neach scale r has been studied in detail, the velocity average [u(x+r)+u(x)]/2\nhas not thus far. Theoretically or experimentally, we find interesting features\nof the velocity average. It satisfies an exact scale-by-scale energy budget\nequation. The flatness factor varies with the scale r in a universal manner.\nThese features are not consistent with the existing assumption that the\nvelocity average is independent of r and represents energy-containing\nlarge-scale motions alone. We accordingly propose that it represents motions\nover scales >= r as long as the velocity difference represents motions at the\nscale r.", "category": "physics_flu-dyn" }, { "text": "Spatters and Spills: Spreading Dynamics for Partially Wetting Droplets: We present a solvable model inspired by dimensional analysis for the\ntime-dependent spreading of droplets that partially wet a substrate, where the\nspreading eventually stops and the contact angle reaches a nonzero equilibrium\nvalue. We separately consider small droplets driven by capillarity and large\ndroplets driven by gravity. To explore both regimes, we first measure the\nequilibrium radius versus a comprehensive range of droplet volumes for four\nhousehold fluids, and we compare the results with predictions based on\nminimizing the sum of gravitational and interfacial energies. The agreement is\ngood, and gives a reliable measurement of an equilibrium contact angle that is\nconsistent in both small and large droplet regimes. Next we use energy\nconsiderations to develop equations of motion for the time dependence of the\nspreading, in both regimes, where the driving forces are balanced against\nviscous drag in the bulk of the droplet and by friction at the moving contact\nline. Our approach leads to explicit prediction of the functional form of the\nspreading dynamics. It successfully describes prior data for a small\ncapillary-driven droplet, and it fits well to new data we obtain for large\ngravity-driven droplets with a wide range of volumes.", "category": "physics_flu-dyn" }, { "text": "Predicting the impact of particle-particle collisions on turbophoresis\n with a reduced number of computational particles: A common feature of wall-bounded turbulent particle-laden flows is enhanced\nparticle concentrations in a thin layer near the wall due to a phenomenon known\nas turbophoresis. Even at relatively low bulk volume fractions,\nparticle-particle collisions regulate turbophoresis in a critical way, making\nsimulations sensitive to collisional effects. Lagrangian tracking of every\nparticle in the flow can become computationally expensive when the physical\nnumber of particles in the system is large. Artificially reducing the number of\nparticles in the simulation can mitigate the computational cost. When\nparticle-particle collisions are an important aspect determining the simulation\noutcome, as in the case when turbophoresis plays an active role, simply\nreducing the number of particles in the simulation significantly alters the\ncomputed particle statistics. This paper introduces a computational particle\ntreatment for particle-particle collisions which reproduces the results of a\nfull simulation with a reduced number of particles. This is accomplished by\nartificially enhancing the particle collision radius based on scaling laws for\nthe collision rates. The proposed method retains the use of deterministic\ncollision models and is applicable for both low and high Stokes number regimes.", "category": "physics_flu-dyn" }, { "text": "How long will a bubble be ?: A soap bubble is a metastable object that eventually breaks. Indeed, the\nsoapy water film thins until rupture, due to drainage and evaporation. In our\nexperimental investigations, floating bubbles at the surface of a liquid bath\nhave been considered. Their lifetime has been measured and reported with\nrespect to their radius. Large bubbles last longer than small ones. Moreover,\nsmall bubbles have more predictable lifetimes than large ones. We propose a\ngeneral equation for that lifetime, based on the lubrication theory. The\nevaporation is shown to be an essential process which determines the bubble\nlifetime.", "category": "physics_flu-dyn" }, { "text": "Segment-Based Wall Treatment Model for Heat Transfer Rate in Smoothed\n Particle Hydrodynamics: In this study, a smoothed particle hydrodynamics (SPH) model that applies a\nsegment-based boundary treatment is used to simulate natural convection. In a\nnatural convection simulated using an SPH model, the wall boundary treatment is\na major issue because accurate heat transfer from boundaries should be\ncalculated. The boundary particle method, which models the boundary by placing\nmultiple layers of particles on and behind the wall boundary, is the most\nwidely used boundary treatment method. Although this method can impose accurate\nboundary conditions, boundary modeling for complex shapes is challenging and\nrequires excessive computational costs depending on the boundary shape. In this\nstudy, we utilize a segment-based boundary treatment method to model the wall\nboundary and apply this method to the energy conservation equation for the wall\nheat transfer model. The proposed method solves the problems arising from the\nuse of boundary particles and simultaneously provides accurate heat transfer\ncalculation results for the wall. In various numerical examples, the proposed\nmethod is verified through a comparison with available experimental results,\nSPH results using the boundary particle method, and finite volume method (FVM)\nresults.", "category": "physics_flu-dyn" }, { "text": "Launching Drifter Observations in the Presence of Uncertainty: Determining the optimal locations for placing extra observational\nmeasurements has practical significance. However, the exact underlying flow\nfield is never known in practice. Significant uncertainty appears when the flow\nfield is inferred from a limited number of existing observations via data\nassimilation or statistical forecast. In this paper, a new computationally\nefficient strategy for deploying Lagrangian drifters that highlights the\ncentral role of uncertainty is developed. A nonlinear trajectory diagnostic\napproach that underlines the importance of uncertainty is built to construct a\nphase portrait map. It consists of both the geometric structure of the\nunderlying flow field and the uncertainty in the estimated state from\nLagrangian data assimilation. The drifters are deployed at the maxima of this\nmap and are required to be separated enough. Such a strategy allows the\ndrifters to travel the longest distances to collect both the local and global\ninformation of the flow field. It also facilitates the reduction of a\nsignificant amount of uncertainty. To characterize the uncertainty, the\nestimated state is given by a probability density function (PDF). An\ninformation metric is then introduced to assess the information gain in such a\nPDF, which is fundamentally different from the traditional path-wise\nmeasurements. The information metric also avoids using the unknown truth to\nquantify the uncertainty reduction, making the method practical. Mathematical\nanalysis exploiting simple illustrative examples is used to validate the\nstrategy. Numerical simulations based on multiscale turbulent flows are then\nadopted to demonstrate the advantages of this strategy over some other methods.", "category": "physics_flu-dyn" }, { "text": "Effect of viscosity on the dynamics of a non-equilibrium bubble in\n free-field and near a free-surface: The effect of viscosity on the behaviour of a non-equilibrium bubble is\ninvestigated experimentally, in two scenarios; firstly, when the bubble is\ngenerated in the bulk of the fluid (termed as ``free-field'' bubble) and\nsecondly when the bubble is generated near a free-surface (termed as\n``free-surface'' bubble). The bubble is created using a low-voltage spark\ncircuit and its dynamics is captured using a high-speed camera with back-lit\nillumination. The viscosity of the surrounding fluid is varied by using\ndifferent grades of silicone oil. For a ``free-field'' bubble, the bubble\noscillates radially and as the viscosity of the liquid increases, the number of\noscillations, as well as the time-period of each oscillation, are increased. At\nhigh viscosities, the bubble also becomes stable and does not disintegrate into\nsmaller bubbles. For ``free-surface'' bubbles, two parameters, namely, the\ninitial distance of the bubble from the free-surface and the viscosity of the\nsurrounding fluid are varied. It is observed that beyond a certain initial\ndistance of the bubble from the free-surface, the bubble behaves as a\n``free-field'' bubble with negligible influence of the free-surface on its\ndynamics. This limiting initial distance decreases as the liquid viscosity is\nincreased and is not dependent on the bubble radius. For these bubbles,\ndifferent behaviours of the free-surface in each liquid are also presented as a\nfunction of the two parameters.", "category": "physics_flu-dyn" }, { "text": "Logarithmic scaling for fluctuations of a scalar concentration in wall\n turbulence: Within wall turbulence, there is a sublayer where the mean velocity and the\nvariance of velocity fluctuations vary logarithmically with the height from the\nwall. This logarithmic scaling is also known for the mean concentration of a\npassive scalar. By using heat as such a scalar in a laboratory experiment of a\nturbulent boundary layer, the existence of the logarithmic scaling is shown\nhere for the variance of fluctuations of the scalar concentration. It is\nreproduced by a model of energy-containing eddies that are attached to the\nwall.", "category": "physics_flu-dyn" }, { "text": "Dynamics of two air bubbles rising in a shear-thinning fluid: In this paper, we have studied the three-dimensional dynamics of two equally\nsized air bubbles rising in a shear-thinning fluid. We have used the combined\nlevel set and volume of fluid (CLSVOF) method to track interface, maintain mass\nbalance and estimate the interface curvature. Additionally, we have\nincorporated a Sharp Surface Force Method (SSF) for surface tension forces.\nThis method significantly suppressed the spurious velocities commonly observed\nwith the conventional volume of fluid method and the Continuum Surface Force\n(CSF). The algorithm is implemented in an in-house code called CUFLOW and runs\non multiple GPUs platform.\n We have explored the effects of fluid rheology on the three-dimensional\ndynamics of two in-line bubbles. Two power-law indices (0.5 and 1) are\ninvestigated to highlight differences in shear-thinning and Newtonian fluids.\nFor a range of parameters examined here, bubbles motion in a shear-thinning\nfluid is seen to be unsteady with significant shape oscillations. Further, we\nhave examined the rise velocity, droplets rise path, transient shapes and found\nthat modification of viscosity by the motion of the leading bubble changes the\ndynamics of the trailing bubble.", "category": "physics_flu-dyn" }, { "text": "Dynamic Mode Decomposition for Aero-Optic Wavefront Characterization: Aero-optical beam control relies on the development of low-latency\nforecasting techniques to quickly predict wavefronts aberrated by the Turbulent\nBoundary Layer (TBL) around an airborne optical system, and its study applies\nto a multi-domain need from astronomy to microscopy for high-fidelity laser\npropagation. We leverage the forecasting capabilities of the Dynamic Mode\nDecomposition (DMD) -- an equation-free, data-driven method for identifying\ncoherent flow structures and their associated spatiotemporal dynamics -- in\norder to estimate future state wavefront phase aberrations to feed into an\nadaptive optic (AO) control loop. We specifically leverage the optimized DMD\n(opt-DMD) algorithm on a subset of the Airborne Aero-Optics Laboratory\nTransonic (AAOL-T) experimental dataset, characterizing aberrated wavefront\ndynamics for 23 beam propagation directions via the spatiotemporal\ndecomposition underlying DMD. Critically, we show that opt-DMD produces an\noptimally de-biased eigenvalue spectrum with imaginary eigenvalues, allowing\nfor arbitrarily long forecasting to produce a robust future-state prediction,\nwhile exact DMD loses structural information due to modal decay rates.", "category": "physics_flu-dyn" }, { "text": "Viscous to Inertial Crossover in Liquid Drop Coalescence: Using an electrical method and high-speed imaging we probe drop coalescence\ndown to 10 ns after the drops touch. By varying the liquid viscosity over two\ndecades, we conclude that at sufficiently low approach velocity where\ndeformation is not present, the drops coalesce with an unexpectedly late\ncrossover time between a regime dominated by viscous and one dominated by\ninertial effects. We argue that the late crossover, not accounted for in the\ntheory, can be explained by an appropriate choice of length-scales present in\nthe flow geometry.", "category": "physics_flu-dyn" }, { "text": "Review of the Onsager \"Ideal Turbulence\" Theory: In his famous undergraduate physics lectures, Richard Feynman remarked about\nthe problem of fluid turbulence: \"Nobody in physics has really been able to\nanalyze it mathematically satisfactorily in spite of its importance to the\nsister sciences\". This statement was already false when Feynman made it.\nUnbeknownst to him, Lars Onsager decades earlier had made an exact mathematical\nanalysis of the high Reynolds-number limit of incompressible fluid turbulence,\nusing a method that would now be described as a non-perturbative\nrenormalization group analysis and discovering the first \"conservation-law\nanomaly\" in theoretical physics. Onsager's results were only cryptically\nannounced in 1949 and he never published any of his detailed calculations.\nOnsager's analysis was finally rescued from oblivion and reproduced by this\nauthor in 1992. The ideas have subsequently been intensively developed in the\nmathematical PDE community, where deep connections emerged with John Nash's\nwork on isometric embeddings. Furthermore, Onsager's method has more recently\nbeen successfully applied to new physics problems, such as compressible fluid\nturbulence and relativistic fluid turbulence, yielding many novel testable\npredictions. This note will explain Onsager's exact analysis of incompressible\nturbulence using modern ideas on renormalization group and conservation-law\nanomalies, and it will also very briefly review subsequent developments.", "category": "physics_flu-dyn" }, { "text": "Observation of nonlinear dispersion relation and spatial statistics of\n wave turbulence on the surface of a fluid: We report experiments on gravity-capillary wave turbulence on the surface of\na fluid. The wave amplitudes are measured simultaneously in time and space\nusing an optical method. The full space-time power spectrum shows that the wave\nenergy is localized on several branches in the wave-vector-frequency space. The\nnumber of branches depend on the power injected within the waves. The\nmeasurement of the nonlinear dispersion relation is found to be well described\nby a law suggesting that the energy transfer mechanisms involved in wave\nturbulence are not only restricted to purely resonant interaction between\nnonlinear waves. The power-law scaling of the spatial spectrum and the\nprobability distribution of the wave amplitudes at a given wave number are also\nmeasured and compared to the theoretical predictions.", "category": "physics_flu-dyn" }, { "text": "Recursive one-way Navier Stokes equations with PSE-like cost: Spatial marching methods, in which the flow state is spatially evolved in the\ndownstream direction, can be used to produce low-cost models of flows\ncontaining a slowly varying direction, such as mixing layers, jets, and\nboundary layers. The parabolized stability equations (PSE) are popular due to\ntheir extremely low cost but can only capture a single instability mode; all\nother modes are damped or distorted by regularization methods required to\nstabilize the spatial march, precluding PSE from properly capturing non-modal\nbehavior, acoustics, and interactions between multiple instability mechanisms.\nThe one-way Navier-Stokes (OWNS) equations properly retain all\ndownstream-traveling modes at a cost that is a fraction of that of global\nmethods but still one to two orders of magnitude higher than PSE. In this\npaper, we introduce a new variant of OWNS whose cost, both in terms of CPU time\nand memory requirements, approaches that of PSE while still properly capturing\nthe contributions of all downstream-traveling modes. The method is formulated\nin terms of a projection operator that eliminates upstream-traveling modes.\nUnlike previous OWNS variants, the action of this operator on a vector can be\nefficiently approximated using a series of equations that can be solved\nrecursively, i.e., successively one after the next, rather than as a coupled\nset. In addition to highlighting the improved cost scaling of our method, we\nderive explicit error expressions and elucidate the relationship with previous\nOWNS variants. The properties, efficiency, and accuracy of our method are\ndemonstrated for both free-shear and wall-bounded flows.", "category": "physics_flu-dyn" }, { "text": "The effect of finite container size on granular jet formation: When an object is dropped into a bed of fine, loosely packed sand, a\nsurprisingly energetic jet shoots out of the bed. In this work we study the\neffect that boundaries have on the granular jet formation. We did this by (i)\ndecreasing the depth of the sand bed and (ii) reducing the container diameter\nto only a few ball diameters. These confinements change the behavior of the\nball inside the bed, the void collapse, and the resulting jet height and shape.\nWe map the parameter space of impact with Froude number, ambient pressure, and\ncontainer dimensions as parameters. From these results we propose a new\nexplanation for the thick-thin structure of the jet reported by several groups\n[J.R.Royer \\textit{et al}., Nature Phys. \\textbf{1}, 164 (2005)], [G.Caballero\n\\textit{et al}., Phys. Rev. Lett. \\textbf{99}, 018001 (2007)] and [J.O.\nMarston, \\textit{et al}., Physics of Fluids \\textbf{20}, 023301 (2008)].", "category": "physics_flu-dyn" }, { "text": "Thermoelectricity in confined liquid electrolytes: The electric field in an extended phase of a liquid electrolyte exposed to a\ntemperature gradient is attributed to different thermophoretic mobilities of\nthe ion species. As shown herein, such Soret-type ion thermodiffusion is not\nrequired to induce thermoelectricity even in the simplest electrolyte if it is\nconfined between charged walls. The space charge of the electric double layer\nleads to selective ion diffusion driven by a temperature-dependent\nelectrophoretic ion mobility, which -- for narrow channels -- may cause\nthermo-voltages larger in magnitude than for the classical Soret equilibrium.", "category": "physics_flu-dyn" }, { "text": "Flow and deformation due to periodic loading in a soft porous material: Soft porous materials, such as biological tissues and soils, are exposed to\nperiodic deformations in a variety of natural and industrial contexts. The\ndetailed flow and mechanics of these deformations have not yet been\nsystematically investigated. Here, we fill this gap by identifying and\nexploring the complete parameter space associated with periodic deformations in\nthe context of a 1D model problem. We use large-deformation poroelasticity to\nconsider a wide range of loading periods and amplitudes. We identify two\ndistinct mechanical regimes, distinguished by whether the loading period is\nslow or fast relative to the characteristic poroelastic timescale. We develop\nanalytical solutions for slow loading at any amplitude and for infinitesimal\namplitude at any period. We use these analytical solutions and a full numerical\nsolution to explore the localisation of the deformation near the permeable\nboundary as the period decreases and the emergence of nonlinear effects as the\namplitude increases. We show that large deformations lead to asymmetry between\nthe loading and unloading phases of each cycle in terms of the distributions of\nporosity and fluid flux.", "category": "physics_flu-dyn" }, { "text": "Active spheroids in viscosity gradients: In this paper, we explore the hydrodynamics of spheroidal active particles in\nviscosity gradients. This work provides a more accurate modeling approach, in\ncomparison to spherical particles, for anisotropic organisms like Paramecium\nswimming through inhomogeneous environments, but more fundamentally examines\nthe influence of particle shape on viscotaxis. We find that spheroidal\nsquirmers generally exhibit dynamics consistent with their spherical analogs,\nirrespective of the classification of swimmers as pushers, pullers, or neutral\nswimmers. However, the slenderness of the spheroids tends to reduce the impact\nof viscosity gradients on their dynamics; when swimmers become more slender,\nthe viscosity difference across their body is reduced, which leads to slower\nreorientation. We also derive the mobility tensor for passive spheroids in\nviscosity gradients generalizing previous results for spheres and slender\nbodies. This work enhances our understanding of how shape factors into the\ndynamics of passive and active particles in viscosity gradients, and offers new\nperspectives that could aid the control of both natural and synthetic swimmers\nin complex fluid environments.", "category": "physics_flu-dyn" }, { "text": "A Magnetically and Electrically Powered Hybrid Micromotor in Conductive\n Solutions: Synergistic Propulsion Effects and Label-Free Cargo Transport and\n Sensing: Electrically powered micro- and nanomotors are promising tools for in-vitro\nsingle-cell analysis. In particular, single cells can be trapped, transported\nand electroporated by a Janus particle (JP) using an externally applied\nelectric field. However, while dielectrophoretic (DEP)-based cargo manipulation\ncan be achieved at high-solution conductivity, electrical propulsion of these\nmicromotors becomes ineffective at solution conductivities exceeding 0.3mS/cm.\nHere, we successfully extended JP cargo manipulation and transport capabilities\nto conductive near-physiological (<6mS/cm) solutions by combining magnetic\nfield-based micromotor propulsion and navigation with DEP-based manipulation of\nvarious synthetic and biological cargos. Combination of a rotating magnetic\nfield and electric field resulted in enhanced micromotor mobility and steering\ncontrol through tuning of the electric field frequency. conditions are\nnecessary. In addition, we demonstrated the micromotors ability of identifying\napoptotic cell among viable and necrotic cells based their dielectrophoretic\ndifference, thus, enabling to analyze the apoptotic status in the single cell\nsamples for drug discovery, cell therapeutics and immunotherapy. We also\ndemonstrated the ability to trap and transport live cells towards regions\ncontaining doxorubicin-loaded liposomes. This hybrid micromotor approach for\nlabel-free trapping, transporting and sensing of selected cells within\nconductive solutions, opens new opportunities in drug delivery and single cell\nanalysis, where close-to-physiological media", "category": "physics_flu-dyn" }, { "text": "Visualization of Kelvin waves on quantum vortices: In superfluid helium, vorticity is quantized and constrained on line-like\nphase singularities, called quantum or quantized vortices. By visualizing the\nmotion of sub-micron frozen particles in superfluid $^{4}$He, we directly\nobserve for the first time the helical Kelvin waves excited after quantized\nvortex reconnections. We compare the data with self-similar solutions of vortex\nfilament models. We report the results in a fluid dynamic video.", "category": "physics_flu-dyn" }, { "text": "Water wave propagation and scattering over topographical bottoms: Here I present a general formulation of water wave propagation and scattering\nover topographical bottoms. A simple equation is found and is compared with\nexisting theories. As an application, the theory is extended to the case of\nwater waves in a column with many cylindrical steps.", "category": "physics_flu-dyn" }, { "text": "Lift and drag forces acting on a particle moving in the presence of slip\n and shear near a wall: The lift and drag forces acting on a small spherical particle moving with a\nfinite slip in single-wall-bounded flows are investigated via direct numerical\nsimulations. The effect of slip velocity on the particle force is analysed as a\nfunction of separation distance for low slip and shear Reynolds numbers\n($10^{-3} \\leq Re_{\\gamma}, Re_{\\text{slip}} \\leq 10^{-1}$) in both quiescent\nand linear shear flows. A generalised lift model valid for arbitrary\nparticle-wall separation distances and $Re_{\\gamma}, Re_{\\text{slip}} \\leq\n10^{-1}$ is developed based on the results of the simulations. The proposed\nmodel can now predict the lift forces in linear shear flows in the presence or\nabsence of slip,and in quiescent flows when slip is present. Existing drag\nmodels are also compared with numerical results for both quiescent and linear\nshear flows to determine which models capture near wall slip velocities most\naccurately for low particle Reynolds numbers. Finally, we compare the results\nof the proposed lift model to previous experimental results of buoyant\nparticles and to numerical results of neutrally-buoyant (force-free) particles\nmoving near a wall in quiescent and linear shear flows. The generalised lift\nmodel presented can be used to predict the behaviour of particle suspensions in\nbiological and industrial flows where the particle Reynolds numbers based on\nslip and shear are $\\mathcal{O}(10^{-1})$ and below.", "category": "physics_flu-dyn" }, { "text": "Two-layer Thermally Driven Turbulence: Mechanisms for Interface Breakup: It is commonly accepted that the breakup criteria of drops or bubbles in\nturbulence is governed by surface tension and inertia. However, also\n{\\it{buoyancy}} can play an important role at breakup. In order to better\nunderstand this role, here we numerically study Rayleigh-B\\'enard convection\nfor two immiscible fluid layers, in order to identify the effects of buoyancy\non interface breakup. We explore the parameter space spanned by the Weber\nnumber $5\\leq We \\leq 5000$ (the ratio of inertia to surface tension) and the\ndensity ratio between the two fluids $0.001 \\leq \\Lambda \\leq 1$, at fixed\nRayleigh number $Ra=10^8$ and Prandtl number $Pr=1$. At low $We$, the interface\nundulates due to plumes. When $We$ is larger than a critical value, the\ninterface eventually breaks up. Depending on $\\Lambda$, two breakup types are\nobserved: The first type occurs at small $\\Lambda \\ll 1$ (e.g. air-water\nsystems) when local filament thicknesses exceed the Hinze length scale. The\nsecond, strikingly different, type occurs at large $\\Lambda$ with roughly $0.5\n< \\Lambda \\le 1$ (e.g. oil-water systems): The layers undergo a periodic\noverturning caused by buoyancy overwhelming surface tension. For both types the\nbreakup criteria can be derived from force balance arguments and show good\nagreement with the numerical results.", "category": "physics_flu-dyn" }, { "text": "Physics of Spin Casting Dilute Solutions: We analyze the evolution of the vertical composition profile during\nhydrodynamic-evaporative film thinning as it typically occurs during spin\ncasting mixtures of non-volatile solutes and volatile solvents. We assume that\nthe solvent dominates the hydrodynamic-evaporative film thinning. The internal\nspatio-temporal evolution of the composition is analyzed with a\ndiffusive-advective approach. The analysis provides transparent physical\ninsights into the influence of the experimental conditions on the evolution of\nthe internal composition. We present power laws that link the process control\nparameters to the composition evolution, process duration, and final solute\ncoverage. The analysis reveals a characteristic Sherwood Number as fundamental\nprocess parameter. It identifies for which stages of the process our analysis\nis quantitatively relevant and discloses the dominance of either diffusion or\nevaporation. The analysis is valid for dilute solutions e.g., for the\ndeposition of solute (sub)monolayers. But it is also relevant for the\ndeposition of thicker (polymer) films.", "category": "physics_flu-dyn" }, { "text": "High fidelity fluid-structure interaction by radial basis functions mesh\n adaption of moving walls: a workflow applied to an aortic valve: Fluid-Structure Interaction (FSI) can be investigated by means of non-linear\nFinite Element Models (FEM), suitable to capture large deflections of\nstructural parts interacting with fluids, and Computational Fluid Dynamics\n(CFD). High fidelity simulations are obtained using the fine spatial resolution\nof both the structural and fluid computational grids. A key enabler to have a\nproper exchange of information between the structural solver and the fluid one\nis the management of the interface at wetted surfaces where the grids are\nusually non matching. A class of applications, known also as one-way FSI\nproblems, involves a complex movement of the walls that is known in advance as\nmeasured or as computed by FEM, and that has to be imposed at the boundaries\nduring a transient CFD solution. Effective methods for the time marching\nadaption of the whole computational grid of the CFD model according to the\nevolving shape of its boundaries are required. A very well established approach\nconsists of a continuum update of the mesh that is regenerated by adding and\nremoving cells to fit the evolution of the moving walls. In this paper, an\ninnovative method based on Radial Basis Functions (RBF) mesh morphing is\nproposed, allowing the retention of the same mesh topology suitable for a\ncontinuum update of the shape. The proposed method is exact at a set of given\nkey configurations and relies on shape blending time interpolation between key\nframes. The study of the complex motion of a Polymeric-Prosthetic Heart Valve\n(P-PHV) is presented using the new framework and considering as a reference the\nestablished approach based on remeshing.", "category": "physics_flu-dyn" }, { "text": "Swimming near Deformable Membranes at Low Reynolds Number: Microorganisms are rarely found in Nature swimming freely in an unbounded\nfluid. Instead, they typically encounter other organisms, hard walls, or\ndeformable boundaries such as free interfaces or membranes. Hydrodynamic\ninteractions between the swimmer and nearby objects lead to many interesting\nphenomena, such as changes in swimming speed, tendencies to accumulate or turn,\nand coordinated flagellar beating. Inspired by this class of problems, we\ninvestigate locomotion of microorganisms near deformable boundaries. We\ncalculate the speed of an infinitely long swimmer close to a flexible surface\nseparating two fluids; we also calculate the deformation and swimming speed of\nthe flexible surface. When the viscosities on either side of the flexible\ninterface differ, we find that fluid is pumped along or against the swimming\ndirection, depending on which viscosity is greater.", "category": "physics_flu-dyn" }, { "text": "Turbulence of capillary waves revisited: Kinetic regime of capillary wave turbulence is classically regarded in terms\nof three-wave interactions with the exponent of power energy spectrum being\n$\\nu=-7/4$ (two-dimensional case). We show that a number of assumptions\nnecessary for this regime to occur can not be fulfilled. Four-wave interactions\nof capillary waves should be taken into account instead, which leads to\nexponents $\\nu=-13/6$ and $\\nu=-3/2$ for one- and two-dimensional wavevectors\ncorrespondingly. It follows that for general dispersion functions of decay\ntype, three-wave kinetic regime need not prevail and higher order resonances\nmay play a major role.\n Accepted for publication.", "category": "physics_flu-dyn" }, { "text": "A method of immersed layers on Cartesian grids, with application to\n incompressible flows: The immersed boundary method (IBM) of Peskin (J. Comput. Phys., 1977), and\nderived forms such as the projection method of Taira and Colonius (J. Comput.\nPhys., 2007), have been useful for simulating flow physics in problems with\nmoving interfaces on stationary grids. However, in their interface treatment,\nthese methods do not distinguish one side from the other, but rather, apply the\nmotion constraint to both sides, and the associated interface force is an\ninseparable mix of contributions from each side. In this work, we define a\ndiscrete Heaviside function, a natural companion to the familiar discrete Dirac\ndelta function (DDF), to define a masked version of each field on the grid\nwhich, to within the error of the DDF, takes the intended value of the field on\nthe respective sides of the interface. From this foundation we develop discrete\noperators and identities that are uniformly applicable to any surface geometry.\nWe use these to develop extended forms of prototypical partial differential\nequations, including Poisson, convection-diffusion, and incompressible\nNavier-Stokes, that govern the discrete masked fields. These equations contain\nthe familiar forcing term of the IBM, but also additional terms that regularize\nthe jumps in field quantities onto the grid and enable us to individually\nspecify the constraints on field behavior on each side of the interface.\nDrawing the connection between these terms and the layer potentials in elliptic\nproblems, we refer to them generically as immersed layers. We demonstrate the\napplication of the method to several representative problems, including\ntwo-dimensional incompressible flows inside a rotating cylinder and external to\na rotating square.", "category": "physics_flu-dyn" }, { "text": "Kelvin wake pattern at small Froude numbers: The surface gravity wave pattern that forms behind a steadily moving\ndisturbance is well known to comprise divergent waves and transverse waves,\ncontained within a distinctive V-shaped wake. In this paper, we are concerned\nwith a theoretical study of the limit of a slow-moving disturbance (small\nFroude numbers), for which the wake is dominated by transverse waves. We\nconsider three configurations: flow past a submerged source singularity, a\nsubmerged doublet, and a pressure distribution applied to the surface. We treat\nthe linearised version of these problems and use the method of stationary phase\nand exponential asymptotics to demonstrate that the apparent wake angle is less\nthan the classical Kelvin angle and to quantify the decrease in apparent wake\nangle as the Froude number decreases. These results complement a number of\nrecent studies for sufficiently fast-moving disturbances (large Froude numbers)\nwhere the apparent wake angle has been also less than the classical Kelvin\nangle. As well as shedding light on the wake angle, we also study the fully\nnonlinear problems for our three configurations under various limits to\ndemonstrate the unique and interesting features of Kelvin wake patterns at\nsmall Froude numbers.", "category": "physics_flu-dyn" }, { "text": "Experimental studies of liquid-liquid dispersion in a turbulent shear\n flow: We study liquid-liquid dispersions in a turbulent Taylor - Couette flow,\nproduced between two counterrotating coaxial cylinders. In pure Water and in\ncounterrotation, Reynolds numbers up to 1.4 10^5 are reached. We first\ncharacterize the single-phase flow, in terms of threshold for transition to\nturbulence, scaling of the torque and measurements of the mean flow and of the\nReynolds stress by stereoscopic PIV. We then study the increase of the\ndissipation in the two-phase flows and find that the torque per unit mass can\nbe twice the torque for a single-phase flow. Long-time behaviours are also\nreported.", "category": "physics_flu-dyn" }, { "text": "Turbulent superstructures in Rayleigh-B\u00e9nard convection: Turbulent Rayleigh-B\\'enard convection displays a large-scale order in the\nform of rolls and cells on lengths larger than the layer height once the\nfluctuations of temperature and velocity are removed. These turbulent\nsuperstructures are reminiscent of the patterns close to the onset of\nconvection. They are analyzed by numerical simulations of turbulent convection\nin fluids at different Prandtl number ranging from 0.005 to 70 and for Rayleigh\nnumbers up to $10^7$. For each case, we identify characteristic scales and\ntimes that separate the fast, small-scale turbulent fluctuations from the\ngradually changing large-scale superstructures. The characteristic scales of\nthe large-scale patterns, which change with Prandtl and Rayleigh number, are\nalso found to be correlated with the boundary layer dynamics, and in particular\nthe clustering of thermal plumes at the top and bottom plates. Our analysis\nsuggests a scale separation and thus the existence of a simplified description\nof the turbulent superstructures in geo- and astrophysical settings.", "category": "physics_flu-dyn" }, { "text": "Direct solutions for normal depths in curved irrigation canals: The normal depth is an important hydraulic element for canal design,\noperation and management. Curved irrigation canals including parabola, U-shaped\nand catenary canals have excellent hydraulic performance and strong ability of\nanti-frost heave, while the normal depths in the governing equations of the\ncurrent common methods have no explicit analytical solution. They are only\nindirect methods by using trial procedures, numerical methods, and graphical\ntools. This study presents new direct formulas for normal depth in curved\nirrigation canals by applying for Marquardt method. The maximum relative error\nof the proposed formulas is less than 1% within the practice range by\ncomparative analysis, and they are simple and convenient for manual\ncalculations. The results may provide the reliable theoretical basis and useful\nreference for the design and operation management of irrigation canals.", "category": "physics_flu-dyn" }, { "text": "Stochastic dynamics of particles trapped in turbulent flows: The long time dynamics of large particles trapped in two inhomogeneous\nturbulent shear flows is studied experimentally. Both flows present a common\nfeature, a shear region that separates two colliding circulations, but with\ndifferent spatial symmetries and temporal behaviors. Because large particles\nare less and less sensitive to flow fluctuations as their size increases, we\nobserve the emergence of a slow dynamics corresponding to back-and-forth\nmotions between two attractors, and a super-slow regime synchronized with flow\nreversals when they exist. Such dynamics is substantially reproduced by a one\ndimensional stochastic model of an over-damped particle trapped in a two-well\npotential, forced by a colored noise. An extended model is also proposed that\nreproduces observed dynamics and trapping without potential barrier: the key\ningredient is the ratio between the time scales of the noise correlation and\nthe particle dynamics. A total agreement with experiments requires the\nintroduction of spatially inhomogeneous fluctuations and a suited confinement\nstrength.", "category": "physics_flu-dyn" }, { "text": "MHD stability of large scale liquid metal batteries: The aim of this paper is to develop a stability theory and a numerical model\nfor the three density-stratified electrically conductive liquid layers. Using\nregular perturbation methods to reduce the full 3d problem to the shallow layer\nmodel, the coupled wave and electric current equations are derived. The problem\nset-up allows the weakly non-linear velocity field action and an arbitrary\nvertical magnetic field. Further linearisation of the coupled equations is used\nfor the linear stability analysis in the case of uniform vertical magnetic\nfield. New analytical stability criteria accounting for the viscous damping are\nderived for particular cases of practical interest and compared to the\nnumerical solutions for variety of materials used in the batteries. The new\ncriteria are equally applicable to the aluminium electrolysis cell MHD\nstability estimates.", "category": "physics_flu-dyn" }, { "text": "Singular vortex pairs follow magnetic geodesics: We consider pairs of point vortices having circulations $\\Gamma_1$ and\n$\\Gamma_2$ and confined to a two-dimensional surface $S$. In the limit of zero\ninitial separation $\\varepsilon$, we prove that they follow a magnetic geodesic\nin unison, if properly renormalized. Specifically, the ``singular vortex pair\"\nmoves as a single charged particle on the surface with a charge of order\n$1/\\varepsilon^2$ in a magnetic field $B$ which is everywhere normal to the\nsurface and of strength $|B|=\\Gamma_1 +\\Gamma_2$. In the case\n$\\Gamma_1=-\\Gamma_2$, this gives another proof of Kimura's conjecture (Kimura\n1999) that singular dipoles follow geodesics.", "category": "physics_flu-dyn" }, { "text": "Quantum computing of reacting flows via Hamiltonian simulation: We report the quantum computing of reacting flows by simulating the\nHamiltonian dynamics. The scalar transport equation for reacting flows is\ntransformed into a Hamiltonian system, mapping the dissipative and\nnon-Hermitian problem in physical space to a Hermitian one in a\nhigher-dimensional space. Using this approach, we develop the quantum spectral\nand finite difference methods for simulating reacting flows in periodic and\ngeneral conditions, respectively. The present quantum computing algorithms\noffer a ``one-shot'' solution for a given time without temporal discretization,\navoiding iterative quantum state preparation and measurement. We compare\ncomputational complexities of the quantum and classical algorithms. The quantum\nspectral method exhibits exponential acceleration relative to its classical\ncounterpart, and the quantum finite difference method can achieve exponential\nspeedup in high-dimensional problems. The quantum algorithms are validated on\nquantum computing simulators with the Qiskit package. The validation cases\ncover one- and two-dimensional reacting flows with a linear source term and\nperiodic or inlet-outlet boundary conditions. The results obtained from the\nquantum spectral and finite difference methods agree with analytical and\nclassical simulation results. They accurately capture the convection,\ndiffusion, and reaction processes. This demonstrates the potential of quantum\ncomputing as an efficient tool for the simulation of reactive flows in\ncombustion.", "category": "physics_flu-dyn" }, { "text": "Transport of condensing droplets in Taylor-Green vortex flow in the\n presence of thermal noise: We study the role of phase change and thermal noise in particle transport in\nturbulent flows. We employ a toy model to extract the main physics: condensing\ndroplets are modelled as heavy particles which grow in size, the ambient flow\nis modelled as a two-dimensional Taylor-Green (TG) flow consisting of an array\nof vortices delineated by separatrices, and thermal noise are modelled as\nuncorrelated Gaussian white noise. In general, heavy inertial particles are\ncentrifuged out of regions of high vorticity and into regions of high strain.\nIn cellular flows, we find, in agreement with earlier results, that droplets\nwith Stokes numbers smaller than a critical value, $St < St_{\\rm{cr}}$, remain\ntrapped in the vortices in which they are initialised, while larger droplets\nmove ballistically away from their initial positions by crossing separatrices.\nWe independently vary the P\\'eclet number $Pe$ characterising the amplitude of\nthermal noise and the condensation rate $\\Pi$ to study their effects on the\ncritical Stokes number for droplet trapping, as well as on the final states of\nmotion of the droplets. We find that the imposition of thermal noise, or of a\nfinite condensation rate, allows droplets of $St < St_{\\rm{cr}}$ to leave their\ninitial vortices. We find that the effects of thermal noise become negligible\nfor growing droplets, and that growing droplets achieve ballistic motion when\ntheir Stokes numbers become $\\mathcal{O}(1)$. We also find an intermediate\nregime prior to attaining the ballistic state, in which droplets move\ndiffusively away from their initial vortices in the presence of thermal noise.", "category": "physics_flu-dyn" }, { "text": "Helicity in axisymmetric vortex breakdown: Vortex breakdown phenomena in the axial vortices is an important feature\nwhich occurs frequently in geophysical flows (tornadoes and hurricanes) and in\nengineering flows (flow past delta wings, Von-Kerman vortex dynamo). We analyze\nhelicity for axisymmetric vortex breakdown and propose a simplified\nformulation. For such cases, negative helicity is shown to conform to the\nvortex breakdown. A model problem has been analyzed to verify the results. The\ntopology of the vortex breakdown is governed entirely by helicity density in\nthe vertical plane. Our proposed methodology may be regarded as the prototype\nfor identifying and characterize the breakdowns/eye in more complicated\nlarge-scale flows such as tornadoes/hurricanes.", "category": "physics_flu-dyn" }, { "text": "Neural networks in feedback for flow analysis, sensor placement and\n control: This work presents a novel methodology for analysis and control of nonlinear\nfluid systems using neural networks. The approach is demonstrated on four\ndifferent study cases being the Lorenz system, a modified version of the\nKuramoto-Sivashinsky equation, a streamwise-periodic 2D channel flow, and a\nconfined cylinder flow. Neural networks are trained as models to capture the\ncomplex system dynamics and estimate equilibrium points through a Newton\nmethod, enabled by backpropagation. These neural network surrogate models\n(NNSMs) are leveraged to train a second neural network, which is designed to\nact as a stabilizing closed-loop controller. The training process employs a\nrecurrent approach, whereby the NNSM and the neural network controller (NNC)\nare chained in closed loop along a finite time horizon. By cycling through\nphases of combined random open-loop actuation and closed-loop control, an\niterative training process is introduced to overcome the lack of data near\nequilibrium points. This approach improves the accuracy of the models in the\nmost critical region for achieving stabilization. Through the use of L1\nregularization within loss functions, the NNSMs can also guide optimal sensor\nplacement, reducing the number of sensors from an initial candidate set. The\ndatasets produced during the iterative training process are also leveraged for\nconducting a linear stability analysis through a modified dynamic mode\ndecomposition approach. The results demonstrate the effectiveness of\ncomputationally inexpensive neural networks in modeling, controlling, and\nenabling stability analysis of nonlinear systems, providing insights into the\nsystem behaviour and offering potential for stabilization of complex fluid\nsystems.", "category": "physics_flu-dyn" }, { "text": "A new engineering theory describing oblique free surface impact by\n flexible plates: Consideration of slamming loads within the structural design of planning\nhulls is of critical importance in ensuring adequate structural performance in\norder to avoid potential catastrophic consequences. However, because of the\nintricacy in the interplay between complex fluid flows and nonlinear structural\ndeformations that accompany the phenomenology of slamming, a general\nengineering theory in slamming has yet to be uncovered, and so design relies on\nspecialized theories. In this paper, we propose one such theory for a design\ncase that has, until now, eluded a proper description. In pursuit of this\ntheory, we employ a specialized implicit, partitioned fluid-structural\ninteraction (FSI) simulation approach, in order to study the underlying\nphysical mechanisms accompanying the oblique impact of a flexible plate during\nwater entry. In the present work, we first present validation results from\nflexible plate water entry experiments, to confirm the veracity of the\ndeveloped FSI solver. Subsequent to validation, we carry out a series of\nnumerical analyses, in an effort to characterize the regimes in impact force\nand plate out-of-plane deformations, as a function of impact velocities and\nplate flexural rigidity. Finally, we use our FSI solver, as a kind of\n\"microscope\", to study the mechanistic evolution of fluid flows and elastic\nplate deformations that occur during slamming. Based on these observations, we\npropose a novel, but simple engineering theory for flexible plates obliquely\nimpacting the water free surface (e.g. high speed porpoising water craft\nreentry).", "category": "physics_flu-dyn" }, { "text": "Similarity between the primary and secondary air-assisted liquid jet\n breakup mechanism: we report an ultrafast synchrotron x-ray phase contrast imaging study of the\nprimary breakup mechanism of a coaxial air-assisted water jet. We demonstrate\nthat there exist great similarities in the phenomenology of primary breakup\nwith that of the secondary breakup. Especially, a membrane-mediated breakup\nmechanism dominates the breakup process for a wide range of air speeds. This\nfinding reveals the intrinsic connections of these two breakup regimes and has\ndeep implications on the unified theoretical approach in treating the breakup\nmechanism of high speed liquid jet.", "category": "physics_flu-dyn" }, { "text": "Rankine--Hugoniot conditions for fluids whose energy depends on space\n and time derivatives of density: By using the Hamilton principle of stationary action, we derive the governing\nequations and Rankine-Hugoniot conditions for continuous media where the\nspecific energy depends on the space and time density derivatives. The\ngoverning system of equations is a time reversible dispersive system of\nconservation laws for the mass, momentum and energy. We obtain additional\nrelations to the Rankine-Hugoniot conditions coming from the conservation laws\nand discuss the well-founded of shock wave discontinuities for dispersive\nsystems.", "category": "physics_flu-dyn" }, { "text": "Metamorphosis of helical magnetorotational instability in the presence\n axial electric current: This paper presents numerical linear stability analysis of a cylindrical\nTaylor-Couette flow of liquid metal carrying axial electric current in a\ngenerally helical external magnetic field. Axially symmetric disturbances are\nconsidered in the inductionless approximation corresponding to zero magnetic\nPrandtl number. Axial symmetry allows us to reveal an entirely new\nelectromagnetic instability. First, we show that the electric current passing\nthrough the liquid can extend the range of helical magnetorotational\ninstability (HMRI) indefinitely by transforming it into a purely\nelectromagnetic instability. Two different electromagnetic instability\nmechanisms are identified. The first is an internal pinch-type instability,\nwhich is due to the interaction of the electric current with its own magnetic\nfield. Axisymmetric mode of this instability requires a free-space component of\nthe azimuthal magnetic field. When the azimuthal component of the magnetic\nfield is purely rotational and the axial component is nonzero, a new kind of\nelectromagnetic instability emerges. The latter driven by the interaction of\nelectric current with a weak collinear magnetic field in a quiescent fluid\ngives rise to a steady meridional circulation coupled with azimuthal rotation.", "category": "physics_flu-dyn" }, { "text": "A modal analysis of the behavior of inertial particles in turbulence: The clustering of small heavy inertial particles subjected to Stokes drag in\nturbulence is known to be minimal at small and large Stokes number and\nsubstantial at $\\rm St = \\mathcal O(1)$. This non-monotonic trend, which has\nbeen shown computationally and experimentally, is yet to be explained\nanalytically. In this study, we obtain an analytical expression for the\nLyapunov exponents that quantitatively predicts this trend. The sum of the\nexponents, which is the normalized rate of change of the signed-volume of a\nsmall cloud of particles, is correctly predicted to be negative and positive at\nsmall and large Stokes numbers, respectively, asymptoting to $\\tau Q$ as $\\tau\n\\to 0$ and $\\tau^{-1/2} |Q|^{1/4}$ as $\\tau \\to \\infty$, where $\\tau$ is the\nparticle relaxation time and $Q(\\tau)$ is the difference between the norm of\nthe rotation- and strain-rate tensors computed along the particle trajectory.\nAdditionally, the trajectory crossing is predicted only in hyperbolic flows\nwhere $Q<0$ for sufficiently inertial particles with a $\\tau$ that scales with\n$|Q|^{-1/2}$. Following the onset of crossovers, a transition from clustering\nto dispersion is predicted correctly. We show these behaviors are not unique to\nthree-dimensional isotropic turbulence and can be reproduced closely by a\none-dimensional mono-harmonic flow, which appears as a fundamental canonical\nproblem in the study of particle clustering. Analysis of this one-dimensional\ncanonical flow shows that the rate of clustering, quantified as the product of\nthe Lyapunov exponent and particle relaxation time, is bounded by $-1/2$,\nbehaving with extreme nonlinearity in the hyperbolic flows and always remaining\npositive in the elliptic flows. These findings, which are stemmed from our\nanalysis, are corroborated by the direct numerical simulations.", "category": "physics_flu-dyn" }, { "text": "A deformed derivative model for turbulent diffusion of contaminants in\n the atmosphere: In the present work, we propose an advection-diffusion equation with\nHausdorff deformed derivatives to stud the turbulent diffusion of contaminants\nin the atmosphere. We compare the performance of our model to fit experimental\ndata against models with classical and Caputo fractional derivatives. We found\nthat the Hausdorff equation gives better results than the tradition\nadvection-diffusion equation when fitting experimental data. Most importantly,\nwe show that our model and the Caputo fractional derivative model display a\nvery similar performance for all experiments. This last result indicates that\nregardless of the kind of non-classical derivative we use, an\nadvection-diffusion equation with non-classical derivative displaying power-law\nmean square displacement is more adequate to describe the diffusion of\ncontaminants in the atmosphere than a model with classical derivatives.\nFurthermore, since Hausdorff derivatives can be related to several deformed\noperators, and since differential equations with the Hausdorff derivatives are\neasier to solve than equations with Caputo and other non-local fractional\nderivatives, our result highlights the potential of deformed derivative models\nto describe the diffusion of contaminants in the atmosphere.", "category": "physics_flu-dyn" }, { "text": "Non-monotonic surface tension leads to spontaneous symmetry breaking in\n a binary evaporating drop: The evaporation of water/1,2-hexanediol binary drops shows remarkable\nsegregation dynamics, with hexanediol-rich spots forming at the rim, thus\nbreaking axisymmetry. While the segregation of hexanediol near the rim can be\nattributed to the preferential evaporation of water, the symmetry-breaking and\nspot formation could not yet be successfully explained. With three-dimensional\nsimulations and azimuthal stability analysis of a minimal model, we investigate\nthe flow and composition in the drop. We show that a slightly non-monotonic\nsurface tension causes the emergence of a counter-rotating Marangoni vortex in\nthe hexanediol-rich rim region, which subsequently becomes azimuthally unstable\nand forms the observed spots. Accurate measurements reveal that the surface\ntension is indeed non-monotonic (~0.3 mN/m). This work provides valuable\ninsight for applications like inkjet printing or spray cooling.", "category": "physics_flu-dyn" }, { "text": "An Aerodynamic Analysis of a Robustly Redesigned Modern Aero-Engine Fan: This paper documents results from a recent computational study aimed at\nde-sensitizing fan stage aerodynamics---in a modern, high bypass ratio\naero-engine---to the effects of rear-seal leakage flows. These flows are the\nresult of seal erosion between a rotor and stator disk in an engine, and\ndeterioration over the life of an engine. The density-matching technique for\noptimization under uncertainty was applied to this problem. This involved RANS\nand adjoint flow solves of a full fan stage carried out at two different\nleakage conditions. Here a detailed analysis of the fan stage aerodynamics is\ncarried out to determine why exactly the new design is more insensitive to the\neffects of leakage flows. Specifically, it is shown that this insensitivity is\nattributed to three main factors: a slight rearward shift in loading, and thus\na reduction in incidence; a reduction in the cross-passage pressure gradient;\nand a re-acceleration of the flow towards the trailing edge, which prevented\nany corner separation.", "category": "physics_flu-dyn" }, { "text": "Viscous Flow Instability of Inflectional Velocity Profile: Rayleigh showed that inviscid flow is unstable if the velocity profile has an\ninflection point in parallel flows. However, whether viscous flows is unstable\nor not is still not proved so far when there is an inflection point in the\nvelocity profile. Fluid viscosity has showed dual role to the flow instability.\nIn this paper, it is demonstrated for the first time that viscous parallel flow\nwith inflectional velocity profile is sufficient for flow instability.", "category": "physics_flu-dyn" }, { "text": "Patterns in transitional shear turbulence. Part 2: Emergence and optimal\n wavelength: Low Reynolds number turbulence in wall-bounded shear flows \\emph{en route} to\nlaminar flow takes the form of oblique, spatially-intermittent turbulent\nstructures. In plane Couette flow, these emerge from uniform turbulence via a\nspatiotemporal intermittent process in which localised quasi-laminar gaps\nrandomly nucleate and disappear. For slightly lower Reynolds numbers, spatially\nperiodic and approximately stationary turbulent-laminar patterns predominate.\nThe statistics of quasi-laminar regions, including the distributions of space\nand time scales and their Reynolds number dependence, are analysed. A smooth,\nbut marked transition is observed between uniform turbulence and flow with\nintermittent quasi-laminar gaps, whereas the transition from gaps to regular\npatterns is more gradual. Wavelength selection in these patterns is analysed\nvia numerical simulations in oblique domains of various sizes. Via lifetime\nmeasurements in minimal domains, and a wavelet-based analysis of wavelength\npredominance in a large domain, we quantify the existence and non-linear\nstability of a pattern as a function of wavelength and Reynolds number. We\nreport that the preferred wavelength maximises the energy and dissipation of\nthe large-scale flow along laminar-turbulent interfaces. This optimal behaviour\nis primarily due to the advective nature of the large-scale flow, with\nturbulent fluctuations playing only a secondary role.", "category": "physics_flu-dyn" }, { "text": "Nonlinear traveling waves as a framework for understanding turbulent\n drag reduction: Nonlinear traveling waves that are precursors to laminar-turbulent transition\nand capture the main structures of the turbulent buffer layer have recently\nbeen found to exist in all the canonical parallel flow geometries. We study the\neffect of polymer additives on these \"exact coherent states\" (ECS), in the\nplane Poiseuille geometry. Many key aspects of the turbulent drag reduction\nphenomenon are found, including: delay in transition to turbulence; drag\nreduction onset threshold; diameter and concentration effects. Furthermore,\nexamination of the ECS existence region leads to a distinct prediction,\nconsistent with experiments, regarding the nature of the maximum drag reduction\nregime. Specifically, at sufficiently high wall shear rates, viscoelasticity is\nfound to completely suppress the normal (i.e. streamwise-vortex-dominated)\ndynamics of the near wall region, indicating that the maximum drag reduction\nregime is dominated by a distinct class of flow structures.", "category": "physics_flu-dyn" }, { "text": "A nonlinear Schr\u00f6dinger equation for water waves on finite depth with\n constant vorticity: A nonlinear Schr\\\"odinger equation for the envelope of two dimensional\nsurface water waves on finite depth with non zero constant vorticity is\nderived, and the influence of this constant vorticity on the well known\nstability properties of weakly nonlinear wave packets is studied. It is\ndemonstrated that vorticity modifies significantly the modulational instability\nproperties of weakly nonlinear plane waves, namely the growth rate and\nbandwidth. At third order we have shown the importance of the coupling between\nthe mean flow induced by the modulation and the vorticity. Furthermore, it is\nshown that these plane wave solutions may be linearly stable to modulational\ninstability for an opposite shear current independently of the dimensionless\nparameter kh, where k and h are the carrier wavenumber and depth respectively.", "category": "physics_flu-dyn" }, { "text": "A Swimming Rheometer: Self-propulsion of a freely-suspended swimmer\n enabled by viscoelastic normal stresses: Self-propulsion at low Reynolds number is notoriously restricted, a concept\nthat is commonly known as the \"scallop theorem\". Here we present a truly\nself-propelled swimmer (force- and torque- free) that, while unable to swim in\na Newtonian fluid due to the scallop theorem, propels itself in a non-Newtonian\nfluid as a result of fluid elasticity. This propulsion mechanism is\ndemonstrated using a robotic swimmer, comprised of a \"head\" sphere and a \"tail\"\nsphere, whose swimming speed is shown to have reasonable agreement with a\nmicrohydrodynamic asymptotic theory and numerical simulations. Schlieren\nimaging demonstrates that propulsion of the swimmer is driven by a strong\nviscoelastic jet at the tail, which develops due to the fore-aft asymmetry of\nthe swimmer. Optimized cylindrical and conic tail geometries are shown to\ndouble the propulsive signal, relative to the optimal spherical tail. Finally,\nwe show that we can use observations of this robot to infer rheological\nproperties of the surrounding fluid. We measure the primary normal stress\ncoefficient at shear rates less than 1 Hz, and show reasonable agreement with\nextrapolated benchtop measurements (between 0.8 to 1.2 Pa sec2 difference). We\nalso discuss how our swimmer can be used to measure the second normal stress\ncoefficient and other rheological properties. The study experimentally\ndemonstrates the exciting potential for a \"swimming rheometer\", bringing\npassive physics-driven fluid sensing to numerous applications in chemical and\nbioengineering.", "category": "physics_flu-dyn" }, { "text": "Numerical Analysis of an Imploding Shock Wave in Solid: Spherical or cylindrical convergent shock waves in imploding materials are\none of the most effective ways to produce extremely high pressures, densities\nand temperatures, hardly attainable in plane shock waves generated by chemical\nhigh explosives or by the impact of high velocities objects. Pressure of the\norder of tens of megabars, densities many times greater than the normal density\nof solids and temperatures of hundreds of thousands degree can be easily\nproduced in the convergence central region of the imploding shock wave.\n In this work we perform a hydrodynamic analysis of a spherical mass of lead\nimploded by an external constant pressure of 1 Mbar acting on its surface. The\naim is to monitor the rise of pressure, density, velocity and temperature with\nthe convergence and reflection of the shock wave at the centre. The analysis\nwas carried out by a hydrodynamic code and by making use of the three-term\nequation of state for solids. This equation of state takes into account the\nelastic pressure, the thermal pressure of atoms and the thermal pressure of\nelectrons in solids submitted to strong shock compressions.", "category": "physics_flu-dyn" }, { "text": "Guidelines for the formulation of Lagrangian stochastic models for\n particle simulations of single-phase and dispersed two-phase turbulent flows: In this paper, we establish a set of criteria which are applied to discuss\nvarious formulations under which Lagrangian stochastic models can be found.\nThese models are used for the simulation of fluid particles in single-phase\nturbulence as well as for the fluid seen by discrete particles in dispersed\nturbulent two-phase flows. A central issue is to put forward reliable\nrequirements which must be met by Lagrangian stochastic models and a new\nelement brought by the present analysis is to address the single- and two-phase\nflow situations from a unified point of view.This analysis does not address the\nquestion of the relative predictive capacities of different models but\nconcentrates on their formulation since advantages and disadvantages of\ndifferent formulations are not always clear. Indeed, hidden in the changes from\none structure to another are some possible pitfalls which can lead to flaws in\nthe construction of practical models and to physically-unsound numerical\ncalculations. A first interest of the present approach is illustrated by\nconsidering some models proposed in the literature and by showing that these\ncriteria help to assess whether these Lagrangian stochastic models can be\nregarded as acceptable descriptions. A second interest is to indicate how\nfuture developments can be safely built, which is also relevant for stochastic\nsubgrid models for particle-laden flows in the context of Large Eddy\nSimulations.", "category": "physics_flu-dyn" }, { "text": "Viscoplastic Lines: Printing a Single Filament of Yield Stress Material\n on a Surface: This study presents the spreading of a single filament of a yield stress\n(viscoplastic) fluid extruded onto a pre-wetted solid surface. The filaments\nspread laterally under surface tension forces until they reach a final\nequilibrium shape when the yield stress dominates. We use a simple experimental\nsetup to print the filaments on a moving surface and measure their final width\nusing optical coherence tomography. Additionally, we present a scaling law for\nthe final width and determine the corresponding pre-factor using asymptotic\nanalysis. We then analyse the level of agreement between the theory and\nexperiments and discuss the possible origins of discrepancies. The process\nstudied here has applications in extrusion-based thermoplastic and bio-3D\nprinting.", "category": "physics_flu-dyn" }, { "text": "Predicting waves in fluids with deep neural network: In this paper, we present a deep learning technique for data-driven\npredictions of wave propagation in a fluid medium. The technique relies on an\nattention-based convolutional recurrent autoencoder network (AB-CRAN). To\nconstruct a low-dimensional representation of wave propagation data, we employ\na denoising-based convolutional autoencoder. The AB-CRAN architecture with\nattention-based long short-term memory cells forms our deep neural network\nmodel for the time marching of the low-dimensional features. We assess the\nproposed AB-CRAN framework against the standard recurrent neural network for\nthe low-dimensional learning of wave propagation. To demonstrate the\neffectiveness of the AB-CRAN model, we consider three benchmark problems,\nnamely, one-dimensional linear convection, the nonlinear viscous Burgers\nequation, and the two-dimensional Saint-Venant shallow water system. Using the\nspatial-temporal datasets from the benchmark problems, our novel AB-CRAN\narchitecture accurately captures the wave amplitude and preserves the wave\ncharacteristics of the solution for long time horizons. The attention-based\nsequence-to-sequence network increases the time-horizon of prediction compared\nto the standard recurrent neural network with long short-term memory cells. The\ndenoising autoencoder further reduces the mean squared error of prediction and\nimproves the generalization capability in the parameter space.", "category": "physics_flu-dyn" }, { "text": "Dependence of the separative power of an optimised Iguassu gas\n centrifuge on the velocity of rotor: Purpose - The purpose of this work is to determine dependence of the\nseparative power of the Iguassu gas centrifuge (GC) on the velocity of the\nrotor.\n Design/methodology/approach - The dependence is determined by means of\ncomputer simulation of the gas flow in the GC and numerical solution of the\ndiffusion equation for the light component of the binary mixture of uranium\nisotopes. 2D axisymmetric model with the sources/sinks of the mass, angular\nmomentum and energy reproducing the scoops was explored for the computer\nsimulation. Parameters of the model correspond to the parameters of the\nso-called Iguassu centrifuge. The separative power has been optimised in\nrelation to the pressure of the gas, temperature of the gas, the temperature\ndrop along the rotor, power of the source of angularmomentum and energy, feed\nflow and geometry of the lower baffle.\n Findings - In the result, the optimised separative power depends only on the\nvelocity, length and diameter of the rotor. The dependence on the velocity is\ndescribed by the power law function with the power law index 2.6 which\ndemonstrate stronger dependence on the velocity than it follows from\nexperimental data. However, the separative power obtained with limitation on\nthe pressure growth with the velocity depends on the velocity on the power\nsimilar to 2 which well agree with the experiments.\n Originality/value - For the first time, the optimised separative power of the\nGCs have been calculated via numerical simulation of the gas flow and diffusion\nof the binary mixture of isotope.", "category": "physics_flu-dyn" }, { "text": "A viscoelastic phase-field model for iceberg calving: Iceberg calving accounts for around half of the ice lost annually from\nAntarctica, but realistic representation of fracture and calving in large-scale\nice sheet models remains a major unsolved problem in glaciology. We present a\nnew phase-field viscoelastic model for fracture that simulates the slow\ndeformation of ice and the distribution and evolution of cracks. Cracks\nnucleate and propagate in response to the evolving stress field, and are\ninfluenced by water pressure below sea level. The model incorporates\nnonlinear-viscous rheology, linear-elastic rheology, and a phase-field\nvariational formulation, which allows simulation of complex fracture phenomena.\nWe show that this approach is capable of simulating the physical process of\ncalving. Numerical experiments supported by a simplified model suggest that\ncalving rate will scale with the fourth power of ice thickness for a floating\nice front that has no variation across flow. The equations make no assumptions\nabout the style of calving, so they would also simulate numerous more realistic\nsettings in Antarctica for which material parameters and three-dimensional\neffects can be expected to influence the calving rate.", "category": "physics_flu-dyn" }, { "text": "An alternative view on dissipation in turbulent flows: An original experimental setup has been elaborated in order to get a better\nview of turbulent flows in a von Karman geometry. The availability of a very\nfast camera allowed to follow in time the evolution of the flows. A surprising\nfinding is that the development of smaller whorls ceases earlier than expected\nand the aspect of the flows remains the same above Reynolds number of a few\nthousand. This fact provides an explanation of the constancy of the reduced\ndissipation in the same range without the need of singularity. Its cause could\nbe in relation with the same type of behavior observed in a rotating frame.", "category": "physics_flu-dyn" }, { "text": "On the fundamentals of Richtmyer-Meshkov dynamics with variable\n acceleration: Richtmyer-Meshkov instability (RMI) plays important role in nature and\ntechnology, from supernovae and fusion to scramjets and nano-fabrication.\nCanonical Richtmyer-Meshkov instability is induced by a steady shock and\nimpulsive acceleration, whereas in realistic environments the acceleration is\nusually variable. This work focuses on RMI induced by acceleration with a\npower-law time-dependence, and applies group theory to solve the classical\nproblem. For early-time dynamics, we find the dependence of RMI growth-rate on\nthe initial conditions and show it is free from the acceleration parameters.\nFor late time dynamics, we find a continuous family of regular asymptotic\nsolutions, including their curvature, velocity, Fourier amplitudes, and\ninterfacial shear, and we study the solutions stability. For each of the\nsolutions, the interface dynamics is directly linked to the interfacial shear,\nand the non-equilibrium velocity field has intense fluid motion near the\ninterface and effectively no motion in the bulk. The quasi-invariance of the\nfastest stable solution suggests that nonlinear coherent dynamics in RMI is\ncharacterized by two macroscopic length-scales - the wavelength and the\namplitude, in excellent agreement with observations. We elaborate new theory\nbenchmarks for experiments and simulations, and put forward a hypothesis on the\nrole of viscous effects in interfacial nonlinear RMI.", "category": "physics_flu-dyn" }, { "text": "Fluid motion induced by surface waves at low Reynolds number: We discuss the scaling laws for the flow generated in a viscous fluid by a\nwave propagating along a solid boundary. This has applications to the\ndisplacement of tiny objects on solids, under the effect of progressive surface\nwaves and for the swimming of microanimals by undulation of ciliae along their\nbody surface.", "category": "physics_flu-dyn" }, { "text": "Efficient survival strategy for zooplankton in turbulence: Zooplankton in a quiescent environment can detect predators by hydrodynamic\nsensing, triggering powerful escape responses. Since turbulent strain tends to\nmask the hydrodynamic signal, the organisms should avoid such regions, but it\nis not known how they accomplish this. We found a simple, robust, and highly\nefficient strategy, that relies on measuring strain gradients. Plankton\nfollowing this strategy show very strong spatial clustering, and align against\nthe local flow velocity, facilitating mate finding and feeding. The strategy\nhas the potential to reconcile competing fitness pressures.", "category": "physics_flu-dyn" }, { "text": "Schlieren study of a sonic jet injected into a supersonic cross flow\n using high-current pulsed LEDs: Benefiting from the development of increasingly advanced high speed cameras,\nflow visualization and analysis nowadays yield detailed data of the flow field\nin many applications. Notwithstanding this progress, for high speed and\nsupersonic flows it is still not trivial to capture high quality images. In\nthis study we present a Schlieren setup that uses pulsed LEDs with high\ncurrents (up to 18 Ampere) to increase the optical output to sufficient levels.\nThe bright and short pulses, down to 130 nanoseconds, allow detailed and sharp\nimaging of the flow with a high spatial resolution adequate for supersonic\nflow. The pulse circuit and pulse width determination are explained in detail.\nAs a test case we studied the near field of a 2 mm diameter sonic jet injected\ntransversely into a supersonic cross flow. This is a model flow for fuel\ninjection in scramjet engines, which is not yet fully understood. Owing to the\nhigh resolution and accuracy of the images produced by the newly developed\nsystem we prove the existence of a large (density) gradient wave traveling in\nthe windward subsonic region between the Mach barrel and the bowshock, which\nhitherto was observed only in some numerical studies but not yet shown in\nexperiments. Furthermore, we demonstrate with this Schlieren setup that\ntime-correlated images can be obtained, with an interframe time of 2\nmicroseconds, so that also flow unsteadiness can be studied such as the\nmovement of shock waves and trajectories of vortices. The obtained results of\nthe jet penetration height are presented as a power law correlation. The\nresults of this study show that the designed setup using a low-cost LED and\nlow-cost control system is a high potential option for application in\nvisualization studies of high speed flows.", "category": "physics_flu-dyn" }, { "text": "Large eddy simulation of a supersonic lifted hydrogen flame with\n perfectly stirred reactor model: Large Eddy Simulation with a Perfectly Stirred Reactor model (LES-PSR) is\ndeveloped to simulate supersonic combustion with high-enthalpy flow conditions.\nThe PSR model considers the viscous heating and compressibility effects on the\nthermo-chemical state, through correcting the chemical source term for progress\nvariable and incorporating absolute enthalpy as the control variable for the\nlook-up table. It is firstly validated by using a priori analysis of the\nviscous heating and compressibility effects. Then an auto-igniting hydrogen\nflame stabilized in supersonic vitiated co-flowing jet is simulated with\nLES-PSR method. The results show that the shock wave structure, overall flame\ncharacteristics, flame-shock interaction and lift-off height are accurately\ncaptured. Good agreements of the velocity and mixture fraction statistics with\nthe experimental data are observed. The results also show that the LES-PSR\nmodel can predict the mean temperature and mole fractions of major species\nquite well in both flame induction and stabilization zones. However, there are\nsome under-predictions of temperature RMS by about 100-150 K, which may be due\nto the chemical non-equilibrium in the H2/O2-enriched combustion product of the\nco-flowing jet. The scatter plots of two probe locations respectively from\ninduction and flame zones show that the respective flame structures in mixture\nfraction space are captured well. However, the flucturations of the temperature\nand species mole fractions are under-predicted in the flame zone. The\nshock-induced auto-igniting spots are captured by the PSR model. These spots\nare highly unsteady and play an important role in flame stabilization. It is\nalso shown that the intense reactions are initiated at mixture fractions around\nthe stoichiometry or fuel-lean values, corresponding to local elevated pressure\n(1.5-2.0 atm) due to shock compression.", "category": "physics_flu-dyn" }, { "text": "Enhancing active wave absorption in RANS models: In this work we review the most common methods for absorbing waves in\nReynolds-Averaged Navier-Stokes (RANS) models. The limitations of active wave\nabsorption, originating from its initial assumption of linear wave theory in\nshallow waters are overcome and the range of applicability is extended to any\nrelative water depth conditions by re-deriving the formulation. The new\nExtended Range Active Wave Absorption (ER-AWA) overperforms the traditional\nimplementation in all the tests performed, which comprise solitary waves and\nregular waves from shallow to deep waters. Moreover, the combined use of a\nrelaxation zone and ER-AWA is tested to further reduce wave reflections. This\nis most often achieved for a given set of parameters, although some case by\ncase tuning of the relaxation zone parameters would be needed to obtain the\nbest overall performance.", "category": "physics_flu-dyn" }, { "text": "Formation of surface nanodroplets under controlled flow conditions: Nanodroplets on a solid surface (i.e. surface nanodroplets) have practical\nimplications for high-throughput chemical and biological analysis,\nlubrications, lab-on-chip devices, and near-field imaging techniques. Oil\nnanodroplets can be produced on a solid-liquid interface in a simple step of\nsolvent exchange in which a good solvent of oil is displaced by a poor solvent.\nIn this work, we experimentally and theoretically investigate the formation of\nnanodroplets by the solvent exchange process under well-controlled flow\nconditions. We find that the contact angle of the nanodroplets is independent\nof the flow condition. However, there are significant effects from the flow\nrate and the flow geometry on the droplet size. We develop a theoretical\nframework to account for these effects. The main idea is that the droplet\nnuclei are exposed to an oil oversaturation pulse during the exchange process.\nThe analysis gives that the volume of the nanodroplets increases with the\nPeclet number $Pe$ of the flow as $\\propto Pe^{3/4}$, which is in good\nagreement with our experimental results. In addition, at fixed flow rate and\nthus fixed Peclet number, larger and less homogeneously distributed droplets\nformed at less narrow channels, due to convection effects originating from the\ndensity difference between the two solutions of the solvent exchange. The\nunderstanding from this work provides valuable guidelines for producing surface\nnanodroplets with desired sizes by controlling the flow conditions.", "category": "physics_flu-dyn" }, { "text": "Asymmetric Vortex Sheet: We present a steady analytical solution of the incompressible Navier-Stokes\nequation for arbitrary viscosity in an arbitrary dimension $d$ of space. It\nrepresents a $d-1$ dimensional vortex \"sheet\" with an asymmetric profile of\nvorticity as a function of the normal coordinate $z$. This profile is related\nto the Hermite polynomials $H_\\mu(z)$ which are analytically continued to the\nnegative fractional index $\\mu = -\\frac{d}{d-1}$. In $d=2$ dimensions, the\nsolution degenerates to a constant vorticity flow. In $ d \\ge 3$ dimensions,\nthe vorticity is confined to the thin layer around the hyperplane with Gaussian\ndecay on one side of the hyperplane and the power decay on another side. One\ncan adjust the common scale of velocity so that the dissipation will stay\nfinite at vanishing viscosity. In this limit, the width $w$ of the viscous\nlawyer will shrink to zero as $\\nu^{\\frac{3}{5}}$ for arbitrary dimension\n$d>3$. In $d=3$ dimensions, this power law is also accompanied by powers of the\nlogarithm.", "category": "physics_flu-dyn" }, { "text": "Artificial Neural Networks Modelling of Wall Pressure Spectra Beneath\n Turbulent Boundary Layers: We analyse and compare various empirical models of wall pressure spectra\nbeneath turbulent boundary layers and propose an alternative machine learning\napproach using Artificial Neural Networks (ANN). The analysis and the training\nof the ANN are performed on data from experiments and high-fidelity simulations\nby various authors, covering a wide range of flow conditions. We present a\nmethodology to extract all the turbulent boundary layer parameters required by\nthese models, also considering flows experiencing strong adverse pressure\ngradients. Moreover, the database is explored to unveil important dependencies\nwithin the boundary layer parameters and to propose a possible set of features\nfrom which the ANN should predict the wall pressure spectra. The results show\nthat the ANN outperforms traditional models in adverse pressure gradients, and\nits predictive capabilities generalise better over the range of investigated\nconditions. The analysis is completed with a deep ensemble approach for\nquantifying the uncertainties in the model prediction and integrated gradient\nanalysis of the model sensitivity to its inputs. Uncertainties and\nsensitivities allow for identifying the regions where new training data would\nbe most beneficial to the model's accuracy, thus opening the path towards a\nself-calibrating modelling approach.", "category": "physics_flu-dyn" }, { "text": "Polymer stress growth in viscoelastic fluids in oscillating extensional\n flows with applications to micro-organism locomotion: Simulations of undulatory swimming in viscoelastic fluids with large\namplitude gaits show concentration of polymer elastic stress at the tips of the\nswimmers.We use a series of related theoretical investigations to probe the\norigin of these concentrated stresses. First the polymer stress is computed\nanalytically at a given oscillating extensional stagnation point in a\nviscoelastic fluid. The analysis identifies a Deborah number (De) dependent\nWeissenberg number (Wi) transition below which the stress is linear in Wi, and\nabove which the stress grows exponentially in Wi. Next, stress and velocity are\nfound from numerical simulations in an oscillating 4-roll mill geometry. The\nstress from these simulations is compared with the theoretical calculation of\nstress in the decoupled (given flow) case, and similar stress behavior is\nobserved. The flow around tips of a time-reversible flexing filament in a\nviscoelastic fluid is shown to exhibit an oscillating extension along particle\ntrajectories, and the stress response exhibits similar transitions. However in\nthe high amplitude, high De regime the stress feedback on the flow leads to non\ntime-reversible particle trajectories that experience asymmetric stretching and\ncompression, and the stress grows more significantly in this regime. These\nresults help explain past observations of large stress concentration for large\namplitude swimmers and non-monotonic dependence on De of swimming speeds.", "category": "physics_flu-dyn" }, { "text": "Equilibrium and non-equilibrium time-reversible dynamical ensembles\n relevant to chiral turbulence: Ideas and theories of turbulence based on modifying the Navier-Stokes\nequation, to obtain equilibrium and non-equilibrium time-reversible dynamical\nensembles relevant to helical turbulence, are presented. Discussions of\ncontrolling helicity to control the aerodynamic force, heat and noise are\npresented, together with the compressible turbulence relevant statistical\nmechanics analysis. A helical time-reversible system for nonequilibrium\ndynamical ensemble is constructed. Applications are also remarked.", "category": "physics_flu-dyn" }, { "text": "A Weak Formulation of Water Waves in Surface Variables: In this short note, we derive a system of two nonlocal equations for the\nwater-wave problem following the work of [AFM06]. Specifically, we consider a\nfluid with a one-dimensional free surface for an irrotational fluid both with,\nand without, surface tension. We show that these equations can be useful for\nderiving direct maps between boundary data at the various interfaces and\nconsider various asymptotic regimes. Finally, for periodic traveling wave\nsolutions, we derive a new single-equation that is derivative free and has the\npotential to simplify the process of finding asymyptotic expansions of the\nsolutions as well as reduce numerical errors when solving computationally.", "category": "physics_flu-dyn" }, { "text": "Imprints of log-periodicity in thermoacoustic systems close to lean\n blowout: In the context of statistical physics, critical phenomena are accompanied by\npower laws having a singularity at the critical point where a sudden change in\nthe state of the system occurs. In this work, we show that lean blowout (LBO)\nin a turbulent thermoacoustic system can be viewed as a critical phenomenon. As\na crucial discovery of the system dynamics approaching LBO, we unravel the\nexistence of the discrete scale invariance (DSI). In this context, we identify\nthe presence of log-periodic oscillations in the temporal evolution of the\namplitude of dominant mode of low-frequency oscillations $(A_f)$ exist in\npressure fluctuations preceding LBO. The presence of DSI indicates the\nrecursive development of blowout. Additionally, we find that $A_f$ shows a\nfaster than exponential growth and becomes singular when blowout occurs. We\nthen present a model that depicts the evolution of $A_f$ based on log-periodic\ncorrections to the power law associated with its growth. Using the model, we\nfind that blowout can be predicted even several seconds earlier. The predicted\ntime of LBO in good agreement with the actual time of occurrence of LBO\nobtained from the experiment.", "category": "physics_flu-dyn" }, { "text": "Effects of polymer additives in the bulk of turbulent thermal convection: We present experimental evidence that a minute amount of polymer additives\ncan significantly enhance heat transport in the bulk region of turbulent\nthermal convection. The effects of polymer additives are found to be the\n\\textit{suppression} of turbulent background fluctuations that give rise to\nincoherent heat fluxes that make no net contribution to heat transport, and at\nthe same time to \\textit{increase} the coherency of temperature and velocity\nfields. The suppression of small-scale turbulent fluctuations leads to more\ncoherent thermal plumes that result in the heat transport enhancement. The fact\nthat polymer additives can increase the coherency of thermal plumes is\nsupported by the measurements of a number of local quantities, such as the\nextracted plume amplitude and width, the velocity autocorrelation functions and\nthe velocity-temperature cross-correlation coefficient. The results from local\nmeasurements also suggest the existence of a threshold value for the polymer\nconcentration, only above which can significant modification of the plume\ncoherent properties and enhancement of the local heat flux be observed.\nEstimation of the plume emission rate suggests that the second effect of\npolymer additives is to stabilize the thermal boundary layers.", "category": "physics_flu-dyn" }, { "text": "Analysis of Boundary Slip in a Flow with an Oscillating Wall: Molecular dynamic (MD) simulation is used to study slip at the fluid-solid\nboundary in an unsteady flow based on the Stokes second problem. An increase in\nslip is observed in comparison to the steady flow for shear rates below the\ncritical shear rate of the corresponding steady flow. This increased slip is\nattributed to fluid inertial forces not represented in a steady flow. An\nunsteady mathematical model for slip is established, which estimates the\nincrement in slip at the boundary. The model shows that slip is also dependent\non acceleration in addition to the shear rate of fluid at the wall. By writing\nacceleration in terms of shear rate, it is shown that slip at the wall in\nunsteady flows is governed by the gradient of shear rate and shear rate of the\nfluid. Non-dimensionalizing the model gives a universal curve which can be used\nto find the slip boundary condition at the fluid-solid interface based on the\ninformation of shear rate and gradient of shear rate of the fluid. A governing\nnon-dimensional number, defined as the ratio of phase speed to speed of sound,\nis identified to help in explaining the mechanism responsible for the\ntransition of slip boundary condition from finite to a perfect slip and\ndetermining when this would occur. Phase lag in fluid velocity relative to wall\nis observed. The lag increases with decreasing time period of wall oscillation\nand increasing wall hydrophobicity. The phenomenon of hysteresis is seen when\nlooking into the variation of slip velocity as a function of wall velocity and\nslip velocity as a function of fluid shear rate. The cause for hysteresis is\nattributed to the unsteady inertial forces of the fluid.", "category": "physics_flu-dyn" }, { "text": "High Weissenberg number simulations with incompressible Smoothed\n Particle Hydrodynamics and the log-conformation formulation: Viscoelastic flows occur widely, and numerical simulations of them are\nimportant for a range of industrial applications. Simulations of viscoelastic\nflows are more challenging than their Newtonian counterparts due to the\npresence of exponential gradients in polymeric stress fields, which can lead to\ncatastrophic instabilities if not carefully handled. A key development to\novercome this issue is the log-conformation formulation, which has been applied\nto a range of numerical methods, but not previously applied to Smoothed\nParticle Hydrodynamics (SPH). Here we present a 2D incompressible SPH algorithm\nfor viscoelastic flows which, for the first time, incorporates a\nlog-conformation formulation with an elasto-viscous stress splitting (EVSS)\ntechnique. The resulting scheme enables simulations of flows at high\nWeissenberg numbers (accurate up to Wi=85 for Poiseuille flow). The method is\nrobust, and able to handle both internal and free-surface flows, and a range of\nlinear and non-linear constitutive models. Several test cases are considerd\nincluded flow past a periodic array of cylinders and jet buckling. This\npresents a significant step change in capabilties compared to previous SPH\nalgorithms for viscoelastic flows, and has the potential to simulate a wide\nrange of new and challenging applications.", "category": "physics_flu-dyn" }, { "text": "Free surface flow due to a submerged source: In this thesis we consider the free surface flow due to a submerged source in\na channel of finite depth. This problem has been considered previously in the\nliterature, with some disagreement about whether or not a train of waves exist\non the free surface for Froude numbers less than unity. The physical\nassumptions behind the accepted model are clearly stated and governing\nequations and boundary conditions derived. Complex variable theory is then\nemployed to obtain a singular nonlinear integral equation, which describes the\nflow field completely.\n The integral equation is evaluated numerically using a collocation method,\nand implemented with mathematical software {\\em Matlab}. The numerical\nsolutions suggest that indeed waves do exist on the free surface, but are\nexponentially small in the limit that the Froude number approaches zero. An\nasymptotic expression is also derived to solve the integral equation, valid for\nsmall Froude numbers. This expansion is calculated to second order, which\nimproves on the leading order solution given previously in the literature. Such\na scheme is unable to predict waves on the free surface, since they are\nexponentially small. It is discussed how exponential asymptotics could be used\nto derive a more accurate analytic solution that describes waves on the free\nsurface.", "category": "physics_flu-dyn" }, { "text": "Vortex-dynamical Interpretation of Anti-phase and In-phase Flickering of\n Dual Buoyant Diffusion Flames: Anti-phase and in-phase flickering modes of dual buoyant diffusion flames\nwere numerically investigated and theoretically analyzed in this study.\nInspired by the flickering mechanism of a single buoyant diffusion flame, for\nwhich the deformation, stretching, or even pinch-off of the flame surface\nresult from the formation and evolution of the toroidal vortices, we attempted\nto understand the anti-phase and in-phase flickering of dual buoyant diffusion\nflames from the perspective of vortex dynamics. The interaction between the\ninner-side shear layers of the two flames was identified to be responsible for\nthe different flickering modes. Specifically, the transition between anti-phase\nand in-phase flickering modes can be predicted by a unified regime nomogram of\nthe normalized flickering frequency versus a characteristic Reynolds number,\nwhich accounts for the viscous effect on vorticity diffusion between the two\ninner-side shear layers. Physically, the transition of the vortical structures\nfrom symmetric (in-phase) to staggered (anti-phase) in a dual-flame system can\nbe interpreted as being similar to the mechanism causing flow transition in the\nwake of a bluff body and forming the Karman vortex street.", "category": "physics_flu-dyn" }, { "text": "Development of Maximum Conical Shock Angle Limit for Osculating Cone\n Waveriders: Hypersonic waveriders are special shapes with leading edges coincident with\nthe body's shock wave, yielding high lift-to-drag ratios. The waverider\ngeometry results from streamline tracing using the solutions of a basic flow\nfield such as the wedge or the cone for specified shock and base curves. The\nbase and shock curves can be independently prescribed in the osculating cone\nmethod enabling a larger design space. Generally, low values of the conical\nshock angle (9-15 degrees) are used. The lack of any method to limit the\nmaximum cone angle for osculating cone waverider motivates this study.\nMathematical expressions are derived for geometrical conditions that result in\nsuccessful osculating cone waverider generation. A power law curve and a Bezier\ncurve are analyzed. Closed-form expressions for the maximum cone shock angle\nare obtained for the power law curve. A numerical procedure to solve the same\nfor the Bezier curve is developed. The results, for a typical Mach number of\n6.0, evidently show that the maximum cone shock angle for successful waverider\ngeneration is significantly lower than the maximum angle for attached shock\nsolutions. The limiting conditions developed will be essential in constraining\nthe waverider sample space for automated multiobjective optimization routines.", "category": "physics_flu-dyn" }, { "text": "Oscillatory instability of fully 3D flow in a cubic diagonally\n lid-driven cavity: A transition to unsteadiness of a flow inside a cubic diagonally lid-driven\ncavity with no-slip boundaries is numerically investigated by a series of\ndirect numerical simulations (DNS) performed on 100^3 and 200^3 stretched\ngrids. It is found that the observed oscillatory instability is setting in via\na subcritical symmetry-breaking Hopf bifurcation. The instability evolves on\ntwo vortices in a coupled manner. Critical values of Reynolds number Recr=2320\nand non-dimensional angular oscillating frequency omegacr=0.249 for transition\nfrom steady to oscillatory flow are accurately estimated. Characteristic\npatterns of the 3D oscillatory flow are presented.", "category": "physics_flu-dyn" }, { "text": "Spontaneous locomotion of a symmetric squirmer: The squirmer is a popular model to analyse the fluid mechanics of a\nself-propelled object, such as a micro-organism. We demonstrate that some\nfore-aft symmetric squirmers can spontaneously self-propel above a critical\nReynolds number. Specifically, we numerically study the effects of inertia on\nspherical squirmers characterised by an axially and fore-aft symmetric\n`quadrupolar' distribution of surface-slip velocity; under creeping-flow\nconditions, such squirmers generate a pure stresslet flow, the stresslet sign\nclassifying the squirmer as either a `pusher' or `puller.' Assuming axial\nsymmetry, and over the examined range of the Reynolds number $Re$ (defined\nbased upon the magnitude of the quadrupolar squirming), we find that\nspontaneous symmetry breaking occurs in the puller case above $Re \\approx\n14.3$, with steady swimming emerging from that threshold consistently with a\nsupercritical pitchfork bifurcation and with the swimming speed growing\nmonotonically with $Re$.", "category": "physics_flu-dyn" }, { "text": "Structure of iso-density sets in supersonic isothermal turbulence: Context: The gas density structure of the cold molecular phase of the\ninterstellar medium is the main controller of star formation. Aims: A\ntheoretical framework is proposed to describe the structural content of the\ndensity field in isothermal supersonic turbulence. Methods: It makes use of\ncorrelation and structure functions of the phase indicator field defined for\ndifferent iso-density values. The relations between these two-point statistics\nand the geometrical features of iso-density sets such as the volume fraction,\nthe surface density, the curvature, and fractal characteristics are provided.\nAn exact scale-by-scale budget equation is further derived revealing the role\nof the turbulent cascade and dilation on the structural evolution of the\ndensity field. Although applicable to many flow situations, this tool is here\nfirst invoked for characterising supersonic isothermal turbulence, using data\nfrom the currently best-resolved numerical simulation. Results: We show that\niso-density sets are surface fractals rather than mass fractals, with\ndimensions that markedly differ between dilute, neutral, and dense regions. The\nsurface-size relation is established for different iso-density values. We\nfurther find that the turbulent cascade of iso-density sets is directed from\nlarge towards smaller scales, in agreement with the classical picture that\nturbulence acts to concentrate more surface into smaller volumes. Intriguingly,\nthere is no range of scales that complies with a constant transfer rate in the\ncascade, challenging our fundamental understanding of interstellar turbulence.\nFinally, we recast the virial theorem in a new formulation drawing an explicit\nrelation between the aforementioned geometrical measures and the dynamics of\niso-density sets.", "category": "physics_flu-dyn" }, { "text": "Development and validation of a transition model based on a mechanical\n approximation: A new 3D transition turbulence model, more accurate and faster than an\nempirical transition model, is proposed. The model is based on the calculation\nof the pre-transitional u'v' due to mean flow shear. The present transition\nmodel is fully described and verified against eight benchmark test cases.\nComputations are performed for the ERCOFTAC flat-plate T3A, T3C and T3L test\ncases. Further, the model is validated for bypass, cross-flow and separation\ninduced transition and compared with empirical transition models. The model\npresents very good results for bypass transition under zero-pressure gradient\nand with pressure gradient flow conditions. Also the model is able to correctly\npredict separation induced transition. However, for very low speed and low\nfree-stream turbulence intensity the model delays separation induced transition\nonset. The model also shows very good results for transition under complex\ncross-flow conditions in three-dimensional geometries. The 3D tested case was\nthe 6:1 prolate-spheroid under three flow conditions.", "category": "physics_flu-dyn" }, { "text": "Analysis of micro-fluidic tweezers in the Stokes regime: Nanowire fluidic tweezers have been developed to gently and accurately\ncapture, manipulate and deliver micro objects. The mechanism behind the capture\nand release process has not yet been well explained. Utilizing the method of\nregularized Stokeslet, we study a cylindrical nanowire tumbling and interacting\nwith spherical particles in the Stokes regime. The capture phenomenon observed\nin experiments is reproduced and illustrated with the trajectories of\nmicro-spheres and fluid tracers. The flow structure and the region of capture\nare quantitatively examined and compared for different sizes of particles,\nvarious tumbling rates, and dimensions of the tweezers. We find that pure\nkinematic effects can explain the mechanism of capture and transport of\nparticles. We further reveal the relation between the capture region and\nstagnation points in the displacement field , i.e., the displacement for tracer\nparticles in the moving frame within one rotation of the wire.", "category": "physics_flu-dyn" }, { "text": "Internal gravity waves near to the sources of disturbances at the\n critical modes of generation: The paper presents the description of the structure of the nearest field of\nthe internal gravity waves at the critical modes of their excitation. Studied\nare the exact solutions both for the elevation component and the vertical\ncomponent of the speed describing the structure of the wave field in the direct\nvicinity of the source. At that the single mode of the elevation is expressed\nthrough the full elliptic integral of the first order, and the single mode of\nthe vertical speed - through McDonald function and the logarithmic functions.\nAs the result of the study it was possible to obtain expressions for the full\nfield representing the sum of the wave modes and expressed through the\nderivatives of the gamma function. The obtained asymptotic and exact\nrepresentations of the solution allow to describe the critical modes of\ngeneration of the internal gravity waves near to the sources of excitations -\nfor the wide ranges of the sources movement velocity .", "category": "physics_flu-dyn" }, { "text": "Unstable Periodically Forced Navier-Stokes Solutions---Towards Nonlinear\n First-Principle Reduced-Order Modeling of Actuator Performance: We advance the computation of physical modal expansions for unsteady\nincompressible flows. Point of departure is a linearization of the\nNavier-Stokes equations around its fixed point in a frequency domain\nformulation. While the most amplified stability eigenmode is readily identified\nby a power method, the technical challenge is the computation of more damped\nhigher-order eigenmodes. This challenge is addressed by a novel method to\ncompute unstable periodically forced solutions of the linearized Navier-Stokes\nsolution. This method utilizes two key enablers. First, the linear dynamics is\ntransformed by a complex shift of the eigenvalues amplifying the flow response\nat the given frequency of interest. Second, the growth rate is obtained from an\niteration procedure. The method is demonstrated for several wake flows around a\ncircular cylinder, a fluidic pinball, i.e. the wake behind a cluster of\ncylinders, a wall-mounted cylinder, a sphere and a delta wing. The example of\nflow control with periodic wake actuation and forced physical modes paves the\nway for applications of physical modal expansions. These results encourage\nGalerkin models of three-dimensional flows utilizing Navier-Stokes based modes.", "category": "physics_flu-dyn" }, { "text": "An Advanced Kinetic Theory For Morphing Continuum With Inner Structures: Advanced kinetic theory with the Boltzmann-Curtiss equation provides a\npromising tool for polyatomic gas flows, especially for fluid flows containing\ninner structures, such as turbulence, polyatomic gas flows and others. Although\na Hamiltonian-based distribution function was proposed for diatomic gas flow, a\ngeneral distribution function for the generalized Boltzmann-Curtiss equations\nand polyatomic gas flow is still out of reach. With assistance from Boltzmann's\nentropy principle, a generalized Boltzmann-Curtiss distribution for polyatomic\ngas flow is introduced. The corresponding governing equations at equilibrium\nstate are derived and compared with Eringen's morphing (micropolar) continuum\ntheory derived under the framework of rational continuum thermomechanics.\nAlthough rational continuum thermomechanics has the advantages of mathematical\nrigor and simplicity, the presented statistical kinetic theory approach\nprovides a clear physical picture for what the governing equations represent.", "category": "physics_flu-dyn" }, { "text": "Leidenfrost temperature: surface thermal diffusivity and effusivity\n effect: Droplet impact on hot surfaces results in either droplet-surface contact or\ndroplet-surface non-contact, i.e., the Leidenfrost state. The Leidenfrost\ndroplet is levitated upon its vapor, deteriorating the heat transfer. The\nLeidenfrost temperature depends on the thermal properties of the surface, which\nare usually characterized by two parameters: the thermal diffusivity and the\nthermal effusivity. In this paper, the effects of these two parameters on the\nLeidenfrost temperature are clarified experimentally by varying the one of\ninterest while keeping the other one constant. The experimental results\nindicate that the Leidenfrost temperature is affected by the thermal effusivity\nrather than by the thermal diffusivity; the higher the thermal effusivity is,\nthe lower the Leidenfrost temperature; and the increase of the Leidenfrost\ntemperature with the droplet impact velocity is greater for the surface with\nthe lower thermal effusivity. To further understand the experimental findings,\na theoretical model is proposed, which considers the transient heat transfer in\nthe surface. The theoretical analysis shows that the Leidenfrost temperature\nscales as the inverse of the thermal effusivity and the root square of the\nimpact velocity, in good agreement with the experimental results.", "category": "physics_flu-dyn" }, { "text": "Quantum vortex identification method and its application to\n Gross-Pitaevskii simulation: A method to identify a quantum vortex in a three-dimensional Gross-Pitaevskii\nsimulation has been developed. A quantum vortex was identified by the use of\neigenvalues and eigenvectors of the Hessian of the mass density, together with\na condition to distinguish a point to constitute a swirling vortex from other\nconfusing data points. This method has been verified to identify vortex axes in\na Gross-Pitaevskii simulation appropriately, being useful to elucidate various\nstatistics associated with turbulent quantum vortices. This method provides us\nwith a unified approach to studying vortex statistics in the turbulence of both\nclassic and quantum fluids. Our study reveals that the maximum radius of a\nswirling region of a quantum vortex can be as large as sixty times the healing\nlength. The characterization of the vortex core radius relative to the healing\nlength is reported for the first time in this paper. Furthermore, the\ngeometrical natures of vortex axes such as the probability density function of\nthe curvature are characterized by the healing length.", "category": "physics_flu-dyn" }, { "text": "Liquid phase stabilization versus bubble formation at a nanoscale-curved\n interface: We investigate the nature of vapor bubble nucleation near a nanoscale-curved\nconvex liquid-solid interface using two models; an equilibrium Gibbs model for\nhomogenous formation, and a nonequilibrium dynamic van der Waals/diffuse\ninterface model for phase change in an initially cool liquid. Vapor bubble\nformation is shown to occur for sufficiently large radius of curvature and is\nsuppressed for smaller radii. Solid-fluid interactions are accounted for and it\nis shown that liquid-vapor interfacial energy, hence Laplace pressure, has\nlimited influence over bubble formation. The dominant factor is the energetic\ncost of creating the solid-vapor interface from the existing solid-liquid\ninterface, as demonstrated via both equilibrium and non-equilibrium arguments.", "category": "physics_flu-dyn" }, { "text": "Impact of time-dependent non-axisymmetric velocity perturbations on\n dynamo action of von-K\u00e1rm\u00e1n-like flows: We have performed numerical simulations of the kinematic induction equation\nin order to examine the dynamo efficiency of an axisymmetric\nvon-K\\'arm\\'an-like flow subject to time-dependent non-axisymmetric velocity\nperturbations. The numerical model is based on the setup of the French\nVon-K\\'arm\\'an-Sodium dynamo (VKS) and on the flow measurements from a model\nwater experiment conducted at the University of Navarra in Pamplona, Spain. Our\nsimulations show that the interactions of azimuthally drifting flow\nperturbations with the fundamental drift of the magnetic eigenmode (caused by\nthe inevitable equatorial symmetry breaking of the basic flow) essentially\ndetermine the temporal behavior of the dynamo state. We find two distinct\nregimes of dynamo action that depend on the (prescribed) drift frequency of an\n($m=2$) vortex-like flow perturbation. For comparatively slowly drifting\nvortices we observe a narrow window with enhanced growth-rates and a drift of\nthe magnetic eigenmode that is synchronized with the perturbation drift. The\nresonance-like enhancement of the growth-rates takes place when the vortex\ndrift frequency roughly equals the drift frequency of the magnetic eigenmode in\nthe unperturbed system. Outside of this small window, the field generation is\nhampered compared to the unperturbed case, and the field amplitude of the\nmagnetic eigenmode is modulated with approximately twice the vortex drift\nfrequency. The abrupt transition between the resonant regime and the modulated\nregime is identified as an spectral exceptional point where eigenvalues\n(growth-rates and frequencies) and eigenfunctions of two previously independent\nmodes collapse.", "category": "physics_flu-dyn" }, { "text": "Sweeping effect and Taylor's hypothesis via correlation function: We performed high-resolution numerical simulations of hydrodynamic turbulence\nwith and without mean velocity ($U_0=0,10$), and demonstrate the sweeping\neffect. For $U_0=0$, the velocity correlation function, $C({\\bf k},\\tau)$\ndecays with time due to eddy viscosity, but it also shows fluctuations due to\nthe sweeping effect. For $U_0=10$, $C({\\bf k},\\tau)$ exhibits damped\noscillations with the frequency of $U_0 k$ and decay time scale corresponding\nto the $U_0=0$ case. A closer examination of $\\Im[C({\\bf k},\\tau)]$ also\ndemonstrates sweeping effect for $U_0=10$. We also demonstrate that the\nfrequency spectra of the velocity fields measured by real-space probes are\nrespectively $f^{-2}$ and $f^{-5/3}$ for $U_0=0$ and 10; these spectra are\nrelated to the Lagrangian and Eulerian space-time correlations.", "category": "physics_flu-dyn" }, { "text": "Time-evolving Impact of Trees on Street Canyon Microclimate: Nowadays, cities are frequently exposed to heatwaves, worsening the outdoor\nthermal comfort and increasing cooling energy demand in summer. Urban forestry\nis seen as one of the viable and preferable solutions to combating extreme heat\nevents and urban heat island (UHI) in times of climate change. While many\ncities have initiated tree-planting programmes in recent years, the evolving\nimpact of trees on street microclimate, in a time span of up to several\ndecades, remains unclear. We investigate the cooling effects of linden trees in\nfive groups, i.e., 10-20, 20-30, 30-40, 40-60, and 60-100 years old. The leaf\narea index (LAI) and leaf area density (LAD) vary nonlinearly as the trees\ngrow, peaking at different ages. Computational fluid dynamics (CFD) simulations\nsolving microclimate are performed for an idealized street canyon with trees of\nvaried age groups. Turbulent airflow, heat and moisture transport, shortwave\nand longwave radiation, shading and transpiration are fully coupled and solved\nin OpenFOAM. The meteorological data, including air temperature, wind speed,\nmoisture, and shortwave radiation of the heatwave in Zurich (June 2019), are\napplied as boundary conditions. The results show that young trees in the age\ngroup of 10-20 years old provide little heat mitigation at the pedestrian level\nin an extreme heat event. Optimal heat mitigation by trees is observed for the\ngroup of 30-60 years old trees. Finally, the potential impact of growing trees\nas a heat mitigation measure on air ventilation is evaluated.", "category": "physics_flu-dyn" }, { "text": "Multi-modal excitation to model the Quasi-Biennial Oscillation: The Quasi-Biennial Oscillation (QBO) of stratospheric winds is the most\nstriking example of mean-flow generation and reversal by the non-linear\ninteractions of internal waves. Previous studies have used an idealized\nmonochromatic forcing to investigate the QBO. Here we instead force a more\nrealistic continuous wave spectrum. Unexpectedly, spreading the wave energy\nacross a wide frequency range leads to more regular oscillations. We also find\nthat different forcing spectra can yield the same QBO. Multi-modal wave forcing\nis thus essential for understanding wave/mean-flow interactions in nature.", "category": "physics_flu-dyn" }, { "text": "High speed flows with Particles on Demand: Boundary Conditions: The particles on demand (PonD) method is a new kinetic theory model that\nallows for simulation of high speed compressible flows. While standard\nLattice-Boltzmann is limited by a fixed reference frame, significantly reducing\nthe range of applicable of Mach numbers, PonD takes advantage of adaptive\nreference frames to get rid of the restrictions of standard LB and is able to\nsimulate flows at high speeds and with large temperature gradients. Previously,\nPonD has been shown to be a viable alternative for simulation of flows with\nstrong discontinuities and for detonation modelling. However, treatment of\nflows with complex boundaries has been lacking. Here, we present PonD augmented\nwith a non-equilibrium extrapolation based boundary condition. We present\nseveral compressible test cases such as shock-vortex interaction in the\nSchardin's Problem and supersonic flow over a two-dimensional cylinder at Mach\nnumbers up to 5. We observe that the results agree well with literature, paving\nthe way for a kinetic theory based approach for simulating compressible flows\nin realistic scenarios.", "category": "physics_flu-dyn" }, { "text": "A strongly coupled immersed boundary method for fluid-structure\n interaction that mimics the efficiency of stationary body methods: Strongly coupled immersed boundary (IB) methods solve the nonlinear fluid and\nstructural equations of motion simultaneously for strongly enforcing the\nno-slip constraint on the body. Handling this constraint requires solving\nseveral large dimensional systems that scale by the number of grid points in\nthe flow domain even though the nonlinear constraints scale only by the small\nnumber of points used to represent the fluid-structure interface. These costly\nlarge scale operations for determining only a small number of unknowns at the\ninterface creates a bottleneck to efficiently time-advancing strongly coupled\nIB methods. In this manuscript, we present a remedy for this bottleneck that is\nmotivated by the efficient strategy employed in stationary-body IB methods\nwhile preserving the favorable stability properties of strongly coupled\nalgorithms -- we precompute a matrix that encapsulates the large dimensional\nsystem so that the prohibitive large scale operations need not be performed at\nevery time step. This precomputation process yields a modified system of\nsmall-dimensional constraint equations that is solved at minimal computational\ncost while time advancing the equations. We also present a parallel\nimplementation that scales favorably across multiple processors. The accuracy,\ncomputational efficiency and scalability of our approach are demonstrated on\nseveral two dimensional flow problems. Although the demonstration problems\nconsist of a combination of rigid and torsionally mounted bodies, the\nformulation is derived in a more general setting involving an arbitrary number\nof rigid, torsionally mounted, and continuously deformable bodies.", "category": "physics_flu-dyn" }, { "text": "Calculation of Evaluation Variables for High Gradient Magnetic\n Separation with an Idealized Capture Model: This paper regards feed mine as a mixture of intergrowths and pure\nnon-magnetic mineral particles, presents a method to calculate the evaluation\nvariables such as grade and recovery in high gradient magnetic separation\n(HGMS). A idealized capture model is constructed in which the interaction\nbetween particles is not taken into account and only for the initial\naggregation condition that the separator has the highest capture efficiency. In\nthe model we adopt the functions that use nominal particle radius and magnetic\nmineral content as independent variables to describe volume fraction\ndistribution and capture efficiency of intergrowths respectively. Through\nadding multi-wire magnetic fields and setting periodic boundary conditions in\nflow field analysis, we modify the computational domain of the single-wire\ncapture theory to a element domain that periodically appears in the multi-wire\nmatrix. By means of finite element software, particle trajectories, flow field\nand magnetic field are clearly exhibited, and then capture efficiency function\nis obtained by interpolation method. The calculated evaluation variables\ntheoretically represent the best performance of magnetic separator for a given\nfeed. They can assist mineral engineers to evaluate or compare the effects of\ndifferent magnetic separation systems in advance. We use removal of iron\nimpurity from kaolin as an example to illustrate the presented calculation\nmethod. The results quantitatively compare the evaluation variables of the\nseparation at different magnetic fields and show that the advantage of higher\nmagnetic field in separation efficiency decreases with the increase of\nsaturation magnetization of magnetic mineral.", "category": "physics_flu-dyn" }, { "text": "Double-diffusive convection and baroclinic instability in a\n differentially heated and initially stratified rotating system: the barostrat\n instability: A water-filled differentially heated rotating annulus with initially prepared\nstable vertical salinity profiles is studied in the laboratory. Based on\ntwo-dimensional horizontal particle image velocimetry (PIV) data, and infrared\ncamera visualizations, we describe the appearance and the characteristics of\nthe baroclinic instability in this original configuration. First, we show that\nwhen the salinity profile is linear and confined between two non stratified\nlayers at top and bottom, only two separate shallow fluid layers can be\ndestabilized. These unstable layers appear nearby the top and the bottom of the\ntank with a stratified motionless zone between them. This laboratory\narrangement is thus particularly interesting to model geophysical or\nastrophysical situations where stratified regions are often juxtaposed to\nconvective ones. Then, for more general but stable initial density profiles,\nstatistical measures are introduced to quantify the extent of the baroclinic\ninstability at given depths and to analyze the connections between this\ndepth-dependence and the vertical salinity profiles. We find that, although the\npresence of stable stratification generally hinders full-depth overturning,\ndouble-diffusive convection can yield development of multicellular sideways\nconvection in shallow layers and subsequently to a multilayered baroclinic\ninstability. Therefore we conclude that by decreasing the characteristic\nvertical scale of the flow, stratification may even enhance the formation of\ncyclonic and anticyclonic eddies (and thus, mixing) in a local sense.", "category": "physics_flu-dyn" }, { "text": "Computational Performance of a LES Solver for Supersonic Jet Flow\n Applications: An in-house large eddy simulation tool is developed in order to reproduce\nhigh fidelity results of compressible jet flows. The large eddy simulation\nformulation is written using the finite difference approach, with an explicit\ntime integration and using a second order spatial discretization. The energy\nequation is carefully discretized in order to model the energy equation of the\nfiltered Navier-Stokes formulation. Such nu\\-me\\-ri\\-cal studies are very\nexpensive and demand high performance computing. Message passage interface\nprotocols are implemented into the code in order to perform parallel\ncomputations. The present work addresses the computational performance of the\nsolver running on up to 400 processors in parallel. Different mesh\nconfigurations, whose size varies from approximately 5.9 million points to\napproximately 1.0 billion points, are evaluate in the current paper. Speedup\nand efficiency curves are evaluated in order to assess the strong scalability\nof the solver.", "category": "physics_flu-dyn" }, { "text": "The Generalized Quasilinear Approximation: Application to Zonal Jets: Quasilinear theory is often utilized to approximate the dynamics of fluids\nexhibiting significant interactions between mean flows and eddies. In this\npaper we present a generalization of quasilinear theory to include dynamic mode\ninteractions on the large scales. This generalized quasilinear (GQL)\napproximation is achieved by separating the state variables into large and\nsmall zonal scales via a spectral filter rather than by a decomposition into a\nformal mean and fluctuations. Nonlinear interactions involving only small zonal\nscales are then removed. The approximation is conservative and allows for\nscattering of energy between small-scale modes via the large scale (through\nnon-local spectral interactions). We evaluate GQL for the paradigmatic problems\nof the driving of large-scale jets on a spherical surface and on the beta-plane\nand show that it is accurate even for a small number of large-scale modes. As\nthis approximation is formally linear in the small zonal scales it allows for\nthe closure of the system and can be utilized in direct statistical simulation\nschemes that have proved an attractive alternative to direct numerical\nsimulation for many geophysical and astrophysical problems.", "category": "physics_flu-dyn" }, { "text": "Accelerations of large inertial particles in turbulence: Understanding the dynamics of material objects advected by turbulent flows is\na long standing question in fluid dynamics. In this perspective article we\nfocus on the characterization of the statistical properties of non-interacting\nfinite-sized massive spherical particles advected by a vigorous turbulent flow.\nWe study the fluctuations and temporal correlations of particle accelerations\nand explore their behaviours with respect to the particle size and the particle\nmass density by means of fully-resolved numerical simulations. We observe that\nthe measured trends can not be interpreted as the simple multiplicative\ncombination of the two dominant effects: the spatial filtering of fluid\naccelerations and the added-mass-adjusted fluid-to-particle density ratio. We\nargue that other hydrodynamical forces or effects, e.g. preferential flow\nsampling, have still a significant role even at the largest particle sizes,\nwhich rich here the integral scale of turbulence.", "category": "physics_flu-dyn" }, { "text": "A fast, matrix-based method to perform omnidirectional pressure\n integration: Experimentally-measured pressure fields play an important role in\nunderstanding many fluid dynamics problems. Unfortunately, pressure fields are\ndifficult to measure directly with non-invasive, spatially resolved\ndiagnostics, and the derivation of pressure from velocity data has proven very\nsensitive to amplification of error. Omnidirectional line integration methods\nare usually more accurate and robust to these effects as compared to implicit\nPoisson equations, but have seen slower uptake due to the higher computational\nand memory costs, particularly in 3D domains. This paper demonstrates how\nomnidirectional line integration approaches can be converted to a matrix\ninversion problem. This novel formulation uses an iterative approach so that\nthe boundary conditions are updated each step, preserving the convergence\nbehavior of omnidirectional schemes while also keeping the computational\nefficiency of Poisson solvers. This method is implemented in Matlab and also as\na GPU-accelerated code in CUDA-C++. The behavior of the new method is\ndemonstrated on 2D and 3D synthetic and experimental data. Three-dimensional\ngrid sizes of up to 125 million grid points are tractable with this method,\nopening exciting opportunities to perform volumetric pressure field estimation\nfrom 3D PIV measurements.", "category": "physics_flu-dyn" }, { "text": "Viscoelastic surface instabilities: We review three different types of viscoelastic surface instabilities: The\nRayleigh -- Plateau, the Saffman -- Taylor and the Faraday instability. These\ninstabilities are classical examples of hydrodynamic surface instabilities. The\naddition of a small amount of polymers to pure water can alter its flow\nbehavior drastically and the type of instability may change not only\nquantitatively but also qualitatively. We will show that some of the observed\nnew phenomena can be explained by the use of simple rheological models that\ncontain most of the underlying physical mechanisms leading to the instability.\nA quantitative description however is often only possible close to the onset of\nthe instability or for weak deviations from Newtonian behavior. A complete\ntheoretical description is still lacking when the system is driven far from\nequilibrium or for fluids with strong non-Newtonian behavior.", "category": "physics_flu-dyn" }, { "text": "Shape evolution of numerically obtained subaqueous barchan dunes: In the realm of granular bedforms, barchan dunes are strong attractors that\ncan be found in rivers, terrestrial deserts and other planetary environments.\nThese bedforms are characterized by a crescentic shape, which, although robust,\npresents different scales according to the environment they are in, their\nlength scale varying from the decimeter under water to the kilometer on Mars.\nIn addition to the scales of bedforms, the transport of grains presents\nsignificant differences according to the nature of the entraining fluid, so\nthat the growth of barchans rests to be fully understood. Given the smaller\nlength and time scales of the aquatic case, subaqueous barchans are the ideal\nobject to study the growth of barchan dunes. In the present paper, we reproduce\nnumerically the experiments of Alvarez and Franklin [Phys. Rev. E 96, 062906\n(2017) and Phys. Rev. Lett. 121, 164503 (2018)] on the shape evolution of\nbarchans from their initiation until they have reached a stable shape. We\ncomputed the bed evolution by using computational fluid dynamics - discrete\nelement method (CFD-DEM), where we coupled DEM with large eddy simulation (LES)\nfor the same initial and boundary conditions of experiments, performed in a\nclosed-conduit channel where initially conical heaps evolved to single barchans\nunder the action of a water flow in turbulent regime. Our simulations captured\nwell the evolution of the initial pile toward a barchan dune in both the\nbedform and grain scales, with the same characteristic time and lengths\nobserved in experiments. In addition, we obtained the local granular flux and\nthe resultant force acting on each grain, the latter not yet previously\nmeasured nor computed. This shows that the present method is appropriate to\nnumerical computations of bedforms, opening new possibilities for accessing\ndata that are not available from current experiments.", "category": "physics_flu-dyn" }, { "text": "Acoustic microstreaming near a plane wall due to a pulsating free/coated\n bubble: Nyborg result revisited and vortices: Acoustic microstreaming due to an oscillating microbubble is analytically\ninvestigated to obtain the circular streaming motion adjacent to a nearby wall.\nClassical theory due to Nyborg is carefully derived in the radial coordinates.\nThe theory is used to obtain the flow field and the vortical motion caused by\nthe microbubble motion. The length of the vertices are decreasing when the\nmicrobubble is excited at distances close to the rigid wall, while the maximum\nshear stress is increasing.", "category": "physics_flu-dyn" }, { "text": "Numerical analysis of the entry flow for continuous shear thickening\n fluids in circular microtubes: Due to their nature, using shear thickening fluids (STFs) in engineering\napplications has sparked an interest in developing energy-dissipating systems,\nsuch as damping devices or shock absorbers. The Rheinforce technology allows\nthe design of customized energy dissipative composites by embedding\nmicrofluidic channels filled with STFs in a scaffold material. One of the\nreasons for using microfluidic channels is that their shape can be numerically\noptimized to control pressure drop (also known as rectifiers); thus, by\ncontrolling the pressure drop, it is possible to control the energy dissipated\nby the viscous effect. Upon impact, the fluid is forced to flow through the\nmicrochannel, experiencing the typical entry flow until it reaches the fully\ndeveloped flow. It is well-known for Newtonian fluid that the entrance flow is\nresponsible for a non-negligible percentage of the total pressure drop in the\nfluid; therefore, an analysis of the fluid flow at the entry region for STFs is\nof paramount importance for an accurate design of the Rheinforce composites.\nThis analysis has been numerically performed before for shear thickening fluids\nmodelled by a power-law model; however, as this constitutive model represents a\ncontinuously growing viscosity between end-viscosity plateau values, it is not\nrepresentative of the characteristic viscosity curve of shear thickening\nfluids, which typically exhibit a three-regions shape\n(thinning-thickening-thinning). For the first time, the influence of these\nthree regions on the entry flow on an axisymmetric pipe is analysed. 2D\nnumerical simulations have been performed for four STFs consisting of four\ndispersions of fumed silica nanoparticles in polypropylene glycol varying\nconcentration (7.5-20wt%) modelled by Khandavalli, et al. [1].", "category": "physics_flu-dyn" }, { "text": "Logarithmic temperature profiles in turbulent Rayleigh-B\u00e9nard\n convection: We report results for the temperature profiles of turbulent Rayleigh-B\\'enard\nconvection (RBC) in the interior of a cylindrical sample of aspect ratio\n$\\Gamma \\equiv D/L = 0.50$ ($D$ and $L$ are the diameter and height\nrespectively). Results from experiment over the Rayleigh number range $4\\times\n10^{12} \\alt Ra \\alt 10^{15}$ for a Prandtl number $\\Pra \\simeq 0.8$ and from\ndirect numerical simulation (DNS) at $Ra = 2 \\times 10^{12}$ for $\\Pra = 0.7$\nare presented. We find that the temperature varies as $A*ln(z/L) + B$ where $z$\nis the distance from the bottom or top plate. This is the case in the classical\nas well as in the ultimate state of RBC. From DNS we find that $A$ in the\nclassical state decreases in the radial direction as the distance from the side\nwall increases and becomes small near the sample center.", "category": "physics_flu-dyn" }, { "text": "Observation of resonant interactions among surface gravity waves: We experimentally study resonant interactions of oblique surface gravity\nwaves in a large basin. Our results strongly extend previous experimental\nresults performed mainly for perpendicular or collinear wave trains. We\ngenerate two oblique waves crossing at an acute angle, while we control their\nfrequency ratio, steepnesses and directions. These mother waves mutually\ninteract and give birth to a resonant wave whose properties (growth rate,\nresonant response curve and phase locking) are fully characterized. All our\nexperimental results are found in good quantitative agreement with four-wave\ninteraction theory with no fitting parameter. Off-resonance experiments are\nalso reported and the relevant theoretical analysis is conducted and validated.", "category": "physics_flu-dyn" }, { "text": "Acoustic band gap and coherence behavior in a periodic two-dimensional\n system: Here we show a complete band gap in acoustic propagation in water with\nregularly placed air-cylinders. Also for the first time, we report that in the\ngap, all cylinders oscillate in phase but exactly out of the phase of a wave\nsource located inside the cylinder array.", "category": "physics_flu-dyn" }, { "text": "Multilayer models for shallow two-phase debris flows with dilatancy\n effects: We present here a multilayer model for shallow grain-fluid mixtures with\ndilatancy effects. It can be seen as a generalization of the depth-averaged\nmodel presented in Bouchut et al. (2016), that includes dilatancy effects by\nconsidering a two-layer model, a mixture grain-fluid layer and an upper fluid\nlayer, to allow the exchange of fluid between them. In the present work the\napproximation of the mixture layer is improved including normal variations of\nthe velocities and concentrations of the two phases thanks to the multilayer\napproach. In the model presented here dilatancy effects induce in particular a\nnon-hydrostatic pressure for both phases related to the excess pore fluid\npressure. We identified here one of the main numerical difficulty of solving\ntwo-phase shallow debris flows models: the strongly non-linear behaviour and\nabrupt changes of the excess pore fluid pressure when starting from\nnon-equilibrium conditions. We propose a simplified approach to approximate the\nexcess pore fluid pressure in the simple case of uniform flows. Our method\nmakes it possible to introduce two or three layers in the normal directions\nwith a reasonable approximation. Analytical solutions for uniform grain-fluid\nflows over inclined planes, with and without side wall friction, are calculated\nand compared to the proposed model. In the numerical results, we observe that\nthe proposed model with a two layer description of the mixture accurately\nrepresents the velocity measured at the surface of the mixture in the\nlaboratory experiments. This is obviously poorly represented by the\ndepth-averaged velocity in single-layer models. Our numerical results show a\nsignificant impact of the parameters involved in dilatancy law, in particular\non the calculation of the time evolution of the excess pore fluid pressure.", "category": "physics_flu-dyn" }, { "text": "Scattering of oblique waves by permeable vertical flexible membrane wave\n barriers: The interaction of obliquely incident surface gravity waves with a vertical\nflexible permeable membrane wave barrier is investigated in the context of\nthree-dimensional linear wave-structure interaction theory. A general\nformulation for wave interaction with permeable submerged vertical membrane is\ngiven. The analytic solution of the physical problem is obtained by using\neigenfunction expansion method, and boundary element method has been used to\nget the numerical solution. In the boundary element method, since the boundary\ncondition on the membrane is not known in advance, membrane motions and\nvelocity potentials are solved simultaneously. From the general formulation of\nthe submerged membrane barrier, the performance of bottom-standing,\nsurface-piercing and fully extended membrane wave barriers are analyzed for\nvarious wave and structural parameters. It is found that the efficiency of the\nsubmerged, surface-piercing and bottom-standing membrane wave barriers can be\nenhanced in waves for certain design conditions. From the analysis of various\nmembrane configurations and parameters, it can be concluded that permeable\nmembrane wave barrier can function as a very effective breakwater if it is\nproperly designed.", "category": "physics_flu-dyn" }, { "text": "Streamwise decay of localized states in channel flow: Channel flow, the pressure driven flow between parallel plates, has exact\ncoherent structures that show various degrees of localization. For states which\nare localized in streamwise direction but extended in spanwise direction, we\nshow that they are exponentially localized, with decay constants that are\ndifferent on the upstream and downstream side. We extend the analysis of Brand\nand Gibson [J. Fluid Mech. 750, R1 (2014)] for stationary states to the case of\nadvected structures that is needed here, and derive expressions for the decay\nin terms of eigenvalues and eigenfunctions of certain second order differential\nequations. The results are in very good agreement with observations on exact\ncoherent structures of different transversal wave lengths.", "category": "physics_flu-dyn" }, { "text": "Characterization of a capillary driven flow in microgravity by means of\n optical technique: The motion of a gas-liquid interface along a solid wall is influenced by the\ncapillary forces resulting from the interface's shape and its interaction with\nthe solid, where it forms a dynamic contact angle. Capillary models play a\nsignificant role in the management of cryogenic propellants in space, where\nsurface tension dominates the behaviour of gas-liquid interfaces. Yet, most\nempirical models have been derived in configurations dominated by viscous\nforces. In this study, we experimentally investigate the wetting of a\nlow-viscosity, highly wetting fluid in a reduced gravity environment. Our setup\nconsisted of a transparent and diverging U-tube in which capillary forces\nsustain the liquid motion. Combining Particle Image Velocimetry (PIV) and\nhigh-speed backlighting visualization, the experimental campaign allowed for\nmeasuring the interface evolution and the velocity field within the liquid\nunder varying gravity levels. This work reports on the preliminary results from\nthe image velocimetry and shows that the velocity profile within the tube is\nclose to parabolic until a short distance from the interface. Nevertheless,\nclassic 1D models for capillary rise face difficulties reproducing the\ninterface dynamics, suggesting that the treatment of the surface tension in\nthese problems must be reviewed.", "category": "physics_flu-dyn" }, { "text": "Front pinning in capillary filling of chemically coated channels: The dynamics of capillary filling in the presence of chemically coated\nheterogeneous boundaries is investigated, both theoretically and numerically.\nIn particular, by mapping the equations of front motion onto the dynamics of a\ndissipative driven oscillator, an analytical criterion for front pinning is\nderived, under the condition of diluteness of the coating spots. The criterion\nis tested against two dimensional Lattice Boltzmann simulations, and found to\nprovide satisfactory agreement as long as the width of the front interface\nremains much thinner than the typical heterogeneity scale of the chemical\ncoating.", "category": "physics_flu-dyn" }, { "text": "Linear Rayleigh-B\u00e9nard stability of a transversely-isotropic fluid: Suspended fibres significantly alter fluid rheology, as exhibited in for\nexample solutions of DNA, RNA and synthetic biological nanofibres. It is of\ninterest to determine how this altered rheology affects flow stability.\nMotivated by the fact thermal gradients may occur in biomolecular analytic\ndevices, and recent stability results, we examine the problem of\nRayleigh-B\\'enard convection of the transversely-isotropic fluid of Ericksen. A\ntransversely-isotropic fluid treats these suspensions as a continuum with an\nevolving preferred direction, through a modified stress tensor incorporating\nfour viscosity-like parameters. We consider the linear stability of a\nstationary, passive, transversely-isotropic fluid contained between two\nparallel boundaries, with the lower boundary at a higher temperature than the\nupper. To determine the marginal stability curves the Chebyshev collocation\nmethod is applied, and we consider a range of initially uniform preferred\ndirections, from horizontal to vertical, and three orders of magnitude in the\nviscosity-like anisotropic parameters. Determining the critical wave and\nRayleigh numbers we find that transversely-isotropic effects delay the onset of\ninstability; this effect is felt most strongly through the incorporation of the\nanisotropic shear viscosity, although the anisotropic extensional viscosity\nalso contributes. Our analysis confirms the importance of anisotropic rheology\nin the setting of convection.", "category": "physics_flu-dyn" }, { "text": "Impact of pressure dissipation on fluid injection into layered aquifers: Carbon dioxide (CO2) capture and subsurface storage is one method for\nreducing anthropogenic CO2 emissions to mitigate climate change. It is well\nknown that large-scale fluid injection into the subsurface leads to a buildup\nin pressure that gradually spreads and dissipates through lateral and vertical\nmigration of water. This dissipation can have an important feedback on the\nshape of the CO2 plume during injection, and the impact of vertical pressure\ndissipation, in particular, remains poorly understood. Here, we investigate the\nimpact of lateral and vertical pressure dissipation on the injection of CO2\ninto a layered aquifer system. We develop a compressible, two-phase model that\ncouples pressure dissipation to the propagation of a CO2 gravity current. We\nshow that our vertically integrated, sharp-interface model is capable of\nefficiently and accurately capturing water migration in a layered aquifer\nsystem with an arbitrary number of aquifers. We identify two limiting cases ---\n`no leakage' and `strong leakage' --- in which we derive analytical expressions\nfor the water pressure field for the corresponding single-phase injection\nproblem. We demonstrate that pressure dissipation acts to suppress the\nformation of an advancing CO2 tongue during injection, resulting in a plume\nwith a reduced lateral extent. The properties of the seals and the number of\naquifers determine the strength of pressure dissipation and subsequent coupling\nwith the CO2 plume. The impact of pressure dissipation on the shape of the CO2\nplume is likely to be important for storage efficiency and security.", "category": "physics_flu-dyn" }, { "text": "Stretching and folding versus cutting and shuffling: An illustrated\n perspective on mixing and deformations of continua: We compare and contrast two types of deformations inspired by mixing\napplications -- one from the mixing of fluids (stretching and folding), the\nother from the mixing of granular matter (cutting and shuffling). The\nconnection between mechanics and dynamical systems is discussed in the context\nof the kinematics of deformation, emphasizing the equivalence between stretches\nand Lyapunov exponents. The stretching and folding motion exemplified by the\nbaker's map is shown to give rise to a dynamical system with a positive\nLyapunov exponent, the hallmark of chaotic mixing. On the other hand, cutting\nand shuffling does not stretch. When an interval exchange transformation is\nused as the basis for cutting and shuffling, we establish that all of the map's\nLyapunov exponents are zero. Mixing, as quantified by the interfacial area per\nunit volume, is shown to be exponentially fast when there is stretching and\nfolding, but linear when there is only cutting and shuffling. We also discuss\nhow a simple computational approach can discern stretching in discrete data.", "category": "physics_flu-dyn" }, { "text": "Thin Film Rupture from the Atomic Scale: The retraction of thin films, as described by the Taylor-Culick (TC) theory,\nis subject to widespread debate, particularly for films at the nanoscale. We\nuse non-equilibrium molecular dynamics simulations to explore the validity of\nthe assumptions used in continuum models, by tracking the evolution of holes in\na film. By deriving a new mathematical form for the surface shape and\nconsidering a locally varying surface tension at the front of the retracting\nfilm, we reconcile the original theory with our simulation data to recover a\ncorrected TC speed valid at the nanoscale.", "category": "physics_flu-dyn" }, { "text": "Small scale anisotropy in turbulent shearless mixing: The generation of small-scale anisotropy in turbulent shearless mixing is\nnumerically investigated. Data from direct numerical simulations at Taylor\nReynolds' numbers between 45 and 150 show that there is not only a significant\ndeparture of the longitudinal velocity derivative moments from the values found\nin homogeneous and isotropic turbulence, but that the variation of skewness has\nan opposite sign for the components across the mixing layer and parallel to it.\nThe anisotropy induced by the presence of a kinetic energy gradient has a very\ndifferent pattern from the one generated by an homogeneous shear. The\ntransversal derivative moments in the mixing are in fact found to be very\nsmall, which highlights that smallness of the transversal moments is not a\nsufficient condition for isotropy.", "category": "physics_flu-dyn" }, { "text": "Numerical Simulation of Vortex-Induced Vibration With Bistable Springs :\n Consistency with the Equilibrium Constraint: We present results from 2-D numerical simulations based on Immersed Boundary\nMethod of a cylinder in uniform fluid flow attached to bistable springs\nundergoing Vortex-Induced Vibrations (VIV). The elastic spring potential for\nthe bistable springs, consisting of 2 potential wells, is defined by the\nspacing between the potential minima and the depth of the potential wells. We\nperform simulations of VIV with linear spring, as well as bistable springs with\ntwo different inter-well separations, over a wide range of reduced velocity. As\nexpected, large oscillation amplitudes correspond to lock-in of the lift force\nwith the natural frequency of the spring-mass system. The range of reduced\nvelocity over which lock-in occurs is significantly higher for VIV with\nbistable springs compared to VIV with linear springs, although the maximum\npossible amplitude appears to be independent of the spring type. For VIV with\nbistable springs, the cylinder undergoes double-well oscillations in the\nlock-in regime. Range of reduced velocity over which lock-in occurs increases\nwhen the inter-well distance is reduced. The vortex shedding patterns and\namplitude trends look similar at the same equivalent reduced velocity for the\ndifferent springs. The results here are consistent with our prior theory, in\nwhich we propose a new \"Equilibrium-Constraint (EC)\" based on average kinetic\nenergy budget of the structure. For a given spring potential, the intersection\nof natural frequency curves with the EC curve yields the possible range of\nreduced velocities over which lock-in should occur. Our numerical simulations\nshow a collapse of the amplitude-versus-structure frequency data for all the\nsimulations onto roughly the same curve, thus supporting the existence of the\nEC, and providing an explanation for the trends in the VIV oscillations.", "category": "physics_flu-dyn" }, { "text": "Comparison of the effect of horizontal vibrations on interfacial waves\n in a two-layer system of inviscid liquids to effective gravity inversion: We study the waves at the interface between two thin horizontal layers of\nimmiscible liquids subject to high-frequency tangential vibrations. Nonlinear\ngoverning equations are derived for the cases of two- and three-dimensional\nflows and arbitrary ratio of layer thicknesses. The derivation is performed\nwithin the framework of the long-wavelength approximation, which is relevant as\nthe linear instability of a thin-layers system is long-wavelength. The dynamics\nof equations is integrable and the equations themselves can be compared to the\nBoussinesq equation for the gravity waves in shallow water, which allows one to\ncompare the action of the vibrational field to the action of the gravity and\nits possible effective inversion.", "category": "physics_flu-dyn" }, { "text": "Turbulent boundary layers around wing sections up to Rec = 1,000,000: Reynolds-number effects in the adverse-pressure-gradient (APG) turbulent\nboundary layer (TBL) developing on the suction side of a NACA4412 wing section\nare assessed in the present work. To this end, we analyze four cases at\nReynolds numbers based on freestream velocity and chord length ranging from Rec\n= 100, 000 to 1,000,000, all of them with 5 degree angle of attack. The results\nof four well-resolved large-eddy simulations (LESs) are used to characterize\nthe effect of Reynolds number on APG TBLs subjected to approximately the same\npressure-gradient distribution (defined by the Clauser pressure-gradient\nparameter beta). Comparisons of the wing profiles with zero-pressure-gradient\n(ZPG) data at matched friction Reynolds numbers reveal that, for approximately\nthe same beta distribution, the lower-Reynolds-number boundary layers are more\nsensitive to pressure-gradient effects. This is reflected in the values of the\ninner-scaled edge velocity Ue+ , the shape factor H, the components of the\nReynolds-stress tensor in the outer region and the outer-region production of\nturbulent kinetic energy. This conclusion is supported by the larger\nwall-normal velocities and outer-scaled fluctuations observed in the lower-Rec\ncases.Thus, our results suggest that two complementing mechanisms contribute to\nthe development of the outer region in TBLs and the formation of large-scale\nenergetic structures: one mechanism associated with the increase in Reynolds\nnumber, and another one connected to the APG. Future extensions of the present\nwork will be aimed at studying the differences in the outer-region energizing\nmechanisms due to APGs and increasing Reynolds number.", "category": "physics_flu-dyn" }, { "text": "Statistical Nonlocality of Dynamically Coherent Structures: We introduce a class of stochastic advection problems amenable to analysis of\nturbulent transport. The statistics of the flow field are represented as a\ncontinuous time Markov process, a choice that captures the intuitive notion of\nturbulence as moving from one coherent structure to another. We obtain closed\nform expressions for the turbulent transport operator without invoking\napproximations. We recover the classical estimate of turbulent transport as a\ndiffusivity tensor, the components of which are the integrated auto-correlation\nof the velocity field, in the limit that the operator becomes local in space\nand time.", "category": "physics_flu-dyn" }, { "text": "Entropic multi-relaxation lattice Boltzmann scheme for turbulent flows: We present three dimensional realizations of the model introduced recently by\n(Karlin, B\\\"osch, Chikatamarla, Phys. Rev. E 2014) and review the role of the\nentropic stabilizer. The presented models achieve outstanding numerical\nstability in presence of turbulent high Reynolds number flows. We report\naccurate results for low order moments for homogeneous isotropic decaying\nturbulence and second order grid convergence for most assessed statistical\nquantities. The explicit and efficient nature of the scheme renders it a very\npromising candidate for both engineering and scientific purposes in the\nvicinity of highly turbulent flows.", "category": "physics_flu-dyn" }, { "text": "Color of turbulence: In this paper, we address the problem of how to account for second-order\nstatistics of turbulent flows using low-complexity stochastic dynamical models\nbased on the linearized Navier-Stokes equations. The complexity is quantified\nby the number of degrees of freedom in the linearized evolution model that are\ndirectly influenced by stochastic excitation sources. For the case where only a\nsubset of velocity correlations are known, we develop a framework to complete\nunavailable second-order statistics in a way that is consistent with\nlinearization around turbulent mean velocity. In general, white-in-time\nstochastic forcing is not sufficient to explain turbulent flow statistics. We\ndevelop models for colored-in-time forcing using a maximum entropy formulation\ntogether with a regularization that serves as a proxy for rank minimization. We\nshow that colored-in-time excitation of the Navier-Stokes equations can also be\ninterpreted as a low-rank modification to the generator of the linearized\ndynamics. Our method provides a data-driven refinement of models that originate\nfrom first principles and captures complex dynamics of turbulent flows in a way\nthat is tractable for analysis, optimization, and control design.", "category": "physics_flu-dyn" }, { "text": "Influence of Sequential Changes in the Crude Oil-Water Interfacial\n Tension on Spontaneous Imbibition in Oil-Wet Sandstone: Crude oil-water interfacial tension in petroleum reservoir is reduced or\nincreased due to surfactant injection or surfactant retention, respectively.\nChanges in the interfacial tension crucially attribute to a governing capillary\npressure and hence an oil displacement in spontaneous imbibition process. The\ncurrent study attempts to elucidate an influence of such changes on spontaneous\nimbibition by replacing surfactant concentration consecutively with two\napproaches: sequential decrease and sequential increase in the interfacial\ntension. Two fluid flow directions were examined simultaneously: co-current and\ncounter-current flows. With strongly oil-wet wettability, capillarity was a\nresisting element to oil displacement and therefore controlled by the oil-water\ninterfacial tension. The sequential-reduced interfacial tension was found to\nweaken such resisting capillary force gradually and resulted in consecutive\nincremental oil production. On the contrary, the sequential-increased\ninterfacial tension initiated the lowest interfacial tension fluid that\nproduced an immediate large amount of oil, but did not much displace further\noil. The current study observed a greater oil recovery obtained from a\nsequential reduction in the interfacial tension scheme (26.9%) compared to a\nconventional single reduction scheme (22.4%), with both schemes attaining same\ninterfacial tension at last. In counter-current imbibition, same\ncharacteristics of oil displacement were observed as in co-current imbibition,\nwith less oil produced and less sensitive to fluid changes due to negligible\ngravitational contribution.", "category": "physics_flu-dyn" }, { "text": "Dynamic wetting experiment with nitrogen in a quasi-capillary tube: This work investigates the wetting dynamics of cryogenic fluids in\ninertia-dominated conditions. We experimentally characterized an oscillating\ngas-liquid interface of liquid nitrogen in a partially filled U-shaped quartz\ntube. The experiments were carried out in controlled cryogenic conditions, with\ninterface oscillations produced by releasing the liquid column from an\nunbalanced position and having nitrogen vapor as the only ullage gas. During\nthe experiments, the interface shape was tracked via image processing and used\nto fit a model from which the contact angle could be accurately determined. The\nresults show that the dynamic contact angle evolution in advancing conditions\nis linearly linked to the Capillary number, with a slope depending on whether\nthe interface moves over a dry or a pre-wet surface. However, the contact angle\nremains close to the one at equilibrium in receding conditions. To analyze the\nrelation between contact angle and interface dynamics, we define an equivalent\ncontact angle as the one that would make a spherical interface produce the same\ncapillary pressure drop as the actual interface shape. The evolution of this\nequivalent contact angle proved to be independent of the evolution of the\nactual one, suggesting that the interface shape is not influenced by it.\nFinally, a theoretical analysis of the interface motion using a simplified\nmodel shows that viscous forces dominate the damping of the interface for small\ntube sizes, while gravity and inertial forces dominate the oscillating dynamics\nof the liquid column for larger tubes.", "category": "physics_flu-dyn" }, { "text": "Hemodynamics of the heart's left atrium based on a Variational\n Multiscale-LES numerical method: In this paper, we investigate the hemodynamics of a left atrium (LA) by\nproposing a computational model suitable to provide physically meaningful fluid\ndynamics indications and detailed blood flow characterization. In particular,\nwe consider the incompressible Navier-Stokes equations in Arbitrary Lagrangian\nEulerian (ALE) formulation to deal with the LA domain under prescribed motion.\nA Variational Multiscale (VMS) method is adopted to obtain a stable formulation\nof the Navier-Stokes equations discretized by means of the Finite Element\nmethod and to account for turbulence modeling based on Large Eddy Simulation\n(LES). The aim of this paper is twofold: on one hand to improve the general\nunderstanding of blood flow in the human LA in normal conditions; on the other,\nto analyse the effects of the turbulence VMS-LES method on a situation of blood\nflow which is neither laminar, nor fully turbulent, but rather transitional as\nin LA. Our results suggest that if relatively coarse meshes are adopted, the\nadditional stabilization terms introduced by the VMS-LES method allow to better\npredict transitional effects and cycle-to-cycle blood flow variations than the\nstandard SUPG stabilization method.", "category": "physics_flu-dyn" }, { "text": "The volume-filtering immersed boundary method: We present a novel framework to deal with static and moving immersed\nboundaries (IB). In this strategy, called Volume-Filtering Immersed Boundary\n(VFIB) method, transport equations are derived by filtering the Navier-Stokes\nequations and accounting for stresses at the solid-fluid interface. The result\nis that boundary conditions that normally apply on the interface are\ntransformed into bodyforces that apply on the right-hand side of the filtered\ntransport equations. The filter width acts as a parameter that controls the\nlevel of resolution. The IB is considered well-resolved if the filter width is\nmuch smaller than the characteristic corrugation scale of the interface. There\nare several innovations in this IB method. First, it sheds light on the role of\nthe internal flow which arises when the transport equations are solved inside\nthe IB. We show that, it is essential to separate stresses due to the external\nand internal fluids in order to get accurate forces and provide a method to do\nso. Second, we show that volumes associated with Lagrangian forcing points on\nthe boundary depend on the local topology of the surface. We provide a method\nto compute these volumes using a triangle tessellation of the interface and the\nsurface density function. Third, we provide an efficient procedure to compute\nthe solid volume fraction, thus, enabling tagging interior/exterior cells. This\nvolume fraction is also involved in the procedure to separate stresses due to\nthe external fluid from the total stresses. Fourth, we show a path forward to\nextend the VFIB method to Large Eddy Simulations involving IBs. Lastly, we\napply the VFIB in several numerical tests involving two- and three- dimensional\nstatic and moving IBs. We show greatly improved results compared to prior IB\nmethods. Further, we test several filter kernels and show that, for\nwell-resolved IBs, the choice of the kernel plays little role.", "category": "physics_flu-dyn" }, { "text": "Fate of bubble clusters rising in a quiescent liquid: We use experiments to study the evolution of bubble clusters in a swarm of\nfreely rising, deformable bubbles. A new machine learning-aided algorithm\nallows us to identify and track bubbles in clusters and measure the cluster\nlifetimes. The results indicate that contamination in the carrier liquid can\nenhance the formation of bubble clusters and prolong the cluster lifetimes. The\nmean bubble rise velocities conditioned on the bubble cluster size are also\nexplored, and we find a positive correlation between the cluster size and the\nrise speed of the bubbles in the cluster, with clustered bubbles rising up to\n$20\\%$ faster than unclustered bubbles.", "category": "physics_flu-dyn" }, { "text": "Adjoint Node-Based Shape Optimization of Free Floating Vessels: The paper is concerned with a node-based, gradient-driven, continuous adjoint\ntwo-phase flow procedure to optimize the shapes of free-floating vessels and\ndiscusses three topics. First, we aim to convey that elements of a\nCahn-Hilliard formulation should augment the frequently employed\nVolume-of-Fluid two-phase flow model to maintain dual consistency. It is seen\nthat such consistency serves as the basis for a robust primal/adjoint coupling\nin practical applications at huge Reynolds and Froude numbers. The second topic\ncovers different adjoint coupling strategies. A central aspect of the\napplication is the floating position, particularly the trim and the sinkage,\nthat interact with a variation of hydrodynamic loads induced by the shape\nupdates. Other topics addressed refer to the required level of density coupling\nand a more straightforward -- yet non-frozen -- adjoint treatment of\nturbulence. The third part discusses the computation of a descent direction\nwithin a node-based environment. We will illustrate means to deform both the\nvolume mesh and the hull shape simultaneously and at the same time obey\ntechnical constraints on the vessel's displacement and its extensions. The\nHilbert-space approach provides smooth shape updates using the established\ncoding infrastructure of a computational fluid dynamics algorithm and provides\naccess to managing additional technical constraints. Verification and\nvalidation follow from a submerged 2D cylinder case. The application includes a\nfull-scale offshore supply vessel at Re=3E+08 and Fn=0.37. Results illustrate\nthat the fully parallel procedure can automatically reduce the drag of an\nalready pre-optimized shape by 9-13 within approximately O(10.000-30.000) CPUh\ndepending on the considered couplings and floatation aspects.", "category": "physics_flu-dyn" }, { "text": "Flexible fiber reveals the two-point statistical properties of\n turbulence: We study the dynamics of a flexible fiber freely moving in a\nthree-dimensional fully-developed turbulent field and present a\nphenomenological theory to describe the interaction between the fiber\nelasticity and the turbulent flow. This theory leads to the identification of\ntwo distinct regimes of flapping, which we validate against Direct Numerical\nSimulations (DNS) fully resolving the fiber dynamics. The main result of our\nanalysis is the identification of a flapping regime where the fiber, despite\nits elasticity, is slaved to the turbulent fluctuations. In this regime the\nfiber can be used to measure two-point statistical observables of turbulence,\nincluding scaling exponents of velocity structure functions, the sign of the\nenergy cascade and the energy flux of turbulence, as well as the characteristic\ntimes of the eddies within the inertial range of scales. Our results are\nexpected to have a deep impact on the experimental turbulence research as a new\nway, accurate and efficient, to measure two-point, and more generally\nmulti-point, statistics of turbulence.", "category": "physics_flu-dyn" }, { "text": "Description of laminar-turbulent transition of an airfoil boundary layer\n measured by differential image thermography using directed percolation theory: The study presented here addresses the challenging problem of\nlaminar-turbulent flow transition in boundary layers. Directed percolation\ntheory has emerged as a promising approach to understand and describe this\ntransition in different scenarios. This study utilizes differential image\nthermography (DIT) to investigate the boundary layer transition on the suction\nside of a heated airfoil, presenting new experimental findings. First, the DIT\nresults underline the ability of capturing the near surface transition for the\nairfoil boundary layer with a high temporal and spatial resolution. Second, the\nevaluation reveals the effectiveness of directed percolation theory in\ndescribing the onset of the transition, showing agreement with all three\nuniversal exponents of (1+1)D directed percolation theory. Third, the study\nshows the applicability of this theory to a wide range of flow situations\nbeyond the parameter space covered in previous examinations. These findings\nunderscore the possible application of directed percolation models in fluid\nmechanics and suggest that the theory could serve as a high precision tool for\ndescribing the transition to turbulence.", "category": "physics_flu-dyn" }, { "text": "Numerical simulation of pulmonary airway reopening by the EOS-based\n multiphase lattice Boltzmann method: The aerosol formation is associated with the rupture of the liquid plug\nduring the pulmonary airway reopening. The fluid dynamics of this process is\ndifficult to predict because the rupture involved complex liquid-gas\ntransition. Equation of state (EOS) plays a key role in the thermodynamic\nprocess of liquid-gas transition. Here, we propose an EOS-based multiphase\nlattice Boltzmann model, in which the nonideal force is directly evaluated by\nEOSs. This multiphase model is used to model the pulmonary airway reopening and\nstudy aerosol formation during exhalation. The numerical model is first\nvalidated with the simulations of Fujioka et al.(2008). and the result is in\nreasonable agreement with their study. Furthermore, two rupture cases with and\nwithout aerosol formation are contrasted and analyzed. It is found that the\ninjury on the epithelium in the case with aerosol formation is essentially the\nsame that of without aerosol formation even while the pressure drop in airway\nincreases by about 67%. Then extensive simulations are performed to investigate\nthe effects of pressure drop, thickness of liquid plug and film on aerosol size\nand the mechanical stresses. The results show that aerosol size and the\nmechanical stresses increase as the pressure drop enlarges and thickness of\nliquid plug become thicken, while aerosol size and the mechanical stresses\ndecrease as thickness of liquid film is thicken. The present multiphase model\ncan be extended to study the generation and transmission of bioaerosols which\ncan carry the bioparticles of influenza or coronavirus.", "category": "physics_flu-dyn" }, { "text": "Stokes flows in a 2D bifurcation: The flow network model is an established approach to approximate\npressure-flow relationships in a network, which has been widely used in many\ncontexts. However, little is known about the impact of bifurcation geometry on\nsuch approximations, so the existing models mostly rely on unidirectional flow\nassumption and Poiseuille's law, and thus neglect the flow details at each\nbifurcation. In this work, we address these limitations by computing Stokes\nflows in a 2D bifurcation using LARS (Lightning-AAA Rational Stokes), a novel\nmesh-free algorithm for solving 2D Stokes flow problems utilising an applied\ncomplex analysis approach based on rational approximation of the Goursat\nfunctions. Using our 2D bifurcation model, we show that the fluxes in two child\nbranches depend on not only pressures and widths of inlet and outlet branches,\nas most previous studies have assumed, but also detailed bifurcation geometries\n(e.g. bifurcation angle), which were not considered in previous studies. The 2D\nStokes flow simulations allow us to represent the relationship between\npressures and fluxes of a bifurcation using an updated flow network, which\nconsiders the bifurcation geometry and can be easily incorporated into previous\nflow network approaches. The errors in the flow conductance of a channel in a\nbifurcation approximated using Poiseuille's law can be greater than 16%, when\nthe centreline length is twice the inlet channel width and the bifurcation\ngeometry is highly asymmetric. In addition, we present details of 2D Stokes\nflow features, such as flow separation in a bifurcation and flows around fixed\nobjects at different locations, which previous flow network models cannot\ncapture. These findings suggest the importance of incorporating detailed flow\nmodelling techniques alongside existing flow network approaches when solving\ncomplex flow problems.", "category": "physics_flu-dyn" }, { "text": "Critical balance in magnetohydrodynamic, rotating and stratified\n turbulence: towards a universal scaling conjecture: It is proposed that critical balance - a scale-by-scale balance between the\nlinear propagation and nonlinear interaction time scales - can be used as a\nuniversal scaling conjecture for determining the spectra of strong turbulence\nin anisotropic wave systems. Magnetohydrodynamic (MHD), rotating and stratified\nturbulence are considered under this assumption and, in particular, a novel and\nexperimentally testable energy cascade scenario and a set of scalings of the\nspectra are proposed for low-Rossby-number rotating turbulence. It is argued\nthat in neutral fluids, the critically balanced anisotropic cascade provides a\nnatural path from strong anisotropy at large scales to isotropic Kolmogorov\nturbulence at very small scales. It is also argued that the kperp^{-2} spectra\nseen in recent numerical simulations of low-Rossby-number rotating turbulence\nmay be analogous to the kperp^{-3/2} spectra of the numerical MHD turbulence in\nthe sense that they could be explained by assuming that fluctuations are\npolarised (aligned) approximately as inertial waves (Alfven waves for MHD).", "category": "physics_flu-dyn" }, { "text": "From thin plates to Ahmed bodies: linear and weakly non-linear stability\n of rectangular prisms: We study the stability of laminar wakes past three-dimensional rectangular\nprisms. The width-to-height ratio is set to $W/H=1.2$, while the\nlength-to-height ratio $1/62000$), these structures are no longer seen as\nbeing dominant; the coherence is clearly lost, both in the near-wall region as\nwell as in the outer layer of the boundary layer. Note, however, that\nlarge-scale streaks in the streamwise velocity, which have their peak energy at\nabout half the boundary-layer thickness, are unambiguously observed.\n In addition to visualisation with classical three-dimensional isosurfaces,\nthe video is also rendered using stereoscopic views using red-cyan anaglyphs.", "category": "physics_flu-dyn" }, { "text": "Two-dimensional flow on the sphere: Equilibrium statistical mechanics predicts that inviscid, two-dimensional,\nincompressible flow on the sphere eventually reaches a state in which spherical\nharmonic modes of degrees $n=1$ and $n=2$ hold all the energy. By a separate\ntheory, such flow is static in a reference frame rotating at angular speed\n$2\\Omega/3$ with respect to the inertial frame. The vorticity field in the\nstatic frame is an accident of the initial conditions, but, once established,\nit lasts forever under the stated assumptions. We investigate the possibility\nof such behavior with a stereographic-coordinate model that conserves energy\nand enstrophy when the viscosity vanishes.", "category": "physics_flu-dyn" }, { "text": "Numerical investigation of high-pressure combustion in rocket engines\n using Flamelet/Progress-variable models: The present paper deals with the numerical study of high pressure LOx/H2 or\nLOx/hydrocarbon combustion for propulsion systems. The present research effort\nis driven by the continued interest in achieving low cost, reliable access to\nspace and more recently, by the renewed interest in hypersonic transportation\nsystems capable of reducing time-to-destination. Moreover, combustion at high\npressure has been assumed as a key issue to achieve better propulsive\nperformance and lower environmental impact, as long as the replacement of\nhydrogen with a hydrocarbon, to reduce the costs related to ground operations\nand increase flexibility. The current work provides a model for the numerical\nsimulation of high- pressure turbulent combustion employing detailed chemistry\ndescription, embedded in a RANS equations solver with a Low Reynolds number\nk-omega turbulence model. The model used to study such a combustion phenomenon\nis an extension of the standard flamelet-progress-variable (FPV) turbulent\ncombustion model combined with a Reynolds Averaged Navier-Stokes equation\nSolver (RANS). In the FPV model, all of the thermo-chemical quantities are\nevaluated by evolving the mixture fraction Z and a progress variable C. When\nusing a turbulence model in conjunction with FPV model, a probability density\nfunction (PDF) is required to evaluate statistical averages of chemical\nquantities. The choice of such PDF must be a compromise between computational\ncosts and accuracy level. State- of-the-art FPV models are built presuming the\nfunctional shape of the joint PDF of Z and C in order to evaluate\nFavre-averages of thermodynamic quantities. The model here proposed evaluates\nthe most probable joint distribution of Z and C without any assumption on their\nbehavior.", "category": "physics_flu-dyn" }, { "text": "Rigorous scaling laws for internally heated convection at infinite\n Prandtl number: New bounds are proven on the mean vertical convective heat transport,\n$\\overline{\\langle wT \\rangle}$, for uniform internally heated (IH) convection\nin the limit of infinite Prandtl number. For fluid in a horizontally-periodic\nlayer between isothermal boundaries, we show that $\\overline{\\langle wT\n\\rangle} \\leq \\frac12 - c R^{-2}$, where $R$ is a nondimensional `flux'\nRayleigh number quantifying the strength of internal heating and $c = 216$.\nThen, $\\overline{\\langle wT \\rangle} = 0$ corresponds to vertical heat\ntransport by conduction alone, while $\\overline{\\langle wT \\rangle} > 0$\nrepresents the enhancement of vertical heat transport upwards due to convective\nmotion. If, instead, the lower boundary is a thermal insulator, then we obtain\n$\\overline{\\langle wT \\rangle} \\leq \\frac12 - c R^{-4}$, with $c\\approx\n0.0107$. This result implies that the Nusselt number $Nu$, defined as the ratio\nof the total-to-conductive heat transport, satisfies $Nu \\lesssim R^{4}$. Both\nbounds are obtained by combining the background method with a minimum principle\nfor the fluid's temperature and with Hardy--Rellich inequalities to exploit the\nlink between the vertical velocity and temperature. In both cases, power-law\ndependence on $R$ improves the previously best-known bounds, which, although\nvalid at both infinite and finite Prandtl numbers, approach the uniform bound\nexponentially with $R$.", "category": "physics_flu-dyn" }, { "text": "Newton saw the truth -- on the nature of fluid flow and viscous\n interaction: The viscous interaction of fluid is understood as the response to\ndeformation, which is proportional to the strain rate. This model has gradually\nbecome the standard since Stokes, and has become the basis of the classical\nflow theory, namely the Navier-Stokes (N-S) equations. However, it has never\nbeen accurately verified in the curved laminar flow. Here, a distinctive\nunambiguous simple experiment is designed to falsify the viscosity model of\ndeformation, and instead a new model is proposed, that is, the viscous friction\noriginates from the slip of fluid layering at molecular scale. Though Newton\ncontributed the initial idea of slip viscosity, the new model cannot be\nformulated without the help of modern differential geometry. From the new\nmodel, the analytical solution of laminar Taylor-Couette (T-C) flow between two\nconcentric cylinders can reproduce the result of the ideal experiment proposed\nby Newton as the outer cylinder being infinite, which was once considered a\nmistake of Newton. A significant difference with the solution of the N-S\nequations when the outer cylinder is relatively large can be used to\ndistinguish the viscosity models, even for the simplest case with both\ncylinders rotating with the same angular velocity. The accurate measurement\ndata by the LDA support the slip model, and the consequent flow theory\ninevitably leads to a new vision in turbulence research.", "category": "physics_flu-dyn" }, { "text": "Magnetohydrodynamic Viscous Flow Over a Shrinking Sheet With Second\n Order Slip Flow Model: In this paper, we investigate the magnetohydrodynamic viscous flow with\nsecond order slip flow model over a permeable shrinking surface. We have\nobtained the closed form of exact solution of Navier-Stokes equations by using\nsimilarity variable technique. The effects of slip, suction and magnetic\nparameter have been investigated in detail. The results show that there are two\nsolution branches, namely lower and upper solution branch. The behavior of\nvelocity and shear stress profiles for different values of slip, suction and\nmagnetic parameters has been discussed through graphs.", "category": "physics_flu-dyn" }, { "text": "Modeling compressed turbulent plasma with rapid viscosity variations: We propose two-equations models in order to capture the dynamics of a\nturbulent plasma undergoing compression and experiencing large viscosity\nvariations.\n The models account for possible relaminarization phases and rapid viscosity\nchanges through closures dependent on the turbulent Reynolds and on the\nviscosity Froude numbers. These closures are determined from a data-driven\napproach using eddy-damped quasi normal markovian simulations. The best model\nis able to mimic the various self-similar regimes identified in\n\\citet{Viciconte2018} and to recover the rapid transition limits identified by\n\\citet{Coleman1991}.", "category": "physics_flu-dyn" }, { "text": "Exact inviscid drag adjoint solution for subcritical flows: This paper describes an exact solution to the drag-based adjoint Euler\nequations in two and three dimensions that is valid for irrotational flows.", "category": "physics_flu-dyn" }, { "text": "Study of an Advection-Reaction-Diffusion equation in a compressible flow\n field: We have studied the front propagation in a one dimensional case of combustion\nby solving numerically an advection-reaction-diffusion equation. The physical\nmodel is simplified so that no coupling phenomena are considered and the\nreacting fluid is a binary mixture of gases. The compressible flow field is\ngiven analytically. We analyse the differences between popular models used in\nfundamental studies of compressible combustion and biological problems. Then,\nwe investigate the effects of compressibility on the front interface dynamics\nfor different reaction types and we characterise the conditions for which the\nreaction stops before its completion.", "category": "physics_flu-dyn" }, { "text": "Ultralow frequency acoustic resonances and its potential for mitigating\n tsunami wave formation: Bubbles display astonishing acoustical properties since they are able to\nabsorb and scatter large amounts of energy coming from waves whose wavelengths\nare two orders of magnitude larger than the bubble size. Thus, as the\ninteraction distance between bubbles is much larger than the bubble size,\nclouds of bubbles exhibit collective oscillations which can scatter acoustic\nwaves three orders magnitude larger than the bubble size. Here we propose\nbubble based systems which resonate at frequencies that match the time scale\nrelevant for seismogenic tsunami wave generation and may mitigate the\ndevastating effects of tsunami waves. Based on a linear approximation, our\nna\\\"ive proposal may open new research paths towards the mitigation of tsunami\nwaves generation.", "category": "physics_flu-dyn" }, { "text": "Contaminant transport by human passage through an air curtain separating\n two sections of a corridor: Part I -- uniform ambient temperature: Air curtains are commonly used as separation barriers to reduce exchange\nflows through an open-door of a building.Here, we investigated the\neffectiveness of an air curtain to prevent the transport of contaminants by a\nperson walking along a corridor from a dirty zone into a clean zone. We\nconducted small-scale waterbath experiments with freshwater, brine and sugar\nsolutions, with the brine as a passive tracer for the contaminant in the wake\nof person. A cylinder representing human walking was pulled between two fixed\npoints in the channel across the air curtain. We observed that the air curtain\ncan prevent up to 40% of the contaminant transport due to the wake of a moving\nperson.We proposed a new way to evaluate the performance of an air curtain in\nterms of the deflection modulus and the effectiveness defined for this\niso-density situation, similar to quantities typically used for the case where\nthe fluid densities in the two zones are different. We observed that the air\ncurtain has an optimal operating condition to achieve a maximum effectiveness.\nDye visualisations and time-resolved particle image velocimetry of the air\ncurtain and the cylinder wake were used to examine the re-establishment process\nof the planar jet after its disruption by the cylinder and we observed that\nsome part of the wake is separated by the re-establishing curtain. We observed\nthat the exchange flux peaks after the cylinder passes the air curtain and\nreduces to a typical value after the re-establishment of the curtain.", "category": "physics_flu-dyn" }, { "text": "Planetary dynamos: The theory of planetary dynamos and its applications to observed phenomena of\nplanetary magnetism are outlined. It is generally accepted that convection\nflows driven by thermal or compositional buoyancy are the most likely source\nfor the sustenance of global planetary magnetic fields. While the existence of\ndynamos in electrically conducting fluid planetary cores provides constraints\non properties of the latter, the lack of knowledge about time dependences of\nthe magnetic fields and about their toroidal components together with the\nrestricted parameter regions accessible to theory have prevented so far a full\nunderstanding of the phenomena of planetary magnetism.", "category": "physics_flu-dyn" }, { "text": "Controlled propulsion and separation of helical particles at the\n nanoscale: Controlling the motion of nano and microscale objects in a fluid environment\nis a key factor in designing optimized tiny machines that perform mechanical\ntasks such as transport of drugs or genetic material in cells, fluid mixing to\naccelerate chemical reactions, and cargo transport in microfluidic chips.\nDirected motion is made possible by the coupled translational and rotational\nmotion of asymmetric particles. A current challenge in achieving directed and\ncontrolled motion at the nanoscale lies in overcoming random Brownian motion\ndue to thermal fluctuations in the fluid. We use a hybrid lattice-Boltzmann\nMolecular Dynamics method with full hydrodynamic interactions and thermal\nfluctuations to demonstrate that controlled propulsion of individual\nnanohelices in an aqueous environment is possible. We optimize the propulsion\nvelocity and the efficiency of externally driven nanohelices. We quantify the\nimportance of the thermal effects on the directed motion by calculating the\nP\\'eclet number for various shapes, number of turns and pitch lengths of the\nhelices. Consistent with the experimental microscale separation of chiral\nobjects, our results indicate that in the presence of thermal fluctuations at\nP\\'eclet numbers $>10$, chiral particles follow the direction of propagation\naccording to its handedness and the direction of the applied torque making\nseparation of chiral particles possible at the nanoscale. Our results provide\ncriteria for the design and control of helical machines at the nanoscale.", "category": "physics_flu-dyn" }, { "text": "Growth rate distribution and intermittency in kinematic turbulent\n dynamos : which moment predicts the dynamo onset?: We consider the generation of magnetic field by a turbulent flow. For the\nlinear induction equation (i.e. the kinematic dynamo problem), we show that the\nstatistical moments of the magnetic field display multiscaling and in\nparticular moments of different order turn unstable for different values of the\ncontrol parameter. On a canonical example, we map the problem onto the\ncalculation of the injected power by a time correlated fluctuating force acting\non a Brownian particle. We are then able to calculate analytically the growth\nrate of the moments of the magnetic field and explain the origin of this\nintermittency. We finally show that the onset for the nonlinear problem is\npredicted by the linear onset of the moment of order 0 + (i.e. the logarithm of\nthe magnetic field)", "category": "physics_flu-dyn" }, { "text": "Validation of a new k-epsilon model with the pressure diffusion effects\n in separated flows: The contribution of the \"rapid\" part of the pressure diffusion to the\nturbulent kinetic energy balance is analyzed, and a new model to describe its\neffect is suggested. A new transport equation for the turbulent kinetic energy\nis derived. The new k-equation does not require any modification in the\nstandard epsilon-equation. A new k-epsilon model, which includes the new\nk-equation and the standard epsilon-equation, is validated in four separated\nflows: a planar diffuser, over a backstep, in a channel with wavy walls, and in\nan axisymmetric combustion chamber. It is shown that a new model reproduces the\nmean velocity, shear stress, and turbulent kinetic energy profiles and the skin\nfriction coefficient in very good agreement with experimental data.", "category": "physics_flu-dyn" }, { "text": "Selective energy and enstrophy modification of two-dimensional decaying\n turbulence: In two-dimensional decaying homogeneous isotropic turbulence, kinetic energy\nand enstrophy are respectively transferred to larger and smaller scales. In\nsuch spatiotemporally complex dynamics, it is challenging to identify the\nimportant flow structures that govern this behavior. We propose and numerically\nemploy two flow modification strategies that leverage the inviscid global\nconservation of energy and enstrophy to design external forcing inputs which\nchange these quantities selectively and simultaneously, and drive the system\ntowards steady-state or other late-stage behavior. One strategy employs only\nlocal flow-field information, while the other is global. We observe various\nflow structures excited by these inputs and compare with recent literature.\nEnergy modification is characterized by excitation of smaller wavenumber\nstructures in the flow than enstrophy modification.", "category": "physics_flu-dyn" }, { "text": "Investigation of Shock Wave Interactions involving Stationary and Moving\n Wedges: The present study investigates the shock wave interactions involving\nstationary and moving wedges using a sharp interface immersed boundary method\ncombined with a fifth order weighted essentially non oscillatory (WENO) scheme.\nInspired by Schardins problem, which involves moving shock interaction with a\nfinite triangular wedge, we study influences of incident shock Mach number and\ncorner angle on the resulting flow physics in both stationary and moving\nconditions. The present study involves three incident shock Mach numbers (1.3,\n1.9, 2.5) and three corner angles (60deg, 90deg, 120deg), while its impact on\nthe vorticity production is investigated using vorticity transport equation,\ncirculation, and rate of circulation production. Further, the results yield\nthat the generation of the vorticity due to the viscous effects are quite\ndominant compared to baroclinic or compressibility effects. The moving cases\npresented involve shock driven wedge problem. The fluid and wedge structure\ndynamics are coupled using the Newtonian equation. These shock driven wedge\ncases show that wedge acceleration due to the shock results in a change in\nreflected wave configuration from Single Mach Reflection (SMR) to Double Mach\nReflection (DMR). The intermediary state between them, the Transition Mach\nReflection (TMR), is also observed in the process. The effect of shock Mach\nnumber and corner angle on Triple Point (TP) trajectory, as well as on the drag\ncoefficient, is analyzed in this study.", "category": "physics_flu-dyn" }, { "text": "Sequence of bifurcations of natural convection of air in a laterally\n heated cube with perfectly insulated horizontal and spanwise boundaries: A sequence of three steady - oscillatory transitions of buoyancy convection\nof air in a laterally heated cube with perfectly thermally insulated horizontal\nand spanwise boundaries is studied. The problem is treated by Newton and\nArnoldi methods based on Krylov subspace iteration. The finite volume grid is\ngradually refined from 100^3 to 256^3 finite volumes. It is shown that the\nprimary instability is characterized by two competing eigenmodes, whose\ntemporal development results in two different oscillatory states that differ by\ntheir symmetries. Bifurcations due to both modes are subcritical. These modes\ndevelop into different oscillatory and then stochastic flow states, which, at\nlarger Grashof number, stabilize and arrive to single stable steady flow. With\nfurther increase of the Grashof number this flow loses it stability again. It\nis argued that in all the three transitions, the instabilities onsets, as well\nas reinstatement of stability, take place owing to an interaction between a\ndestabilizing centrifugal mechanism and stabilizing effect of thermal\nstratification.", "category": "physics_flu-dyn" }, { "text": "Localised eigenmodes in a moving frame of reference representing\n convective instability: When representing convective instability mechanisms with the streamwise\nBiGlobal stability approach, results suffer from a sensitivity to the\nstreamwise domain truncation length and boundary conditions. The methodology\nproposed in this paper resolves this sensitivity by considering a moving frame\nof reference. In that frame, the spectrum features discrete eigenvalues whose\ncorresponding eigenfunctions decay exponentially in both the up- and downstream\ndirections. Therefore, the truncation boundaries can be placed far enough that\nboth variations in the domain length and artificial boundary conditions have no\nimpact. The discrete nature of the spectrum enables the use of local and\nnon-local stability methods to perform an independent approximation of the\nBiGlobal eigensolutions via global mode theory. We demonstrate that retrieving\nset-up-independent solutions in the stationary frame of reference is likely\nimpossible for the considered flow.", "category": "physics_flu-dyn" }, { "text": "Structure function tensor equations in inhomogeneous turbulence: Exact budget equations for the second-order structure function tensor\n$\\langle \\delta u_i \\delta u_j \\rangle$ are used to study the two-point\nstatistics of velocity fluctuations in inhomogeneous turbulence. The\nAnisotropic Generalized Kolmogorov Equations (AGKE) describe the production,\ntransport, redistribution and dissipation of every Reynolds stress component\noccurring simultaneously among different scales and in space, i.e. along\ndirections of statistical inhomogeneity. The AGKE are effective to study the\ninter-component and multi-scale processes of turbulence. In contrast to more\nclassic approaches, such as those based on the spectral decomposition of the\nvelocity field, the AGKE provide a natural definition of scales in the\ninhomogeneous directions, and describe fluxes across such scales too. Compared\nto the Generalized Kolmogorov Equation, which is recovered as their half trace,\nthe AGKE can describe inter-component energy transfers occurring via the\npressure-strain term and contain also budget equations for the off-diagonal\ncomponents of $\\langle \\delta u_i \\delta u_j \\rangle$.\n The non-trivial physical interpretation of the AGKE terms is demonstrated\nwith three examples. First, the near-wall cycle of a turbulent channel flow at\n$Re_\\tau=200$ is considered. The off-diagonal component $\\langle -\\delta u\n\\delta v \\rangle$, which can not be interpreted in terms of scale energy, is\ndiscussed in detail. Wall-normal scales in the outer turbulence cycle are then\ndiscussed by applying the AGKE to channel flows at $Re_\\tau=500$ and $1000$. In\na third example, the AGKE are computed for a separating and reattaching flow.\nThe process of spanwise-vortex formation in the reverse boundary layer within\nthe separation bubble is discussed for the first time.", "category": "physics_flu-dyn" }, { "text": "Hysteretic behavior of electrical conductivity in packings of particles: We address the problem of predicting saturation-dependent electrical\nconductivity {\\sigma} in packings of spheres during drainage and imbibition.\nThe effective-medium approximation (EMA) and the universal power law of\npercolation for {\\sigma} are used, respectively, at higher and low water\nsaturations to predict the conductivity, with the crossover between the two\noccurring at some intermediate saturation Swx. The main input to the theory is\na single parameter that we estimate using the capillary pressure data. The\npredictions are compared with experimental, as well as numerical data for three\ndistinct types of packings. The results for drainage in all the packings\nindicate that the universal power law of percolation is valid over the entire\nrange of Sw. For imbibition, however, the universal power law crosses over to\nthe EMA at Swx = 0.5. We also find that the effect of the pore-size\ndistribution on the {\\sigma}-Sw relation is minimal during both drainage and\nimbibition.", "category": "physics_flu-dyn" }, { "text": "Numerical solution of the Newtonian plane Couette flow with linear\n dynamic wall slip: An efficient numerical approach based on weighted average finite differences\nis used to solve the Newtonian plane Couette flow with wall slip, obeying a\ndynamic slip law that generalizes the Navier slip law with the inclusion of a\nrelaxation term. Slip is exhibited only along the fixed plate, and the motion\nis triggered by the motion of the other plate. Three different cases are\nconsidered for the motion of the moving plate, i.e., constant speed,\noscillating speed, and a single-period sinusoidal speed. The velocity and the\nvolumetric flow rate are calculated in all cases and comparisons are made with\nthe results of other methods and available results in the literature. The\nnumerical outcomes confirm the damping with time and the lagging effects\narising from the Navier and dynamic wall slip conditions and demonstrate the\nhysteretic behavior of the slip velocity in following the harmonic boundary\nmotion.", "category": "physics_flu-dyn" }, { "text": "Skating on a Film of Air: Drops Impacting on a Surface: Drops impacting on a surface are ubiquitous in our everyday experience. This\nimpact is understood within a commonly accepted hydrodynamic picture: it is\ninitiated by a rapid shock and a subsequent ejection of a sheet leading to\nbeautiful splashing patterns. However, this picture ignores the essential role\nof the air that is trapped between the impacting drop and the surface. Here we\ndescribe a new imaging modality that is sensitive to the behavior right at the\nsurface. We show that a very thin film of air, only a few tens of nanometers\nthick, remains trapped between the falling drop and the surface as the drop\nspreads. The thin film of air serves to lubricate the drop enabling the fluid\nto skate on the air film laterally outward at surprisingly high velocities,\nconsistent with theoretical predictions. Eventually this thin film of air must\nbreak down as the fluid wets the surface. We suggest that this occurs in a\nspinodal-like fashion, and causes a very rapid spreading of a wetting front\noutwards; simultaneously the wetting fluid spreads inward much more slowly,\ntrapping a bubble of air within the drop. Our results show that the dynamics of\nimpacting drops are much more complex than previously thought and exhibit a\nrich array of unexpected phenomena that require rethinking classical paradigms.", "category": "physics_flu-dyn" }, { "text": "Particles on Demand method: theoretical analysis, simplification\n techniques and model extensions: The Particles on Demand method [B. Dorschner, F. B\\\"{o}sch and I. V. Karlin,\n{\\it Phys. Rev. Lett.} {\\bf 121}, 130602 (2018)] was recently formulated with a\nconservative finite volume discretization and validated against challenging\nbenchmarks. In this work, we rigorously analyze the properties of the reference\nframe transformation and its implications on the accuracy of the model. Based\non these considerations, we propose strategies to boost the efficiency of the\nscheme and to reduce the computational cost. Additionally, we generalize the\nmodel such that it includes a tunable Prandlt number via quasi-equilibrium\nrelaxation. Finally, we adapt concepts from the multi-scale semi-Lagrangian\nlattice Boltzmann formulation to the proposed framework, further improving the\npotential and the operating range of the kinetic model. Numerical simulations\nof high Mach compressible flows demonstrate excellent accuracy and stability of\nthe model over a wide range of conditions.", "category": "physics_flu-dyn" }, { "text": "A framework of the transport model for high-order eddy viscosity tensor\n in two-dimensional turbulent flow: Motivated by the concept of eddy viscosity tensor in improved Boussinesq\nhypothesis, a transport model of high-order eddy viscosity tensor in 2D-3C\nturbulence structure is derived from the second-order moment model by tensorial\nanalysis.", "category": "physics_flu-dyn" }, { "text": "Transformation hydrodynamic metamaterials: Rigorous arguments on form\n invariance and structured design with spatial variance: The method of transformation optics has been a powerful tool to manipulate\nphysical fields if governing equations are formally invariant under coordinate\ntransformations. However, regulation of hydrodynamics is still far from\nsatisfactory due to the lack of rigorous arguments on the validation of\ntransformation theory for various categories of fluids. In this paper, we\nsystematically investigate the applicability of transformation optics to fluid\nmechanics. We find that the Stokes equation and the Navier-Stokes equations,\nrespectively describing the Stokes flow and general flow, will alter their\nforms under curvilinear transformations. On the contrary, the Hele-Shaw flow\ncharacterized with shallow geometries rigidly retain the form of its governing\nequation under arbitrary transformations. Based on the derived transformation\nrules, we propose the design of multilayered structures with spatially varying\ncell depth, instead of engineering the rank-2 shear viscosity tensor, to\nrealize the required anisotropy of transformation Hele-Shaw hydrodynamic\nmetamaterials. The theoretical certify and fabrication method revealed in this\nwork may pave an avenue for precisely controlling flow distribution with the\nconcept of artificial structure design.", "category": "physics_flu-dyn" }, { "text": "Diffuse interface modelling of soluble surfactants in two-phase flow: Phase field models for two-phase flow with a surfactant soluble in possibly\nboth fluids are derived from balance equations and an energy inequality so that\nthermodynamic consistency is guaranteed. Via a formal asymptotic analysis, they\nare related to sharp interface models. Both cases of dynamic as well as\ninstantaneous adsorption are covered. Flexibility with respect to the choice of\nbulk and surface free energies allows to realise various isotherms and\nrelations of state between surface tension and surfactant. Some numerical\nsimulations display the effectiveness of the presented approach.", "category": "physics_flu-dyn" }, { "text": "Direct numerical simulations of turbulent channel flow under\n transcritical conditions: Turbulent flows under transcritical conditions are present in regenerative\ncooling systems of rocker engines and extraction processes in chemical\nengineering. The turbulent flows and the corresponding heat transfer phenomena\nin these complex processes are still not well understood experimentally and\nnumerically. The objective of this work is to investigate the turbulent flows\nunder transcritical conditions using DNS of turbulent channel flows. A fully\ncompressible solver is used in conjunction with a Peng-Robinson real-fluid\nequation of state to describe the transcritical flows. A channel flow with two\nisothermal walls is simulated with one heated and one cooled boundary layers.\nThe grid resolution adopted in this study is slightly finer than that required\nfor DNS of incompressible channel flows. The simulations are conducted using\nboth fully (FC) and quasi-conservative (QC) schemes to assess their performance\nfor transcritical wall-bounded flows. The instantaneous flows and the\nstatistics are analyzed and compared with the canonical theories. It is found\nthat results from both FC and QC schemes qualitatively agree well with\nnoticeable difference near the top heated wall, where spurious oscillations in\nvelocity can be observed. Using the DNS data, we then examine the usefulness of\nTownsend attached eddy hypothesis in the context of flows at transcritical\nconditions. It is shown that the streamwise energy spectrum exhibits the\ninverse wavenumber scaling and that the streamwise velocity structure function\nfollows a logarithmic scaling, thus providing support to the attached eddy\nmodel at transcritical conditions.", "category": "physics_flu-dyn" }, { "text": "Inertial range scaling of inhomogeneous turbulence: We investigate how inhomogeneity influences the $k^{-5/3}$ inertial range\nscaling of turbulent kinetic energy spectra (with $k$ the wavenumber). For weak\nstatistical inhomogeneity, the energy spectrum can be described as an\nequilibrium spectrum plus a perturbation. Theoretical arguments suggest that\nthis latter contribution scales as $k^{-7/3}$. This prediction is assessed\nusing direct numerical simulations of three-dimensional Kolmogorov flow.", "category": "physics_flu-dyn" }, { "text": "Viscous coalescence of droplets: a Lattice Boltzmann study: The coalescence of two resting liquid droplets in a saturated vapor phase is\ninvestigated by Lattice Boltzmann simulations in two and three dimensions. We\nfind that, in the viscous regime, the bridge radius obeys a t^{1/2}-scaling law\nin time with the characteristic time scale given by the viscous time. Our\nresults differ significantly from the predictions of existing analytical\ntheories of viscous coalescence as well as from experimental observations.\nWhile the underlying reason for these deviations is presently unknown, a simple\nscaling argument is given that describes our results well.", "category": "physics_flu-dyn" }, { "text": "A dynamic subgrid-scale modeling framework for large eddy simulation\n using approximate deconvolution: We put forth a dynamic modeling framework for sub-grid parametrization of\nlarge eddy simulation of turbulent flows based upon the use of the approximate\ndeconvolution procedure to compute the Smagorinsky constant self-adaptively\nfrom the resolved flow quantities. Our numerical assessments for solving the\nBurgers turbulence problem shows that the proposed approach could be used as a\nviable tool to address the turbulence closure problem due to its flexibility.", "category": "physics_flu-dyn" }, { "text": "Advection-dominated transport past isolated disordered sinks: stepping\n beyond homogenization: We investigate the transport of a solute past isolated sinks in a bounded\ndomain when advection is dominant over diffusion, evaluating the effectiveness\nof homogenization approximations when sinks are distributed uniformly randomly\nin space. Corrections to such approximations can be non-local, non-smooth and\nnon-Gaussian, depending on the physical parameters (a P\\'eclet number Pe,\nassumed large, and a Damk\\\"ohler number Da) and the compactness of the sinks.\nIn one spatial dimension, solute distributions develop a staircase structure\nfor large Pe, with corrections being better described with credible intervals\nthan with traditional moments. In two and three dimensions, solute\ndistributions are near-singular at each sink (and regularized by sink size),\nbut their moments can be smooth as a result of ensemble averaging over variable\nsink locations. We approximate corrections to a homogenization approximation\nusing a moment-expansion method, replacing the Green's function by its\nfree-space form, and test predictions against simulation. We show how, in two\nor three dimensions, the leading-order impact of disorder can be captured in a\nhomogenization approximation for the ensemble mean concentration through a\nmodification to Da that grows with diminishing sink size.", "category": "physics_flu-dyn" }, { "text": "Shape and dynamics of seepage erosion in a horizontal granular bed: We investigate erosion patterns observed in a horizontal granular bed\nresulting from seepage of water motivated by observation of beach rills and\nchannel growth in larger scale landforms. Our experimental apparatus consists\nof a wide rectangular box filled with glass beads with a narrow opening in one\nof the side walls from which eroded grains can exit. Quantitative data on the\nshape of the pattern and erosion dynamics are obtained with a laser-aided\ntopography technique. We show that the spatial distribution of the source of\ngroundwater can significantly impact the shape of observed patterns. An\nelongated channel is observed to grow upstream when groundwater is injected at\na boundary adjacent to a reservoir held at constant height. An amphitheater\n(semi-circular) shape is observed when uniform rainfall infiltrates the\ngranular bed to maintain a water table. Bifurcations are observed as the\nchannels grow in response to the ground water. We further find that the\nchannels grow by discrete avalanches as the height of the granular bed is\nincreased above the capillary rise, causing the deeper channels to have rougher\nfronts. The spatio-temporal distribution of avalanches increase with bed height\nwhen partial saturation of the bed leads to cohesion between grains. However,\nthe overall shape of the channels is observed to remain unaffected indicating\nthat seepage erosion is robust to perturbation of the erosion front.", "category": "physics_flu-dyn" }, { "text": "Liquid metal flow controls at liquid metal experiment: Liquid metal flow behavior under magnetic field and electric current is\ninvestigated in experiment and numerical simulations. Several aspects of the\nresulted Lorentz force action are discussed and demonstrated. The enhanced flow\nmixing induced by the non-uniform current density appears to be crucial for the\nheat transfer efficiency. Also the outflow heat flux is strongly affected by\nthe action of the \\JxB force.", "category": "physics_flu-dyn" }, { "text": "The analogue Hawking effect in rotating polygonal hydraulic jumps: Rotation of non-circular hydraulic jumps is a recent experimental observation\nthat lacks a theory based on first principles. Here we furnish a basic theory\nof this phenomenon founded on the shallow-water model of the circular hydraulic\njump. The breaking of the axial symmetry morphs the circular jump into a\npolygonal state. Variations on this state rotate the polygon in the azimuthal\ndirection. The dependence of the rotational frequency on the flow rate and on\nthe number of polygon vertices agrees with known experimental results. We also\npredict how the rotational frequency varies with viscosity. Finally, we\nestablish a correspondence between the rotating polygonal structure and the\nHawking effect in an analogue white hole. The rotational frequency of the\npolygons affords a direct estimate of the frequency of the thermal Hawking\nradiation.", "category": "physics_flu-dyn" }, { "text": "Mixing in thermally stratified nonlinear spin-up with sources and sinks: Stratified spin-up experiments in enclosed cylinders have reported the\npresence of small pockets of well-mixed fluids but quantitative measurements of\nthe mixedness of the fluid has been lacking. Previous numerical simulations\nhave not addressed these measurements. Here we present numerical simulations\nthat address how the combined effect of spin-up and thermal boundary conditions\nenhances or hinders mixing of a fluid in a cylinder. Measurements of efficiency\nof mixing are based on the variance of temperature and explained in terms of\nthe potential energy available. The numerical simulations of the Navier--Stokes\nequations for the problem with different sets of thermal boundary conditions at\nthe horizontal walls helped shed some light on the physical mechanisms of\nmixing, for which a clear explanation was lacking.", "category": "physics_flu-dyn" }, { "text": "Lattice Boltzmann modeling of two-phase flow in macroporous media with\n application to porous asphalt: Porous asphalt (PA) is an open-graded porous material with a porosity of 20%,\nallowing fast drainage of rain and improving driving and acoustic conditions.\nHowever, the high porosity leads to significant contact with water resulting in\na shorter life expectancy. To improve the durability and performance of PA, the\ndistribution of water and its residence time have to be understood which\nentails capturing diverse multiphase phenomena. For these reasons, a numerical\nstudy is performed to analyze in detail the fluid transport mechanisms at play\nin PA, towards estimating the liquid distribution inside the nanometer- to\nmillimeter-sized pore structure of PA. In this study, LBM is used for a\ndetailed analysis of multiphase flow in complex porous domains. A multiphase\nsingle component LBM method, with parallel computing, has been developed\ndifferent phase separation phenomena on surfaces and in porous media. The LBM\nis validated with Laplace law and dynamic capillary intrusion test and then the\ncapillary uptake simulations are validated with analytical solutions, varying\ncontact angles, tube shapes and sizes. Pore meniscus and corner arc menisci are\nstudied in both square and triangular tubes. In order to address the behavior\nof rain droplets on a PA surface, run-off, wetting, pinning and evaporation of\nsingle droplet are considered in terms of effects of variation of contact\nangle, surface wetting heterogeneity and structure. Finally, gravity-driven\ndrainage in PA is studied with LBM in accordance with temporal evolution of\nwater distribution by comparing with experimental data, showing good agreement.\nThis study allows a better understanding of the diverse multiphase flow\nphenomena occurring in complex porous media, namely PA, at pore scale in\nsaturated and unsaturated states, providing information towards improving the\ndurability and performance of PA.", "category": "physics_flu-dyn" }, { "text": "Numerical study on wide gap Taylor Couette flow with flow transition: This study aims to investigate the possible sources of non-axisymmetric\ndisturbances and their propagation mechanism in Taylor Couette flow (TCF) for\nwide gap problems using direct numerical simulation with a radius ratio of 0.5\nand Reynolds number (Re) ranging from 60 to 650. Here, attention is focused on\nthe viscous layer (VL) thickness in near-wall regions and its spatial\ndistribution along the axial direction to gain an insight into the origin and\npropagation of non-axisymmetric disturbances. The results show that an\naxisymmetric Taylor-vortex flow occurs when Re is between 68 and 425. Above Re\n= 425, transition from axisymmetric to non-axisymmetric flow is observed up to\nRe = 575 before the emergence of wavy-vortex flow. From the variation of VL\nthickness with Re, the VL does not experience any significant changes in the\nflow separation region of the inner wall, as well as jet impingement region of\nboth the inner and outer walls. However, a sudden increase in VL thickness in\nthe flow separation region of the outer wall reveals possible source of\nnon-axisymmetric disturbances in the flow separation region of the outer wall.\nThese disturbances develop into the periodic secondary flow as the axisymmetric\nflow transforms into non-axisymmetric flow and this leads to the emergence of\nazimuthal wave. The periodic secondary flow contributes to sudden increase in\nthe natural wavelength and rapid reduction in the strength of two\ncounter-rotating Taylor vortices. This in turn leads to a substantial reduction\nof torque in the transition flow vis-a-vis axisymmetric Taylor-vortex flow.", "category": "physics_flu-dyn" }, { "text": "Large bubble rupture sparks fast liquid jet: This Letter presents the novel experimental observation of long and narrow\njets shooting out in disconnecting large elongated bubbles. We investigate this\nphenomenon by carrying out experiments with various viscosities, surface\ntensions, densities and nozzle radii. We propose a universal scaling law for\nthe jet velocity, which unexpectedly involves the bubble height to the power\n3/2. This anomalous exponent suggests an energy focusing phenomenon. We\ndemonstrate experimentally that this focusing is purely gravity-driven and\nindependent of the pinch-off singularity.", "category": "physics_flu-dyn" }, { "text": "Augmenting a Physics-Informed Neural Network for the 2D Burgers Equation\n by Addition of Solution Data Points: We implement a Physics-Informed Neural Network (PINN) for solving the\ntwo-dimensional Burgers equations. This type of model can be trained with no\nprevious knowledge of the solution; instead, it relies on evaluating the\ngoverning equations of the system in points of the physical domain. It is also\npossible to use points with a known solution during training. In this paper, we\ncompare PINNs trained with different amounts of governing equation evaluation\npoints and known solution points. Comparing models that were trained purely\nwith known solution points to those that have also used the governing\nequations, we observe an improvement in the overall observance of the\nunderlying physics in the latter. We also investigate how changing the number\nof each type of point affects the resulting models differently. Finally, we\nargue that the addition of the governing equations during training may provide\na way to improve the overall performance of the model without relying on\nadditional data, which is especially important for situations where the number\nof known solution points is limited.", "category": "physics_flu-dyn" }, { "text": "Discrete shedding of secondary vortices along a modified Kaden spiral: When an object is accelerated in a fluid, a primary vortex is formed through\nthe roll-up of a shear layer. This primary vortex does not grow indefinitely\nand will reach a limiting size and strength. Additional vorticity beyond the\ncritical limit will end up in a trailing shear layer and accumulate into\nsecondary vortices. The secondary vortices are typically considerably smaller\nthan the primary vortex. Here, we focus on the formation, shedding, and\ntrajectory of secondary vortices generated by a rotating rectangular plate in a\nquiescent fluid using time-resolved particle image velocimetry. The Reynolds\nnumber (Re) is varied from 840 to 11150. At low Re, the shear layer is a\ncontinuous uninterrupted layer of vorticity that rolls up into a single\ncoherent primary vortex. At Re=1955, the shear layer becomes unstable. For\nRe>4000, secondary vortices are discretely released from the plate tip. First,\nwe demonstrate that the roll-up of the shear layer, the trajectory of the\nprimary vortex, and the path of secondary vortices can be predicted by a\nmodified Kaden spiral. Second, the timing of the secondary vortex shedding is\nanalysed. The time interval between the release of successive secondary\nvortices is not constant during the rotation but increases the more vortices\nhave been shed. The shedding time interval also increases with decreasing\nReynolds number. The increased time interval under both conditions is due to a\nreduced circulation feeding rate.", "category": "physics_flu-dyn" }, { "text": "A robust single-pixel particle image velocimetry based on fully\n convolutional networks with cross-correlation embedded: Particle image velocimetry (PIV) is essential in experimental fluid dynamics.\nIn the current work, we propose a new velocity field estimation paradigm, which\nachieves a synergetic combination of the deep learning method and the\ntraditional cross-correlation method. Specifically, the deep learning method is\nused to optimize and correct a coarse velocity guess to achieve a\nsuper-resolution calculation. And the cross-correlation method provides the\ninitial velocity field based on a coarse correlation with a large interrogation\nwindow. As a reference, the coarse velocity guess helps with improving the\nrobustness of the proposed algorithm. This fully convolutional network with\nembedded cross-correlation is named as CC-FCN. CC-FCN has two types of input\nlayers, one is for the particle images, and the other is for the initial\nvelocity field calculated using cross-correlation with a coarse resolution.\nFirstly, two pyramidal modules extract features of particle images and initial\nvelocity field respectively. Then the fusion module appropriately fuses these\nfeatures. Finally, CC-FCN achieves the super-resolution calculation through a\nseries of deconvolution layers to obtain the single-pixel velocity field. As\nthe supervised learning strategy is considered, synthetic data sets including\nground-truth fluid motions are generated to train the network parameters.\nSynthetic and real experimental PIV data sets are used to test the trained\nneural network in terms of accuracy, precision, spatial resolution and\nrobustness. The test results show that these attributes of CC-FCN are further\nimproved compared with those of other tested PIV algorithms. The proposed model\ncould therefore provide competitive and robust estimations for PIV experiments.", "category": "physics_flu-dyn" }, { "text": "Re-visiting the single-phase flow model for liquid steel ladle stirred\n by gas: Ladle stirring is an important step of the steelmaking process to homogenize\nthe temperature and the chemical composition of the liquid steel and to remove\ninclusions before casting. Gas is injected from the bottom of the bath to\ninduce a turbulent flow of the liquid steel. Multiphase modeling of ladle\nstirring can become computationally expensive, especially when used within\noptimal flow control problems. This paper focuses therefore on single-phase\nflow models. It aims at improving the existing models from the literature.\nSimulations in a 2d axial-symmetrical configuration, as well as, in a real 3d\nlaboratory-scale ladle, are performed. The results obtained with the present\nmodel are in a relative good agreement with experimental data and suggest that\nit can be used as an efficient model in optimal flow control problems.", "category": "physics_flu-dyn" }, { "text": "Sound induced by a simple impact oscillator: Acoustic radiation due to vibration and impact of a spring-mass-damper\noscillator whose motion is constrained by a barrier is analyzed at a field\npoint in a free field. Impact between the mass and the barrier is modeled using\na coefficient of restitution model. Non-linear behavior of the oscillator is\nobserved due to motion constraint. Steady state response is studied using a\nbifurcation diagram. For small amplitudes of oscillation, the pressure\nperturbation by a vibrating mass in a compressible fluid is modeled as an\nacoustic dipole with its center at the equilibrium position of the mass and its\naxis aligned with the motion of the oscillator. The boundary condition for the\nacoustic domain is an acoustic free-field condition. It is observed that the\nunsteady acoustic pressure resulting from the impact force is a few orders of\nmagnitude greater relative to the pressure field resulting from vibration alone\nbefore or after impact. We also analyzed the influence of coefficient of\nrestitution, damping ratio, the ration of base excitation frequency to the\nnatural frequency, and the ratio of the distance of the barrier to the base\nexcitation amplitude on the acoustic radiation. Damping ratio and coefficient\nof restituion are shown to be the most significant paramters that affect the\nacoustic radiation from the vibro-impact oscillator.", "category": "physics_flu-dyn" }, { "text": "A Universal Fractional Model of Wall-Turbulence: Modeling of wall-bounded turbulent flows is still an open problem in\nclassical physics, with only modest progress made in the last few decades\nbeyond the so-called `log law', which describes only the intermediate region in\nwall-bounded turbulence, i.e., $30-50 y^+ \\text{ to } 0.1-0.2 R^+$ (in wall\nunits) in a pipe of radius $R$. Here we propose a fundamentally new approach\nbased on fractional calculus to model the {\\em entire} mean velocity profile\nfrom the wall to the centerline of the pipe. Specifically, we represent the\nReynolds stresses with a non-local fractional derivative of {\\em variable\norder} that decays with the distance from the wall. Surprisingly, we find that\nthis variable fractional order has a universal form for all Reynolds numbers\nand for three different flow types, i.e., channel flow, Couette flow, and pipe\nflow. We first use existing data bases from direct numerical simulations (DNS)\nto learn the variable fractional order function, and subsequently we test it\nagainst other DNS data and experimental measurements, including the Princeton\nsuperpipe experiments. Taken together, our findings reveal the continuous and\ndecaying change of rate of turbulent diffusion from the wall as well as the\nstrong non-locality of turbulent interactions that intensify away from the\nwall.", "category": "physics_flu-dyn" }, { "text": "A simple mechanism for controlling vortex breakdown in a closed flow: This work is focused to study the development and control of the laminar\nvortex breakdown of a flow enclosed in a cylinder. We show that vortex\nbreakdown can be controlled by the introduction of a small fixed rod in the\naxis of the cylinder. Our method to control the onset of vortex breakdown is\nsimpler than those previously proposed, since it does not require any auxiliary\ndevice system. The effect of the fixed rods may be understood using a simple\nmodel based on the failure of the quasi-cylindrical approximation. We report\nexperimental results of the critical Reynolds number for the appearance of\nvortex breakdown for different radius of the fixed rods and different aspect\nratios of the system. Good agreement is found between the theoretical and\nexperimental results.", "category": "physics_flu-dyn" }, { "text": "Direct Numerical Simulation of Electrokinetic Instability and Transition\n to Chaotic Motion: A new type of instability - electrokinetic instability - and an unusual\ntransition to chaotic motion near a charge-selective surface was studied by\nnumerical integration of the Nernst-Planck-Poisson-Stokes system and a weakly\nnonlinear analysis near the threshold of instability. Two kinds of initial\nconditions were considered: (a) white noise initial conditions to mimic \"room\ndisturbances\" and subsequent natural evolution of the solution; (b) an\nartificial monochromatic ion distribution with a fixed wave number to simulate\nregular wave patterns. The results were studied from the viewpoint of\nhydrodynamic stability and bifurcation theory. The threshold of\nelectroconvective movement was found by the linear spectral stability theory,\nthe results of which were confirmed by numerical simulation of the entire\nsystem. The following regimes, which replace each other as the potential drop\nbetween the selective surfaces increases, were obtained: one-dimensional steady\nsolution; two-dimensional steady electroconvective vortices (stationary point\nin a proper phase space); unsteady vortices aperiodically changing their\nparameters (homoclinic contour); periodic motion (limit cycle); and chaotic\nmotion. The transition to chaotic motion did not include Hopf bifurcation.\nNumerical resolution of the thin concentration polarization layer showed\nspike-like charge profiles along the surface, which could be, depending on the\nregime, either steady or aperiodically coalescent.\n The numerical investigation confirmed the experimentally observed absence of\nregular (near-sinusoidal) oscillations for the overlimiting regimes. There is a\nqualitative agreement of the experimental and the theoretical values of the\nthreshold of instability, the dominant size of the observed coherent\nstructures, and the experimental and theoretical volt-current characteristics.", "category": "physics_flu-dyn" }, { "text": "Birth and growth of cavitation bubbles within water under tension\n confined in a simple synthetic tree: Water under tension, as can be found in several systems including tree\nvessels, is metastable. Cavitation can spontaneously occur, nucleating a\nbubble. We investigate the dynamics of spon- taneous or triggered cavitation\ninside water filled microcavities of a hydrogel. Results show that a stable\nbubble is created in only a microsecond timescale, after transient\noscillations. Then, a diffusion driven expansion leads to filling of the\ncavity. Analysis reveals that the nucleation of a bubble releases a tension of\nseveral tens of MPa, and a simple model captures the different time scales of\nthe expansion process.", "category": "physics_flu-dyn" }, { "text": "The optimal displacement of immiscible two-phase fluids in a pore\n doublet: The displacement of multiphase fluid flow in a pore doublet is a fundamental\nproblem, and is also of importance in understanding of the transport mechanisms\nof multiphase flows in the porous media. During the displacement of immiscible\ntwo-phase fluids in the pore doublet, the transport process is not only\ninfluenced by the capillary and viscous forces, but also affected by the\nchannel geometry. In this paper, we first present a mathematical model to\ndescribe the two-phase fluid displacement in the pore doublet where the effects\nof capillary force, viscous force and the geometric structure are included.\nThen we derive an analytical solution of the model for the first time, and find\nthat the displacement process is dominated by the capillary number, the\nviscosity ratio and the radius ratio. Furthermore, we define the optimal\ndisplacement that the wetting fluids in two daughter channels break through the\nbranches simultaneously (both of them have the same breakthrough time), and\nalso obtain the critical capillary number corresponding to the optimal\ndisplacement, which is related to the radius ratio of two daughter channels and\nviscosity ratio of two immiscible fluids. Finally, it is worthy noting that the\npresent analytical results on the displacement in the pore doublet can be used\nto explain and understand the phenomenon of preferential imbibition or\npreferential flow in porous media.", "category": "physics_flu-dyn" }, { "text": "Leapfrogging Kelvin waves: Two vortex rings can form a localized configuration whereby they continually\npass through one another in an alternating fashion. This phenomenon is called\nleapfrogging. Using parameters suitable for superfluid helium-4, we describe a\nrecurrence phenomenon that is similar to leapfrogging, which occurs for two\ncoaxial straight vortex filaments with the same Kelvin wave mode. For\nsmall-amplitude Kelvin waves we demonstrate that our full Biot-Savart\nsimulations closely follow predictions obtained from a simpified model that\nprovides an analytical approximation developed for nearly parallel vortices.\nOur results are also relevant to thin-cored helical vortices in classical\nfluids.", "category": "physics_flu-dyn" }, { "text": "Multi-level segment analysis: definition and application in turbulent\n systems: For many complex systems the interaction of different scales is among the\nmost interesting and challenging features. It seems not very successful to\nextract the physical properties in different scale regimes by the existing\napproaches, such as structure-function and Fourier spectrum method.\nFundamentally these methods have their respective limitations, for instance\nscale mixing, i.e. the so-called infrared and ultraviolet effects. To make\nimprovement in this regard, a new method, multi-level segment analysis (MSA)\nbased on the local extrema statistics, has been developed. Benchmark\n(fractional Brownian motion) verifications and the important case tests\n(Lagrangian and two-dimensional turbulence) show that MSA can successfully\nreveal different scaling regimes, which has been remaining quite controversial\nin turbulence research. In general the MSA method proposed here can be applied\nto different dynamic systems in which the concepts of multiscaling and\nmultifractal are relevant.", "category": "physics_flu-dyn" }, { "text": "Density reconstruction from schlieren images through Bayesian\n nonparametric models: This study proposes a radically alternate approach for extracting\nquantitative information from schlieren images. The method uses a scaled,\nderivative enhanced Gaussian process model to obtain true density estimates\nfrom two corresponding schlieren images with the knife-edge at horizontal and\nvertical orientations. We illustrate our approach on schlieren images taken\nfrom a wind tunnel sting model, a supersonic aircraft in flight, and a\nhigh-order numerical shock tube simulation.", "category": "physics_flu-dyn" }, { "text": "Branch cuts of Stokes wave on deep water. Part I: Numerical solution and\n Pad\u00e9 approximation: Complex analytical structure of Stokes wave for two-dimensional potential\nflow of the ideal incompressible fluid with free surface and infinite depth is\nanalyzed. Stokes wave is the fully nonlinear periodic gravity wave propagating\nwith the constant velocity. Simulations with the quadruple and variable\nprecisions are performed to find Stokes wave with high accuracy and study the\nStokes wave approaching its limiting form with $2\\pi/3$ radians angle on the\ncrest. A conformal map is used which maps a free fluid surface of Stokes wave\ninto the real line with fluid domain mapped into the lower complex half-plane.\nThe Stokes wave is fully characterized by the complex singularities in the\nupper complex half-plane. These singularities are addressed by rational\n(Pad\\'e) interpolation of Stokes wave in the complex plane. Convergence of\nPad\\'e approximation to the density of complex poles with the increase of the\nnumerical precision and subsequent increase of the number of approximating\npoles reveals that the only singularities of Stokes wave are branch points\nconnected by branch cuts. The converging densities are the jumps across the\nbranch cuts. There is one branch cut per horizontal spatial period $\\lambda$ of\nStokes wave. Each branch cut extends strictly vertically above the\ncorresponding crest of Stokes wave up to complex infinity. The lower end of\nbranch cut is the square-root branch point located at the distance $v_c$ from\nthe real line corresponding to the fluid surface in conformal variables. The\nlimiting Stokes wave emerges as the singularity reaches the fluid surface.\nTables of Pad\\'e approximation for Stokes waves of different heights are\nprovided. These tables allow to recover the Stokes wave with the relative\naccuracy of at least $10^{-26}$. The tables use from several poles to about\nhundred poles for highly nonlinear Stokes wave with $v_c/\\lambda\\sim 10^{-6}.$", "category": "physics_flu-dyn" }, { "text": "Discrete and mesoscopic regimes of finite-size wave turbulence: Bounding volume results in discreteness of eigenmodes in wave systems. This\nleads to a depletion or complete loss of wave resonances (three-wave,\nfour-wave, etc.), which has a strong effect on Wave Turbulence, (WT) i.e. on\nthe statistical behavior of broadband sets of weakly nonlinear waves. This\npaper describes three different regimes of WT realizable for different levels\nof the wave excitations: Discrete, mesoscopic and kinetic WT. Discrete WT\ncomprises chaotic dynamics of interacting wave \"clusters\" consisting of\ndiscrete (often finite) number of connected resonant wave triads (or quarters).\nKinetic WT refers to the infinite-box theory, described by well-known\nwave-kinetic equations. Mesoscopic WT is a regime in which either the discrete\nand the kinetic evolutions alternate, or when none of these two types is purely\nrealized. We argue that in mesoscopic systems the wave spectrum experiences a\nsandpile behavior. Importantly, the mesoscopic regime is realized for a broad\nrange of wave amplitudes which typically spans over several orders on\nmagnitude, and not just for a particular intermediate level.", "category": "physics_flu-dyn" }, { "text": "On the noise prediction for serrated leading edges: An analytical model is developed for the prediction of noise radiated by an\naerofoil with leading edge serration in a subsonic turbulent stream. The model\nmakes use of the Fourier Expansion and Schwarzschild techniques in order to\nsolve a set of coupled differential equations iteratively and express the\nfar-field sound power spectral density in terms of the statistics of incoming\nturbulent upwash velocity. The model has shown that the primary noise reduction\nmechanism is due to the destructive interference of the scattered pressure\ninduced by the leading edge serrations. It has also shown that in order to\nachieve significant sound reduction, the serration must satisfy two geometrical\ncriteria related to the serration sharpness and hydrodynamic properties of the\nturbulence. A parametric study has been carried out and it is shown that\nserrations can reduce the overall sound pressure level at most radiation\nangles, particularly at downstream angles close to the aerofoil surface. The\nsound directivity results have also shown that the use of leading edge\nserration does not particularity change the dipolar pattern of the far-field\nnoise at low frequencies, but it changes the cardioid directivity pattern\nassociated with radiation from straight-edge scattering at high frequencies to\na tilted dipolar pattern.", "category": "physics_flu-dyn" }, { "text": "Droplet actuation induced by coalescence: experimental evidences and\n phenomenological modeling: This paper considers the interaction between two droplets placed on a\nsubstrate in immediate vicinity. We show here that when the two droplets are of\ndifferent fluids and especially when one of the droplet is highly volatile, a\nwealth of fascinating phenomena can be observed. In particular, the interaction\nmay result in the actuation of the droplet system, i.e. its displacement over a\nfinite length. In order to control this displacement, we consider droplets\nconfined on a hydrophilic stripe created by plasma-treating a PDMS substrate.\nThis controlled actuation opens up unexplored opportunities in the field of\nmicrofluidics. In order to explain the observed actuation phenomenon, we\npropose a simple phenomenological model based on Newton's second law and a\nsimple balance between the driving force arising from surface energy gradients\nand the viscous resistive force. This simple model is able to reproduce\nqualitatively and quantitatively the observed droplet dynamics.", "category": "physics_flu-dyn" }, { "text": "Detailed comparative study and a mechanistic model of resuspension of\n spherical particles from rough and smooth surfaces: Resuspension of solid particles by a tornado-like vortex from surfaces of\ndifferent roughness is studied using a three-dimensional particle tracking\nvelocimetry (3D-PTV) method. By utilizing the three-dimensional information on\nparticle positions, velocities and accelerations before, during and after the\nresuspension (lift-off) event, we demonstrate that the resuspension efficiency\nis significantly higher from the rough surface, and propose a mechanistic model\nof this peculiar effect. The results indicate that for all Reynolds numbers\ntested, the resuspension rate, as well as particle velocities and\naccelerations, are higher over the rough surface, as compared to the smooth\ncounterpart. The results and the model can help to improve modeling and\nanalysis of resuspension rates in engineering and environmental applications.", "category": "physics_flu-dyn" }, { "text": "Pore cross-talk in colloidal filtration: Blockage of pores by particles is found in many processes, including\nfiltration and oil extraction. We present filtration experiments through a\nlinear array of ten channels with one dimension which is sub-micron, through\nwhich a dilute dispersion of Brownian polystyrene spheres flows under the\naction of a fixed pressure drop. The growth rate of a clog formed by particles\nat a pore entrance systematically increases with the number of already\nsaturated (entirely clogged) pores, indicating that there is an interaction or\n\"cross-talk\" between the pores. This observation is interpreted based on a\nphenomenological model, stating that a diffusive redistribution of particles\noccurs along the membrane, from clogged to free pores. This one-dimensional\nmodel could be extended to two-dimensional membranes.", "category": "physics_flu-dyn" }, { "text": "An estimate of the circulation generated by a bluff body: A loss in circulation is sometimes cited in connection with bluff-body wakes\nas a result of comparing the circulation actually observed downstream with a\nwell-known theoretical estimate of the total circulation generated by a\ncylinder. In an effort to better understand this reported loss in circulation,\nan alternative estimate of the circulation generated by a cylinder is derived\nby integrating the velocity on a closed loop containing the attached boundary\nlayer. Predictions of the dimensionless circulation for a cylinder in cross\nflow are less than the previous theoretical estimate and agree with observed\nvalues. This suggests that the total circulation generated by bluff bodies may\nhave been overestimated in the past, and that comparison of observed values\nwith this overestimate is the origin of the perceived \"loss\" in circulation.", "category": "physics_flu-dyn" }, { "text": "Adaptive and model-based control theory applied to convectively unstable\n flows: Research on active control for the delay of laminar-turbulent transition in\nboundary layers has made a significant progress in the last two decades, but\nthe employed strategies have been many and dispersed. Using one framework, we\nreview model-based techniques, such as linear-quadratic regulators, and\nmodel-free adaptive methods, such as least-mean square filters. The former are\nsupported by a elegant and powerful theoretical basis, whereas the latter may\nprovide a more practical approach in the presence of complex disturbance\nenvironments, that are difficult to model. We compare the methods with a\nparticular focus on efficiency, practicability and robustness to uncertainties.\nEach step is exemplified on the one-dimensional linearized Kuramoto-Sivashinsky\nequation, that shows many similarities with the initial linear stages of the\ntransition process of the flow over a flat plate. Also, the source code for the\nexamples are provided.", "category": "physics_flu-dyn" }, { "text": "Divergence of critical fluctuations on approaching catastrophic phase\n inversion in turbulent emulsions: Catastrophic phase inversion, the sudden breakdown of a dense emulsion,\noccurs when the dispersed majority phase irreversibly exchanges role with the\ncontinuous minority phase. This common process has been extensively studied\nover the past decades and yet its fundamental physical mechanism has remained\nlargely unexplored. Here we experimentally and numerically study the dynamics\nof catastrophic phase inversion as it occurs when the volume fraction of the\ndispersed phase exceeds a critical volume fraction (typically around 92% in\nexperiments). Our data accurately quantify the abrupt change of both the global\ntorque and average droplet size at approaching and across the phase inversion\npoint, exhibiting strong hysteresis. Most importantly, we reveal that the\nfluctuations in the global torque diverge as a power-law while approaching the\ncritical volume-fraction and we connect their growth to the formation of highly\nheterogeneous spatial droplet structures. The present finding, unveiling the\ntight connection between fluctuations in dynamic heterogeneity and the critical\ndivergence of torque fluctuation, paves the way to a quantitative description\nof catastrophic phase inversion as an out-of-equilibrium critical-like\nphenomena.", "category": "physics_flu-dyn" }, { "text": "Formation, dissolution and properties of surface nanobubbles: Surface nanobubbles are stable gaseous phases in liquids that form on solid\nsubstrates. While their existence has been confirmed, there are many open\nquestions related to their formation and dissolution processes along with their\nstructures and properties, which are difficult to investigate experimentally.\nTo address these issues, we carried out molecular dynamics simulations based on\natomistic force fields for systems comprised of water, air (N2 and O2), and a\nHighly Oriented Pyrolytic Graphite (HOPG) substrate. Our results provide\ninsights into the formation/dissolution mechanisms of nanobubbles and estimates\nfor their density, contact angle, and surface tension. We found that the\nformation of nanobubbles is driven by an initial nucleation process of air\nmolecules and the subsequent coalescence of the formed air clusters. The\nclusters form favorably on the substrate, which provides an enhanced stability\nto the clusters. In contrast, nanobubbles formed in the bulk either move\nrandomly to the substrate and spread or move to the water--air surface and pop\nimmediately. Moreover, nanobubbles consist of a condensed gaseous phase with a\nsurface tension smaller than that of an equivalent system under atmospheric\nconditions, and contact angles larger than those in the equivalent nanodroplet\ncase. We anticipate that this study will provide useful insights into the\nphysics of nanobubbles and will stimulate further research in the field by\nusing all-atom simulations.", "category": "physics_flu-dyn" }, { "text": "On-Shell Description of Unsteady Flames: The problem of non-perturbative description of unsteady premixed flames with\narbitrary gas expansion is solved in the two-dimensional case. Considering the\nflame as a surface of discontinuity with arbitrary local burning rate and gas\nvelocity jumps given on it, we show that the front dynamics can be determined\nwithout having to solve the flow equations in the bulk. On the basis of the\nThomson circulation theorem, an implicit integral representation of the gas\nvelocity downstream is constructed. It is then simplified by a successive\nstripping of the potential contributions to obtain an explicit expression for\nthe vortex component near the flame front. We prove that the unknown potential\ncomponent is left bounded and divergence-free by this procedure, and hence can\nbe eliminated using the dispersion relation for its on-shell value (i.e., the\nvalue along the flame front). The resulting system of integro-differential\nequations relates the on-shell fuel velocity and the front position. As\nlimiting cases, these equations contain all theoretical results on flame\ndynamics established so far, including the linear equation describing the\nDarrieus-Landau instability of planar flames, and the nonlinear\nSivashinsky-Clavin equation for flames with weak gas expansion.", "category": "physics_flu-dyn" }, { "text": "Effect of baffles on pressurization and thermal stratification in a LN2\n tank under micro-gravity: Researches on the impact of existing baffles on sloshing suppression of\ntwo-phase fluids in storage tanks have been widely conducted in literature.\nHowever, few studies focus on the effect of the baffles on self-pressurization\nor thermal stratification of the fluids in containers. This paper uses Volume\nof Fluid (VOF) method to simulate the thermodynamic and fluid dynamic behavior\nof liquid nitrogen in a tank with different baffle structures under\nmicrogravity environment. Groups of gravity levels, fill levels and distances,\nangles and gaps of baffles, are compared and analyzed. Up to 54\\% difference in\npressurization can be observed by optimizing the baffle structure and metrics,\nwhich is significant to achieve the highest performance of storage fluid\ncontrol in the tank.", "category": "physics_flu-dyn" }, { "text": "On pressure impulse of a laser-induced underwater shock wave: We experimentally examine a laser-induced underwater shock wave with a\nspecial attention to pressure impulse, the time integral of pressure evolution.\n%total pressure variation associated with the shock wave. Plasma formation,\nshock-wave expansion, and pressure in water are observed simultaneously using a\ncombined measurement system that obtains high-resolution nanosecond-order image\nsequences. These detailed measurements reveal a non-spherically-symmetric\ndistribution of pressure peak. In contrast, remarkably, pressure impulse is\nfound to distribute symmetrically for a wide range of experimental parameters\neven when the shock waves are emitted from an elongated region. The structure\nis determined to be a collection of multiple spherical shock waves originated\nfrom point-like plasmas in the elongated region.", "category": "physics_flu-dyn" }, { "text": "A method for Hamiltonian truncation: A four-wave example: A method for extracting finite-dimensional Hamiltonian systems from a class\nof 2+1 Hamiltonian mean field theories is presented. These theories possess\nnoncanonical Poisson brackets, which normally resist Hamiltonian truncation,\nbut a process of beatification by coordinate transformation near a reference\nstate is described in order to perturbatively overcome this difficulty. Two\nexamples of four-wave truncation of Euler's equation for scalar vortex dynamics\nare given and compared: one a direct non-Hamiltonian truncation of the\nequations of motion, the other obtained by beatifying the Poisson bracket and\nthen truncating.", "category": "physics_flu-dyn" }, { "text": "Gas-Particle Dynamics in High-Speed Flows: High-speed disperse multiphase flows are present in numerous environmental\nand engineering applications with complex interactions between turbulence,\nshock waves, and particles. Compared to its incompressible counterpart,\ncompressible two-phase flows introduce new scales of motion that challenge\nsimulations and experiments. This review focuses on gas-particle interactions\nspanning subsonic to supersonic flow conditions. An overview of existing Mach\nnumber-dependent drag laws is presented, with origins from 18th-century cannon\nfirings, and new insights from particle-resolved numerical simulations. The\nequations of motion and phenomenology for a single particle are first reviewed.\nMulti-particle systems spanning dusty gases to dense suspensions are then\ndiscussed from numerical and experimental perspectives.", "category": "physics_flu-dyn" }, { "text": "Direct numerical simulation of acoustic turbulence: Zakharov-Sagdeev\n spectrum: We present the results of direct numerical simulation of three-dimensional\nacoustic turbulence in medium with weak positive dispersion. It is shown that\nat the beginning of the long-wavelength region in the turbulence energy\ndistribution in the $k$-space, there are formed jets in the form of narrow\ncones. At larger wavenumbers, the cones broaden, and the distribution\naccordingly tends to isotropic. In this region of wavenumbers, the\nangle-averaged turbulence spectrum acquires a power-law character, $E(k)\\propto\nk^{-\\alpha}$, with the exponent close to $3/2$, which corresponds to the\nZakharov-Sagdeev weak acoustic turbulence spectrum.", "category": "physics_flu-dyn" }, { "text": "Modeling air entrainment in plunging jet using 3DYNAFS: As the liquid jet plunges into a free surface, significant air is entrained\ninto the water and forms air pockets. These air pockets eventually break up\ninto small bubbles, which travel downstream to form a bubbly wake. To better\nunderstand the underlying flow physics involved in the bubble entrainment, in\nthe linked videos, air entrainment due to a water jet plunging onto a pool of\nstationary water was numerically studied by using the 3DYNAFS software suit.\n The flow field is simulated by directly solving the Navier-Stokes equations\nthrough the viscous module, 3DYNAFS-VIS, using a level set method for capturing\nthe free surface. The breakup of entrained air pockets and the resulting bubbly\nflow were modeled by coupling 3DYNAFS-VIS with a Lagrangian multi-bubble\ntracking model, 3DYNAFS-DSM (Hsiao & Chahine, 2003), which emits bubbles into\nthe liquid according to local liquid/gas interface flow conditions based on the\nsub-grid air entrainment modeling proposed by Ma et al. (2011), and tracks all\nbubbles in the liquid flow using equations of motion and bubble dynamics.\nFurther breakup of the dispersed bubbles into smaller ones is modeled by\nincorporating the bubble breakup model developed by Martinez-Bazan et al.\n(1999). The software was parallelized by using a hybrid MPI-OpenMP scheme.\n The video describes the impinging of a water jet with a diameter of 4 cm at a\nvelocity of 4 m/s, the entrainment and breakup of the air pockets accompanied\nby strong production and interaction with vorticity structures, and the\nsubsequent bubbly flows. This video has been submitted to the Gallery of Fluid\nMotion 2011, which is an annual showcase of fluid dynamics videos.", "category": "physics_flu-dyn" }, { "text": "Generalized-alpha scheme in the PFEM for velocity-pressure and\n displacement-pressure formulations of the incompressible Navier-Stokes\n equations: Despite the increasing use of the Particle Finite Element Method (PFEM) in\nfluid flow simulation and the outstanding success of the Generalized-alpha time\nintegration method, very little discussion has been devoted to their combined\nperformance. This work aims to contribute in this regard by addressing three\nmain aspects. Firstly, it includes a detailed implementation analysis of the\nGeneralized-alpha method in PFEM. The work recognizes and compares different\nimplementation approaches from the literature, which differ mainly in the terms\nthat are alpha-interpolated (state variables or forces of momentum equation)\nand the type of treatment for the pressure in the time integration scheme.\nSecondly, the work compares the performance of the Generalized-alpha method\nagainst the Backward Euler and Newmark schemes for the solution of the\nincompressible Navier-Stokes equations. Thirdly, the study is enriched by\nconsidering not only the classical velocity-pressure formulation but also the\ndisplacement-pressure formulation that is gaining interest in the\nfluid-structure interaction field. The work is carried out using various 2D and\n3D benchmark problems such as the fluid sloshing, the solitary wave\npropagation, the flow around a cylinder, and the collapse of a cylindrical\nwater column.", "category": "physics_flu-dyn" }, { "text": "Effect of confinement on flow around a rotating elliptic cylinder in\n laminar flow regime: The flow phenomena around a rotating elliptic cylinder in a channel is\nstudied numerically. The value of the confinement parameter \\beta is varied as\n\\frac{1}{k}, where k = 2, 4, 6, and 8 respectively, to demonstrate the\nvortex-shedding patterns around the cylinder in the downstream wake. The\nnon-dimensional rotation rate \\alpha takes up 0.5, 1, and 2 as its value.\nAdditionally, the Reynolds number (\\textit{Re}) based on the cylinder diameter\nis taken to be 50, 100, and 150 respectively. A parametric study is performed\nto explain the changes in drag coefficient \\textit{(C_{D})}, lift coefficient\n\\textit{(C_{L})}, and moment coefficient \\textit{(C_{M})} with variations of\n\\beta, \\alpha, and \\textit{Re}. The Fast-Fourier transform (FFT) of the\ntime-periodic lift signals is presented to understand the shedding frequency\ncharacteristics, and the \\textit{C_{M}} values are analyzed for cases of\nautorotation. Despite the introduction of significant confinement and cylinder\nrotation, complete suppression of vortex shedding is not observed for the\nconsidered parameter space. Autorotation is observed and becomes prominent with\ndecrease in non-dimensional rotation rate and increase in confinement and\nReynolds number.", "category": "physics_flu-dyn" }, { "text": "Bingham Fluid Flow through Oscillatory Porous Plate with Ion-Slip and\n Hall Current: The numerical approach has been performed to study the Bingham fluid flow\nthrough an oscillatory porous plate with Ion-Slip and Hall current. Initially,\nat time; t = 0 both the fluid and the upper plate are at rest. At time; t > 0\nthe upper plate begins to oscillate in its own plane while the lower plate is\nstationary. The lower plate temperature is constant while the upper plate\ntemperature has oscillated. A uniform magnetic field is applied perpendicular\nto the plates. To obtain the dimensionless equations from the governing\nnon-linear partial differential equations, the usual transformations have been\nused. The explicit finite difference technique has been applied to solve the\nobtained dimensionless equations. The MATLAB R2015a has been used for numerical\nsimulation. For the accuracy of the numerical technique, the stability and\nconvergence criteria have been discussed and the system has found to be\nconverged for P_r>=0.08, Beta_i>=2, H_a<=20, K_o<=8 (k~=2) and R_e>=0.011 with\nBeta_e=0.10, E_c=0.10, Delta(Y)=0.05 and Delta(Tau)=0.0001. The steady-state\nsolution has achieved at the dimensionless time=2.00. At the steady-state time,\nthe effect of several parameters on the flow patterns, local shear stress and\nthe Nusselt number have been shown graphically.", "category": "physics_flu-dyn" }, { "text": "Wave topology brought to the coast: Since the pioneering work of Kelvin on Laplace tidal equations, a zoology of\ntrapped waves have been found in the context of coastal dynamics. Among them,\nthe one originally computed by Kelvin plays a particular role, as it is an\nunidirectional mode filling a frequency gap between different wave bands. The\nexistence of such Kelvin waves is robust to changes in the boundary shape and\nin changes of the underlying model for the coast. This suggests a topological\ninterpretation that has yet up to now remained elusive. Here we rectify the\nsituation, by taking advantage of a reformulation of the shallow water dynamics\nthat highlights an analogy with the celebrated Haldane model in condensed\nmatter physics. For any profile of bottom topography, the number of modes that\ntransit from one wave band to another in the dispersion relation is predicted\nby computing a first Chern number describing the topology of complex eigenmodes\nin a dual, simpler wave problem.", "category": "physics_flu-dyn" }, { "text": "The propagation and decay of a coastal vortex on a shelf: A coastal eddy is modelled as a barotropic vortex propagating along a coastal\nshelf. If the vortex speed matches the phase speed of any coastal trapped shelf\nwave modes, a shelf wave wake is generated leading to a flux of energy from the\nvortex into the wave field. Using a simply shelf geometry, we determine\nanalytic expressions for the wave wake and the leading order flux of wave\nenergy. By considering the balance of energy between the vortex and wave field,\nthis energy flux is then used to make analytic predictions for the evolution of\nthe vortex speed and radius under the assumption that the vortex structure\nremains self similar. These predictions are examined in the asymptotic limit of\nsmall rotation rate and shelf slope and tested against numerical simulations.\n If the vortex speed does not match the phase speed of any shelf wave, steady\nvortex solutions are expected to exist. We present a numerical approach for\nfinding these nonlinear solutions and examine the parameter dependence of their\nstructure.", "category": "physics_flu-dyn" }, { "text": "Diffusive interaction of multiple surface nanobubbles and nanodroplets:\n shrinkage, growth, and coarsening: Surface nanobubbles are nanoscopic spherical-cap shaped gaseous domains on\nimmersed substrates which are stable, even for days. After the stability of a\n{\\it single} surface nanobubble has been theoretically explained, i.e. contact\nline pinning and gas oversaturation are required to stabilize them against\ndiffusive dissolution [Lohse and Zhang, Phys.\\ Rev.\\ E 91, 031003 (R) (2015)],\nhere we focus on the {\\it collective} diffusive interaction of {\\it multiple}\nnanobubbles. For that purpose we develop a finite difference scheme for the\ndiffusion equation with the appropriate boundary conditions and with the\nimmersed boundary method used to represent the growing or shrinking bubbles.\nAfter validation of the scheme against the exact results of Epstein and Plesset\nfor a bulk bubble [J. Chem. Phys. 18, 1505 (1950)] and of Lohse and Zhang for a\nsurface bubble, the framework of these simulations is used to describe the\ncoarsening process of competitively growing nanobubbles. The coarsening process\nfor such diffusively interacting nanobubbles slows down with advancing time and\nthus increasing bubble distance. The present results for surface nanobubbles\nare also applicable for immersed surface nanodroplets, for which better\ncontrolled experimental results of the coarsening process exist.", "category": "physics_flu-dyn" }, { "text": "Symmetric form for the hyperbolic-parabolic system of fourth-gradient\n fluid model: The fourth-gradient model for fluids-associated with an extended molecular\nmean-field theory of capillarity-is considered. By producing fluctuations of\ndensity near the critical point like in computational molecular dynamics, the\nmodel is more realistic and richer than van der Waals' one and other models\nassociated with a second order expansion. The aim of the paper is to prove-with\na fourth-gradient internal energy already obtained by the mean field\ntheory-that the quasi-linear system of conservation laws can be written in an\nHermitian symmetric form implying the stability of constant solutions. The\nresult extends the symmetric hyperbolicity property of governing-equations'\nsystems when an equation of energy associated with high order deformation of a\ncontinuum medium is taken into account.", "category": "physics_flu-dyn" }, { "text": "OpenPIV-Matlab -- An open-source software for particle image\n velocimetry; test case: birds' aerodynamics: We present an open-source MATLAB package, entitled OpenPIV-Matlab, for\nanalyzing particle image velocimetry (PIV) data. We extend the PIV analysis\nwith additional tools for post-processing the PIV results including the\nestimation of aero/hydrodynamic forces from the PIV data of a wake behind an\nimmersed (bluff or streamlined) body. The paper presents a detailed description\nof the packages, covering the three main parts: generating two-dimensional two\ncomponent velocity fields from pairs of images (OpenPIV-Matlab), spatial and\ntemporal flow analysis based on the velocity fields (Spatial and Temporal\nAnalysis Toolbox), and wake flow analysis along with the force estimates\n(getWAKE Toolbox). A complete analysis with a variety of post-processing\ncapabilities is demonstrated using time-resolved PIV wake data of a freely\nflying European starling (\\emph{Sturnus vulgaris}) in a wind tunnel.", "category": "physics_flu-dyn" }, { "text": "Toroidal and poloidal energy in rotating Rayleigh-B\u00e9nard convection: We consider rotating Rayleigh-B\\'enard convection of a fluid with a Prandtl\nnumber of $Pr = 0.8$ in a cylindrical cell with an aspect ratio $\\Gamma = 1/2$.\nDirect numerical simulations were performed for the Rayleigh number range $10^5\n\\leq Ra \\leq 10^9$ and the inverse Rossby number range $0 \\leq 1/Ro \\leq 20$.\nWe propose a method to capture regime transitions based on the decomposition of\nthe velocity field into toroidal and poloidal parts. We identify four different\nregimes. First, a buoyancy dominated regime occurring as long as the toroidal\nenergy $e_{tor}$ is not affected by rotation and remains equal to that in the\nnon-rotating case, $e^0_{tor}$. Second, a rotation influenced regime, starting\nat rotation rates where $e_{tor} > e^0_{tor}$ and ending at a critical inverse\nRossby number $1/Ro_{cr}$ that is determined by the balance of the toroidal and\npoloidal energy, $e_{tor} = e_{pol}$. Third, a rotation dominated regime, where\nthe toroidal energy $e_{tor}$ is larger than both, $e_{pol}$ and $e^0_{tor}$.\nFourth, a geostrophic turbulence regime for high rotation rates where the\ntoroidal energy drops below the value of non-rotating convection.", "category": "physics_flu-dyn" }, { "text": "Analysis of Multipoint Correlations in Direct Numerical Simulation: We examine the Markov properties of the three velocity components of a\nturbulent flow generated by a DNS simulation of the flow around an airfoil\nsection. The spectral element code Nektar has been used to generate a well\nresolved flow field around an fx79w-151a airfoil profile at a Reynolds number\nof Re=5000 and an angle of attack of {\\alpha} = 12{\\deg}. Due to a homogeneous\ngeometry in the spanwise direction, a Fourier expansion has been used for the\nthird dimension of the simulation. In the wake of the profile the flow field\nshows a von Karman street like behavior with the vortices decaying in the wake\nwhich trigger a turbulent field. Time series of the 3D flow field were\nextracted from the flow at different locations to analyze the stochastic\nfeatures. In particular the existence of Markov properties in the flow have\nbeen shown for different cases in the surrounding of the airfoil. This is of\nbasic interest as it indicates that fine structures of turbulence can be\nreplaced by stochastic processes. Turbulent and Markovian scales are being\ndetermined in the turbulent field and limits of standard Gaussian Langevin\nprocesses are being determined by the reconstruction of a flow field in time\nand space.", "category": "physics_flu-dyn" }, { "text": "Vortices of Electro-osmotic Flow in Heterogeneous Porous Media: Traditional models of electrokinetic transport in porous media are based on\nhomogenized material properties, which neglect any macroscopic effects of\nmicroscopic fluctuations. This perspective is taken not only for convenience,\nbut also motivated by the expectation of irrotational electro-osmotic flow,\nproportional to the electric field, for uniformly charged surfaces (or constant\nzeta potential) in the limit of thin double layers. Here, we show that the\ninherent heterogeneity of porous media generally leads to macroscopic vortex\npatterns, which have important implications for convective transport and\nmixing. These vortical flows originate due to competition between\npressure-driven and electro-osmotic flows, and their size are characterized by\nthe correlation length of heterogeneity in permeability or surface charge. The\nappearance of vortices is controlled by a single dimensionless control\nparameter, defined as the ratio of a typical electro-osmotic velocity to the\ntotal mean velocity.", "category": "physics_flu-dyn" }, { "text": "A fast immersed boundary method for external incompressible viscous\n flows using lattice Green's functions: A new parallel, computationally efficient immersed boundary method for\nsolving three-dimensional, viscous, incompressible flows on unbounded domains\nis presented. Immersed surfaces with prescribed motions are generated using the\ninterpolation and regularization operators obtained from the discrete delta\nfunction approach of the original (Peskin's) immersed boundary method. Unlike\nPeskin's method, boundary forces are regarded as Lagrange multipliers that are\nused to satisfy the no-slip condition. The incompressible Navier-Stokes\nequations are discretized on an unbounded staggered Cartesian grid and are\nsolved in a finite number of operations using lattice Green's function\ntechniques. These techniques are used to automatically enforce the natural\nfree-space boundary conditions and to implement a novel block-wise adaptive\ngrid that significantly reduces the run-time cost of solutions by limiting\noperations to grid cells in the immediate vicinity and near-wake region of the\nimmersed surface. These techniques also enable the construction of practical\ndiscrete viscous integrating factors that are used in combination with\nspecialized half-explicit Runge-Kutta schemes to accurately and efficiently\nsolve the differential algebraic equations describing the discrete momentum\nequation, incompressibility constraint, and no-slip constraint. Linear systems\nof equations resulting from the time integration scheme are efficiently solved\nusing an approximation-free nested projection technique. The algebraic\nproperties of the discrete operators are used to reduce projection steps to\nsimple discrete elliptic problems, e.g. discrete Poisson problems, that are\ncompatible with recent parallel fast multipole methods for difference\nequations. Numerical experiments on low-aspect-ratio flat plates and spheres at\nReynolds numbers up to 3,700 are used to verify the accuracy and physical\nfidelity of the formulation.", "category": "physics_flu-dyn" }, { "text": "Rayleigh-Taylor Instability in a Compressible Fluid: Rayleigh-Taylor instability in a compressible fluid is reconsidered. The\ndensity is allowed to vary with pressure under the barotropy assumption. For\nthe case with equal speeds of sound in the two superposed fluids, in order to\ngive a non-trivial compressibility correction to the Rayleigh-Taylor growth\nrate, the compressibility correction is calculated to $O(g^2/k^2a^4)$. To this\norder, compressibility effects are found to reduce the growth rate.", "category": "physics_flu-dyn" }, { "text": "Assessment of the Effects of Azimuthal Mode Number Perturbations upon\n the Implosion Processes of Fluids in Cylinders: Fluid instabilities arise in a variety of contexts and are often unwanted\nresults of engineering imperfections. In one particular model for a magnetized\ntarget fusion reactor, a pressure wave is propagated in a cylindrical annulus\ncomprised of a dense fluid before impinging upon a plasma and imploding it.\nPart of the success of the apparatus is a function of how axially-symmetric the\nfinal pressure pulse is upon impacting the plasma. We study a simple model for\nthe implosion of the system to study how imperfections in the pressure imparted\non the outer circumference grow due to geometric focusing. Our methodology\nentails linearizing the compressible Euler equations for mass and momentum\nconservation about a cylindrically symmetric problem and analyzing the\nperturbed profiles at different mode numbers. The linearized system gives rise\nto singular shocks and through analyzing the perturbation profiles at various\ntimes, we infer that high mode numbers are dampened through the propagation. We\nalso study the Linear Klein-Gordon equation in the context of stability of\nlinear cylindrical wave formation whereby highly oscillatory, bounded behaviour\nis observed in a far field solution.", "category": "physics_flu-dyn" }, { "text": "Implementation of a fully nonlinear Hamiltonian Coupled-Mode Theory, and\n application to solitary wave problems over bathymetry: This paper deals with the implementation of a new, efficient,\nnon-perturbative, Hamiltonian coupled-mode theory (HCMT) for the fully\nnonlinear, potential flow (NLPF) model of water waves over arbitrary\nbathymetry, Papoutsellis and Athanassoulis (2017) (arXiv:1704.03276).\nApplications considered herein concern the interaction of solitary waves with\nbottom topographies and vertical walls both in two- and three-dimensional\nenvironments. The essential novelty is a new representation of the\nDirichlet-to-Neumann operator, needed to close the Hamiltonian evolution\nequations. This representation emerges from the treatment of the substrate\nkinematical problem by means of exact semi-separation of variables in the\nirregular fluid domain, established recently by Athanassoulis & Papoutsellis\n(2017) (https://doi.org/10.1098/rspa.2017.0017). The HCMT ensures an efficient\ndimensional reduction of the exact NLFP, being able to treat an arbitrary\nbathymetry as simply as the flat-bottom case. A key point for the efficient\nimplementation of HCMT is the fast and accurate evaluation of the space-time\nvarying coefficients appearing in some of its equations. All varying\ncoefficients are calculated analytically, resulting in a refined version of the\ntheory, characterized by improved accuracy at significantly reduced\ncomputational time. This improved version of HCMT is first validated against\nexisting experimental results and other computations, and subsequently applied\nto new solitary wave-bottom interaction problems. The latter include: i) the\ninvestigation of a new type of Bragg scattering effect, appearing when a\nsolitary wave propagates over a seabed with a sinusoidal patch, and ii) the\ndisintegration, focusing and reflection of a solitary wave moving over a\nthree-dimensional bathymetry consisting of parallel banks and troughs, and\nimpinging on a vertical wall.", "category": "physics_flu-dyn" }, { "text": "Influence of gravitational forces and fluid flows on a shape of surfaces\n of a viscous fluid of capillary size: The Navier-Stokes equations and boundary conditions for viscous fluids of\ncapillary size are formulated in curvilinear coordinates associated with a\ngeometry of the fluid-gas interface. As a result, the fluid dynamics of drops\nand menisci can be described taking into account an influence of gravitational\nforces and flows on the surface shape. This gives a convenient basis for\nrespective numerical studies. Estimations of the effects are presented for the\ncase of an evaporating sessile drop.", "category": "physics_flu-dyn" }, { "text": "Liouville chains: new hybrid vortex equilibria of the 2D Euler equation: A large class of new exact solutions to the steady, incompressible Euler\nequation on the plane is presented. These hybrid solutions consist of a set of\nstationary point vortices embedded in a background sea of Liouville-type\nvorticity that is exponentially related to the stream function. The input to\nthe construction is a \"pure\" point vortex equilibrium in a background\nirrotational flow. Pure point vortex equilibria also appear as a parameter $A$\nin the hybrid solutions approaches the limits $A\\to 0,\\infty$. While $A\\to 0$\nreproduces the input equilibrium, $A\\to\\infty$ produces a new pure point vortex\nequilibrium. We refer to the family of hybrid equilibria continuously\nparametrised by $A$ as a \"Liouville link\". In some cases, the emergent point\nvortex equilibrium as $A\\to\\infty$ can itself be the input for a second family\nof hybrid equilibria linking, in a limit, to yet another pure point vortex\nequilibrium. In this way, Liouville links together form a \"Liouville chain\". We\ndiscuss several examples of Liouville chains and demonstrate that they can have\na finite or an infinite number of links. We show here that the class of hybrid\nsolutions found by Crowdy (2003) and by Krishnamurthy et al. (2019) form the\nfirst two links in one such infinite chain. We also show that the stationary\npoint vortex equilibria recently studied by Krishnamurthy et al. (2020) can be\ninterpreted as the limits of a Liouville link. Our results point to a rich\ntheoretical structure underlying this class of equilibria of the 2D Euler\nequation.", "category": "physics_flu-dyn" }, { "text": "Structures and Dynamics of Lone Schur Flows with Vorticity but no Swirls: We study the dynamics and indications of the flows with all the eigenvalues\nof the velocity gradients being real, thus `lone', \\textit{i.e.}, without\nforming the complex conjugate pairs associated to the swirls. A generic\nprototype is the `lone Schur flow (LSF)' whose velocity gradient tensor is\nuniformly of Schur form but free of complex eigenvalues. A (partial)\nintegral-differential equation governing such LSF is established, and a\nsemi-analytical algorithm is accordingly designed for computation. Simulated\nevolutions of example LSFs in 2- and 3-spaces show rich dynamics and vortical\nstructures, but no obvious swirls (nor even the homoclinic loops in whatever\ndistorted forms) could be found. We discovered the flux loop scenario and the\nanisotropic analogy of the incompressible turbulence at or close to the\ncritical dimension $D_c =4/3$ decimated from 2-space.", "category": "physics_flu-dyn" }, { "text": "Weakly nonlinear analysis of pattern formation in active suspensions: We consider the Saintillan--Shelley kinetic model of active rodlike particles\nin Stokes flow (Saintillan & Shelley 2008a,b), for which the uniform, isotropic\nsuspension of pusher particles is known to be unstable in certain settings.\nThrough weakly nonlinear analysis accompanied by numerical simulations, we\ndetermine exactly how the isotropic steady state loses stability in different\nparameter regimes. We study each of the various types of bifurcations admitted\nby the system, including both subcritical and supercritical Hopf and pitchfork\nbifurcations. Elucidating this system's behavior near these bifurcations\nprovides a theoretical means of comparing this model with other physical\nsystems which transition to turbulence, and makes predictions about the nature\nof bifurcations in active suspensions that can be explored experimentally.", "category": "physics_flu-dyn" }, { "text": "Wake interactions between two side-by-side circular cylinders with\n different sizes: Flows over two side-by-side circular cylinders exhibit fascinating flow\nphysics due to complex interactions between the coupled wakes. However, their\nmutual interference effects have not been elucidated in a quantitative manner\nthus far. In this paper, we create a mismatch between the two wakes by\nintroducing a size difference in the cylinder pair, such that the effects of\none wake on the other can be distinguished. Depending on the size and gap\nratios between the two cylinders, the coupled wake exhibits distinct dynamical\nfeatures including the quasi-periodic flow, synchronized flow, and chaotic\nflow. Through advanced spectral analysis of lift coefficients and dynamic mode\ndecomposition of the flow fields, we reveal that the quasi-periodic flows are\nmainly composed of two primary frequencies associated with vortex shedding in\nthe near wakes of the two cylinders. Both wakes impose their own frequencies on\nthe other, resulting in the beating phenomenon in the lift coefficients. The\ntriad interactions between the two wakes generate the sideband frequencies,\nwhich are associated with modal structures that are mostly active in the far\nwake. The transition from quasi-periodicity to synchronization is dominated by\nthe vortex shedding behind the larger cylinder, to which the wake of the\nsmaller cylinder locks in. These results reveal new insights on the coupled\nwakes of two circular cylinders, and are pivotal for understanding more general\nwake interaction problems.", "category": "physics_flu-dyn" }, { "text": "Generalized theory for numerical instability of the Gaussian-filtered\n Navier-Stokes equations as a model system for large eddy simulation of\n turbulence: The Gaussian-filtered Navier-Stokes equations are examined theoretically and\na generalized theory of their numerical stability is proposed. Using the exact\nexpansion series of subfilter-scale stresses or integration by parts, the terms\ndescribing the interaction between the mean and fluctuation portions in a\nstatistically steady state are theoretically rewritten into a closed form in\nterms of the known filtered quantities. This process involves high-order\nderivatives with time-independent coefficients. Detailed stability analyses of\nthe closed formulas are presented for determining whether a filtered system is\nnumerically stable when finite difference schemes or others are used to solve\nit. It is shown that by the Gaussian filtering operation, second and higher\neven-order derivatives are derived that always exhibit numerical instability in\na fixed range of directions; hence, if the filter widths are unsuitably large,\nthe filtered Navier-Stokes equations can in certain cases be unconditionally\nunstable even though there is no error in modeling the subfilter-scale stress\nterms. As is proved by a simple example, the essence of the present discussion\ncan be applied to any other smooth filters; that is, such a numerical\ninstability problem can arise whenever the dependent variables are smoothed out\nby a filter.", "category": "physics_flu-dyn" }, { "text": "Puff-like instability in laminar to turbulence supercritical transition\n of round jets: We explore the laminar to turbulence transition of round jets at low Reynolds\nnumber (Re < 1000) using a novel experimental setup and linear stability theory\n(LST). The setup has a large domain and a low disturbance environment which\nincreases the critical Re to approximately 500, permitting the appearance of a\nhitherto unknown puff-like instability (PFI). The instability is identified in\nthe self-similar region of the jet through clean flow visualizations (FV) and\nfurther corroborated by particle image velocimetry (PIV) measurements. For 400<\nRe <700, the flow exhibits PFI embedded in a puff-train encapsulated by a\nlaminar flow analogous to 'puffs' in pipe-flow transition; the latter being\nsymbolic of the finite-amplitude disturbance transition scenario. The\nobservation that PFI convects close to the average local velocity with an\ninflectional velocity profile further strengthens the analogy. LST predicts\nthat PFI is effectively a superposition of the helical mode (HM) pair with\nazimuthal wavenumbers n=+-1. Hence, PFI can also appear in the\ninfinitesimal-amplitude supercritical route to transition of linearly unstable\nflows.", "category": "physics_flu-dyn" }, { "text": "Solitary water waves created by variations in bathymetry: We study the flow of water waves over bathymetry that varies periodically\nalong one direction. We derive a linearized, homogenized model and show that\nthe periodic bathymetry induces an effective dispersion, distinct from the\ndispersion inherently present in water waves. We relate this dispersion to the\nwell-known effective dispersion introduced by changes in the bathymetry in\nnon-rectangular channels. Numerical simulations using the (non-dispersive)\nshallow water equations reveal that a balance between this effective dispersion\nand nonlinearity can create solitary waves. We derive a KdV-type equation that\napproximates the behavior of these waves in the weakly-nonlinear regime. We\nshow that, depending on geometry, dispersion due to bathymetry can be much\nstronger than traditional water wave dispersion and can prevent wave breaking\nin strongly nonlinear regimes. Computational experiments using depth-averaged\nwater wave models %, as well as confirm the analysis and suggest that\nexperimental observation of these solitary waves is possible.", "category": "physics_flu-dyn" }, { "text": "Quantfying rich patterns in agglomeration of floating beads: Macroscopic spherical particles spontaneously form rich patterns on a\nstanding Faraday wave. These patterns are found to follow a very systematic\ntrend depending on the floater concentration $\\phi$: The same floaters that\naccumulate at amplitude maxima (antinodes) of the wave at low $\\phi$,\nsurprisingly move towards the nodal lines when $\\phi$ is beyond a certain\nvalue. In more detail, circular irregularly packed antinode clusters at low\n$\\phi$ give way to loosely packed filamentary structures at intermediate\n$\\phi$, and are then followed by densely packed grid-shaped node clusters at\nhigh $\\phi$. Here, we successfully characterize the morphology of these rich\npatterns using a metric analysis, i.e., the Minkowski functionals. We modify\nthe Minkowski functionals such that we are able to measure the physical\nquantities of the clusters such as area, perimeter, and aspect ratio.", "category": "physics_flu-dyn" }, { "text": "Predicting convection configurations in coupled fluid-porous systems: A ubiquitous arrangement in nature is a free-flowing fluid coupled to a\nporous medium, for example a river or lake lying above a porous bed. Depending\non the environmental conditions, thermal convection can occur and may be\nconfined to the clear fluid region, forming shallow convection cells, or it can\npenetrate into the porous medium, forming deep cells. Here, we combine three\ncomplementary approaches -- linear stability analysis, fully nonlinear\nnumerical simulations, and a coarse-grained model -- to determine the\ncircumstances that lead to each configuration. The coarse-grained model yields\nan explicit formula for the transition between deep and shallow convection in\nthe physically relevant limit of small Darcy number. Near the onset of\nconvection, all three of the approaches agree, validating the predictive\ncapability of the explicit formula. The numerical simulations extend these\nresults into the strongly nonlinear regime, revealing novel hybrid\nconfigurations in which the flow exhibits a dynamic shift from shallow to deep\nconvection. This hybrid shallow-to-deep convection begins with small, random\ninitial data, progresses through a metastable shallow state, and arrives at the\npreferred steady-state of deep convection. We construct a phase diagram that\nincorporates information from all three approaches and depicts the regions in\nparameter space that give rise to each convective state.", "category": "physics_flu-dyn" }, { "text": "Mechanisms of dynamic near-wake modulation of a utility-scale wind\n turbine: The current study uses large eddy simulations to investigate the transient\nresponse of a utility-scale wind turbine wake to dynamic changes in atmospheric\nand operational conditions, as observed in previous field-scale measurements.\nMost wind turbine wake investigations assume quasi-steady conditions, but real\nwind turbines operate in a highly stochastic atmosphere, and their operation\n(e.g., blade pitch, yaw angle) changes constantly in response. Furthermore,\ndynamic control strategies have been recently proposed to optimize wind farm\npower generation and longevity. Therefore, improved understanding of dynamic\nwake behaviors is essential. First, changes in blade pitch are investigated and\nthe wake expansion response is found to display hysteresis as a result of flow\ninertia. The timescales of the wake response to different pitch rates are\nquantified. Next, changes in wind direction with different timescales are\nexplored. Under short timescales, the wake deflection is in the opposite\ndirection of that observed under quasi-steady conditions. Finally, yaw changes\nare implemented at different rates, and the maximum inverse wake deflection and\ntimescale are quantified, showing a clear dependence on yaw rate. To gain\nfurther physical understanding of the mechanism behind the inverse wake\ndeflection, the streamwise vorticity in different parts of the wake is\nquantified. The results of this study provide guidance for the design of\nadvanced wake flow control algorithms. The lag in wake response observed for\nboth blade pitch and yaw changes shows that proposed dynamic control strategies\nmust implement turbine operational changes with a timescale on the order of the\nrotor timescale or slower.", "category": "physics_flu-dyn" }, { "text": "About Dissipative and Pseudo Port-Hamiltonian Formulations of\n Irreversible Newtonian Compressible Flows: In this paper we consider the problem of obtaining a general port-Hamiltonian\nformulation of Newtonian fluids. We propose the port-Hamiltonian models to\ndescribe the energy flux of rotational three-dimensional isentropic and\nnon-isentropic fluids, whose boundary flows and efforts can be used for control\npurposes or for power-preserving interconnection with other physical systems.\nIn case of two-dimensional flows, we include the considerations about the\noperators associated with fluid vorticity, preserving the port-Hamiltonian\nstructure of the models proposed.", "category": "physics_flu-dyn" }, { "text": "Experimental evidence for surface tension origin of the circular\n hydraulic jump: For more than a century, the consensus has been that the thin-film hydraulic\njump that can be seen in kitchen sinks is created by gravity. However, we\nrecently reported that these jumps are created by surface tension, and gravity\ndoes not play a significant role. In this paper, {we present experimental data\nfor hydraulic jump experiments conducted in a micro-gravity environment\n($\\approx 2\\%$ of Earth's gravity) (Avedisian \\& Zhao 2000; Painter et al.\n2007; Phillips et al. 2008). The existence of a hydraulic jump in micro-gravity\nunequivocally confirms that gravity is not the principal force causing the\nformation of the kitchen sink hydraulic jump.} We also present thirteen sets of\nexperimental data conducted under terrestrial gravity reported in the\nliterature for jumps in the steady-state for a range of liquids with different\nphysical parameters, flow rates and experimental conditions. There is good\nagreement with {Bhagat et al.}'s theoretical predictions. We also show that\nbeyond a critical flow rate, $Q_C^* \\propto \\gamma^2 /\\nu \\rho^2 g$, gravity\ndoes influence the hydraulic jumps. At lower flow rates, at the scale of the\nkitchen sink, surface tension is the dominating force. We discuss previously\nreported phenomenological and predictive models of hydraulic jumps and show\nthat the phenomenological model -- effectively a statement of continuity of\nradial momentum across the jump -- does not allow the mechanism of the origin\nof the jump to be identified. However, combining the phenomenological model and\n{Bhagat et al.}'s theory allows us to predict the height of the jump.", "category": "physics_flu-dyn" }, { "text": "Bursting Drops: For decades, researchers worldwide have investigated phenomena related to\nnatural, artificial oil leakages such as oil drop formation within water\nbodies, their rise, and oil slick evolution after they breach the water-air\ninterface. Despite this, the event leading to slick formation -the bursting of\noil drops at the liquid-air interface has remained unnoticed thus far. In this\nwork, we investigate this and report a counterintuitive jetting reversal that\nreleases a daughter oil droplet inside the bulk as opposed to the upwards\nshooting jets observed in bursting air bubbles. We show that the daughter\ndroplet size thus produced can be correlated to the bulk liquid properties and\nthat its formation can be suppressed by increasing the bulk viscosity, by an\noverlaying layer of oil or by the addition of microparticles. We further\ndemonstrate the significance of our results by synthesizing colloidal pickered\ndroplets and show applications of bursting compound drops in double emulsions\nand studies on raindrop impact on a slick. These results could be immensely\ntransformative for diverse areas, including climatology, oceanic, atmospheric\nsciences, colloidal synthesis and drug delivery.", "category": "physics_flu-dyn" }, { "text": "Visco-potential free-surface flows and long wave modelling: In a recent study [DutykhDias2007] we presented a novel visco-potential free\nsurface flows formulation. The governing equations contain local and nonlocal\ndissipative terms. From physical point of view, local dissipation terms come\nfrom molecular viscosity but in practical computations, rather eddy viscosity\nshould be used. On the other hand, nonlocal dissipative term represents a\ncorrection due to the presence of a bottom boundary layer. Using the standard\nprocedure of Boussinesq equations derivation, we come to nonlocal long wave\nequations. In this article we analyse dispersion relation properties of\nproposed models. The effect of nonlocal term on solitary and linear progressive\nwaves attenuation is investigated. Finally, we present some computations with\nviscous Boussinesq equations solved by a Fourier type spectral method.", "category": "physics_flu-dyn" }, { "text": "An Eigenvalue-Free Implementation of the Log-Conformation Formulation: The log-conformation formulation, although highly successful, was from the\nbeginning formulated as a partial differential equation that contains an, for\nPDEs unusual, eigenvalue decomposition of the unknown field. To this day, most\nnumerical implementations have been based on this or a similar eigenvalue\ndecomposition, with Knechtges et al. (2014) being the only notable exception\nfor two-dimensional flows.\n In this paper, we present an eigenvalue-free algorithm to compute the\nconstitutive equation of the log-conformation formulation that works for two-\nand three-dimensional flows. Therefore, we first prove that the challenging\nterms in the constitutive equations are representable as a matrix function of a\nslightly modified matrix of the log-conformation field. We give a proof of\nequivalence of this term to the more common log-conformation formulations.\nBased on this formulation, we develop an eigenvalue-free algorithm to evaluate\nthis matrix function. The resulting full formulation is first discretized using\na finite volume method, and then tested on the confined cylinder and\nsedimenting sphere benchmarks.", "category": "physics_flu-dyn" }, { "text": "Turbulence decay towards the linearly-stable regime of Taylor-Couette\n flow: Taylor-Couette (TC) flow is used to probe the hydrodynamical stability of\nastrophysical accretion disks. Experimental data on the subcritical stability\nof TC are in conflict about the existence of turbulence (cf. Ji et al. Nature,\n444, 343-346 (2006) and Paoletti et al., A$\\&$A, 547, A64 (2012)), with\ndiscrepancies attributed to end-plate effects. In this paper we numerically\nsimulate TC flow with axially periodic boundary conditions to explore the\nexistence of sub-critical transitions to turbulence when no end-plates are\npresent. We start the simulations with a fully turbulent state in the unstable\nregime and enter the linearly stable regime by suddenly starting a\n(stabilizing) outer cylinder rotation. The shear Reynolds number of the\nturbulent initial state is up to $Re_s \\sim10^5$ and the radius ratio is\n$\\eta=0.714$. The stabilization causes the system to behave as a damped\noscillator and correspondingly the turbulence decays. The evolution of the\ntorque and turbulent kinetic energy is analysed and the periodicity and damping\nof the oscillations are quantified and explained as a function of shear\nReynolds number. Though the initially turbulent flow state decays,\nsurprisingly, the system is found to absorb energy during this decay.", "category": "physics_flu-dyn" }, { "text": "Observed dependence of characteristics of liquid-pool fires on swirl\n magnitude: One dozen vertically oriented thin rectangular vanes, 62 cm tall and 15.2 cm\nwide, were placed 27 cm from the center of heptane and ethanol pool fires in\ncontinuously fed, floor-flush pans 3.2 cm and 5.1 cm in diameter in the\nlaboratory. The vanes were all oriented at the same fixed angles from the\nradial direction, for 9 different angles, ranging from 0 to 85 degrees, thereby\nimparting 9 different levels of circulation to the air entrained by each pool\nfire. The different swirl levels were observed to engender dramatically\ndifferent pool-fire structures. Moderate swirl suppresses the global puffing\ninstability, replacing it by a global helical instability that generates a tall\nfire whirl, the height of which increases with increasing circulation. Except\nfor the largest heptane pool, higher swirl levels produced vortex breakdown,\nresulting in the emergence of a bubble-like recirculation region with a ring\nvortex encircling the axis. Measured burning rates increase with increasing\nswirl levels as a consequence of the associated increasing inflow velocities\nreducing the thickness of the boundary layer within which combustion occurs\nright above the liquid surface, eventually forming detached edge flames in the\nboundary layer that move closer to the axis as the circulation is increased.\nStill higher circulation reduces the burning rate by decreasing the surface\narea of the liquid covered by the flame, thereby reducing the height of the\nfire whirl. Even higher circulation causes edge-flame detachment, resulting in\nformation of the blue whirl identified in recent literature, often meandering\nover the surface of the liquid in the present experiments. This sequence of\nevents is documented herein.", "category": "physics_flu-dyn" }, { "text": "Three-dimensional receptivity of hypersonic sharp and blunt cones to\n free-stream planar waves using hierarchical input-output analysis: Understanding the receptivity of hypersonic flows to free-stream disturbances\nis crucial for predicting laminar to turbulent boundary layer transition.\nInput-output analysis as a receptivity tool considers which free-stream\ndisturbances lead to the largest response from the boundary layer using the\nglobal linear dynamics. Two technical challenges are addressed. First, we\nextend recent work by Kamal et al. (Kamal, 2023) and restrict the allowable\nforcing to physically realizable inputs via a free-stream boundary modification\nto the classic input-output formulation. Second, we develop a hierarchical\ninput-output (H-IO) analysis which allows us to solve the three-dimensional\nproblem at a fraction of the computational cost otherwise associated with\ndirectly inverting the fully three-dimensional resolvent operator. Next, we\nconsider Mach 5.8 flows over a sharp cone and two blunt cones with 3.6 mm and\n7.2 mm spherically blunt tips. H-IO correctly predicts that the sharp cone\nboundary layer is most receptive to slow acoustic waves at an optimal incidence\nangle of 10 degrees, validating the method. We then investigate the effect of\nfree-stream disturbances on the blunt cone boundary layer, and identify two\ndistinct vorticity-dominated receptivity mechanisms for the oblique first mode\ninstability at 10 kHz and an entropy layer instability at 40 and 70 kHz. Our\nresults reveal these receptivity processes to be highly three-dimensional in\nnature, involving both the nose-tip and excitation along narrow bands at\ncertain azimuthal angles along the oblique shock downstream. We interpret these\nprocesses in terms of critical angles from linear shock/perturbation\ninteraction theory. Finally, we show how these novel receptivity processes vary\nwith frequency and nose tip bluntness, and demonstrate how this methodology\nmight be applied to transition prediction from first principles.", "category": "physics_flu-dyn" }, { "text": "Nonaxisymmetric instabilities in turbulent electromagnetic pumps: The stability of annular electromagnetic pumps (EMPs) is investigated through\n3D direct numerical simulations. When induction effects dominate dissipative\nprocesses, linear induction EMPs become unstable and lead to non-axisymmetric\nstates relatively different from what was predicted by previous theoretical\nmodels. We show that the 3D destabilization of the flow is deeply connected to\nthe axisymmetric state, and always appears as a secondary bifurcation from a\nlocally stalled flow, even if the applied magnetic field is axisymmetric.\nFinally, we model a configuration aiming to optimize the efficiency of the\npump, by imposing electrical currents on both sides (inner and outer cylinders)\nof the pump. This configuration increases the efficiency of the pump, but\ngenerates a complex dynamics, associated to periodic oscillation of a large\nvortex flow localized in the active region of the pump. Finally, we demonstrate\nthe existence of an upper bound on the efficiency of such electromagnetic pump,\nwhich can never exceed 50%.", "category": "physics_flu-dyn" }, { "text": "Exploration in Booster Reuse Employing Aerosurfaces: A new idea for reusing rockets, based on the Energia 2 launch vehicle, using\ndeployable wings is proposed. A mission design is planned with an expected\nmaximum altitude of 500 m. Following that systems engineering was used to\nmanage the design process of separate components, which in turn led to the\nmanufacturing plans for the separate components. The sizing studies were used\nto determine how big each component would need to be, and a weight and balances\nsheet was used to determine the position of the wings for optimal static margin\nand stability percentage. A derivation was made using basic Newtonian mechanics\nfor determining the maximum mass of the rocket given a target altitude.\nFollowing that Computer Assisted Design (CAD) drawings were created of critical\ncomponents before outlining the necessary structural tests and the flight test\nplan. A MatLab script to determine static stability was used which confirmed\nthe stability of the booster at angle of attacks from 0 to 5 degrees. A MatLab\nscript to determine glide ratio and range was used on the booster and Energia 2\nto show scalability, and which was found feasible.", "category": "physics_flu-dyn" }, { "text": "Planar bubble plumes from an array of nozzles: Measurements and\n modelling: Bubble curtains are widely used for sound mitigation during offshore pile\ndriving to protect marine life. However, the lack of well validated\nhydrodynamic models is a major factor in the inability to predict the sound\nattenuation of a bubble curtain a priori. We present a new dataset resulting\nfrom bubble curtain measurements carried out in a 10m deep and 31m wide\nfreshwater tank. The data describe the evolution of the void fraction profile\nand the bubble size distribution along the height of the bubble curtain. On\nthis basis, a new relationship is developed for the dependence of the\nentrainment parameter of the bubble curtain on the air flowrate. In addition,\nwe have extended a recently developed integral model for round bubble plumes to\nseamlessly capture the transition from initially individual round plumes to a\nplanar plume after their merger. With additional modifications to the\nentrainment relation, the effective slip velocity and the initial condition for\nthe bubble size distribution, the new model is found to be in good agreement\nwith the data. In particular, the bubble size distribution sufficiently distant\nfrom the source is found to be independent of the gas flowrate, both in the\ndata and in the model.", "category": "physics_flu-dyn" }, { "text": "Characterization of the reactive flow field dynamics in a gas turbine\n injector using high frequency PIV: The present work details the analysis of the aerodynamics of an experimental\nswirl stabilized burner representative of gas turbine combustors. This analysis\nis carried out using High Frequency PIV (HFPIV) measurements in a reactive\nsituation. While this information is usually available at a rather low rate,\ntemporally resolved PIV measurements are necessary to better understand highly\nturbulent swirled flows, which are unsteady by nature. Thanks to recent\ntechnical improvements, a PIV system working at 12 kHz has been developed to\nstudy this experimental combustor flow field. Statistical quantities of the\nburner are first obtained and analyzed, and the measurement quality is checked,\nthen a temporal analysis of the velocity field is carried out, indicating that\nlarge coherent structures periodically appear in the combustion chamber. The\nfrequency of these structures is very close to the quarter wave mode of the\nchamber, giving a possible explanation for combustion instability coupling.", "category": "physics_flu-dyn" }, { "text": "Set-valued solutions for non-ideal detonation: The existence and structure of steady gaseous detonation propagating in a\npacked bed of solid inert particles are analyzed in the one-dimensional\napproximation by taking into consideration frictional and heat losses between\nthe gas and the particles. A new formulation of the governing equations is\nintroduced that eliminates the well-known difficulties with numerical\nintegration across the sonic singularity in the reactive Euler equations. The\nnew algorithm allows us to determine that the detonation solutions as the loss\nfactors are varied have a set-valued nature at low detonation velocities when\nthe sonic constraint disappears from the solutions. These set-valued solutions\ncorrespond to a continuous spectrum of the eigenvalue problem that determines\nthe velocity of the detonation.", "category": "physics_flu-dyn" }, { "text": "Temperature-gradient-induced electrokinetic flow and thermoelectricity\n of electrolyte solutions in a capillaries: A systematic theoretical study of temperature-gradient-induced electrokinetic\nflow and thermoelectric potential of electrolyte solutions in a\nmicro-/nanocapillary is presented. The study is based on a semi-analytical\nmodel developed by simultaneously solving the energy equation and the\nPoisson-Nernst-Planck/Navier-Stokes equations with the lubrication theory. The\nsemi-analytical model is shown to be mainly governed by eight parameters,\nincluding two temperature-related parameters (temperature and its gradient),\ntwo electrokinetic parameters ($\\zeta$ potential and the ratio of capillary\nradius to the Debye length $\\kappa_0a$) and four physical properties of cation\nand anion (i.e. Soret coefficient difference $\\Delta S_T$, average Soret\ncoefficient $S_T$, normalized difference in diffusivities $\\chi$ and intrinsic\nPeclet number $\\lambda$). It is found that the thermoelectric field is induced\nby three effects, which are respectively due to (1) the difference in the Soret\ncoefficients of cation and anion; (2) the selective ion diffusion resulting\nfrom the temperature-modified Boltzmann distribution of ions; (3) the advective\ntransport of ions caused by the fluid flow. The first thermoelectric effect\nprevails for lower $\\zeta$ potentials or large $\\kappa_0a$, while the second is\ndominant for higher $\\zeta$ potentials with very small $\\kappa_0a$. The first\ntwo thermoelectric effects can cooperate or counteract depending on the sign of\n$\\zeta\\Delta S_T$. Finally, the temperature-gradient-induced electrokinetic\nflow is found to be a superposition of an electroosmotic flow component due to\nthe thermoelectric field and a thermoosmotic flow component due to the combined\neffects of osmotic pressure and dielectric body force. These two flow\ncomponents may cooperate or counteract depending on values of $\\zeta$ and\n$\\kappa_0a$.", "category": "physics_flu-dyn" }, { "text": "Direct numerical simulation of the very large anisotropic scales in a\n turbulent channel: Over the last decades the knowledge on the small scales of turbulent wall\nflows has experienced a significant advance, especially in the near-wall region\nwhere the highest production of turbulent energy and the maximum turbulence\nintensity occur. The development of computers has played an important role in\nthis progress, making direct numerical simulations affordable (Kim, Moin &\nMoser, 1987), and offering wider observational possibilities than most\nlaboratory experiments. The large scales have received less attention, and it\nhas not been until recently that their significance and their real size have\nbeen widely recognized, thanks in part to the experiments by Hites (1997) and\nKim & Adrian (1999), and to the compilation of experimental and numerical data\nby Jimenez (1998). The requirements of both a very large box and a high\nReynolds number has made direct numerical simulation of the VLAS unapproachable\nuntil today. The purpose of this report is to serve as a preliminary\ndescription of a newly compiled numerical database of the characteristics of\nthe large scales in turbulent channel flow at moderate Reynolds numbers.", "category": "physics_flu-dyn" }, { "text": "Circulation Statistics in Homogeneous and Isotropic Turbulence: This is the final version of a Thesis presented to the PostGrad Program in\nPhysics of the Physics Institute of the Federal University of Rio de Janeiro\n(UFRJ), as a necessary requirement for the title of Ph.D. in Science (Physics).\nThe development of the Vortex Gas Model (VGM) introduces a novel statistical\nframework for describing the characteristics of velocity circulation. In this\nmodel, the underlying foundations rely on the statistical attributes of two\nfundamental constituents. The first is a GMC field that governs intermittent\nbehavior and the second constituent is a Gaussian Free field responsible for\nthe partial polarization of the vortices in the gas. The model is revisited in\na more sophisticated language, where volume exclusion among vortices is\naddressed. These additions were subsequently validated through numerical\nsimulations of turbulent Navier-Stokes equations. This revised approach\nharmonizes with the multifractal characteristics exhibited by circulation\nstatistics, offering a compelling elucidation for the phenomenon of\nlinearization of the statistical circulation moments, observed in recent\nnumerical simulation.\n In the end, a field theoretical approach, known as\nMartin-Siggia-Rose-Janssen-de Dominicis (MSRJD) functional method is carried\nout in the context of circulation probability density function. This approach\ndelves into the realm of extreme circulation events, often referred to as\nInstantons, through two distinct methodologies: The First investigates the\nlinear solutions and, by a renormalization group argument a time-rescaling\nsymmetry is discussed. Secondly, a numerical strategy is implemented to tackle\nthe nonlinear instanton equations in the axisymmetric approximation. This\napproach addresses the typical topology exhibited by the velocity field\nassociated with extreme circulation events.", "category": "physics_flu-dyn" }, { "text": "The Generation of Gravity-Capillary Solitary Waves by a Pressure Source\n Moving at a Trans-critical Speed: The unsteady response of a water free surface to a localized pressure source\nmoving at constant speed $U$ in the range $0.95c_\\mathrm{min} \\lesssim U \\leq\n1.02 c_\\mathrm{min}$, where $c_\\mathrm{min}$ is the minimum phase speed of\nlinear gravity-capillary waves in deep water, is investigated through\nexperiments and numerical simulations. This unsteady response state, which\nconsists of a V-shaped pattern behind the source and features periodic shedding\nof pairs of depressions from the tips of the V, was first observed\nqualitatively by Diorio et al. (Phys. Rev. Let., 103, 214502, 2009) and called\nstate III. In the present investigation, cinematic shadowgraph and\nrefraction-based techniques are utilized to measure the temporal evolution of\nthe free surface deformation pattern downstream of the source as it moves along\na towing tank, while numerical simulations of the model equation described by\nCho et al. (J. Fluid Mech., 672, 288-306, 2011) are used to extend the\nexperimental results over longer times than are possible in the experiments.\nFrom the experiments, it is found that the speed-amplitude characteristics and\nthe shape of the depressions are nearly the same as those of the freely\npropagating gravity-capillary lumps of inviscid potential theory. The decay\nrate of the depressions is measured from their height-time characteristics,\nwhich are well fitted by an exponential decay law with an order 1 decay\nconstant. It is found that the shedding period of the depression pairs\ndecreases with increasing source strength and speed. As the source speed\napproaches $c_\\mathrm{min}$, this period tends to about 1~s for all source\nmagnitudes. At the low-speed boundary of state III, a new response with\nunsteady asymmetric shedding of depressions is found. This response is also\npredicted by the model equation.", "category": "physics_flu-dyn" }, { "text": "Extinction and re-initiation of methane detonation in dilute coal\n particle suspensions: In this study, methane detonation propagation in dilute coal particle\nsuspensions is studied based on Eulerian-Lagrangian method. Two-dimensional\nconfiguration is considered, and a skeletal chemical mechanism (24 species and\n104 reactions) is applied for methane combustion. The gas and particulate phase\nequations are solved using an OpenFOAM code for two-phase compressible reacting\nflow, RYrhoCentralFOAM. The effects of char combustion on methane detonation\ndynamics are investigated and devolatized coal particles are modelled. The\nresults show that propagation of the methane detonation wave in coal particle\nsuspensions are considerably affected by coal particle concentration and size.\nDetonation extinction occurs when the coal particle size is small and\nconcentration is high. The averaged lead shock speed generally decreases with\nincreased particle concentration and decreased particle size. Mean structure of\nmethane and coal particle hybrid detonation is analysed, based on the gas and\nparticle quantities. It is found that char combustion proceeds in the subsonic\nregion behind the detonation wave and heat release is relatively distributed\ncompared to that from gas phase reaction. Moreover, for 1 {\\mu}m particle, if\nthe particle concentration is beyond a threshold value, detonation\nre-initiation occurs after it is quenched at the beginning of the coal dust\nsuspensions. This is caused by hot spots from the shock focusing along the\nreaction front in a decoupled detonation and these shocks are generated from\nchar combustion behind the lead shock. A regime map of detonation propagation\nand extinction is predicted. It is found that the re-initiation location\ndecreases with the particle concentration and approaches a constant value when\nthe concentration exceeds 1000 g/m3. The results from this study are useful for\nprevention and suppression of methane/coal dust hybrid explosion.", "category": "physics_flu-dyn" }, { "text": "Reinforcement Learning for Active Flow Control in Experiments: We demonstrate experimentally the feasibility of applying reinforcement\nlearning (RL) in flow control problems by automatically discovering active\ncontrol strategies without any prior knowledge of the flow physics. We consider\nthe turbulent flow past a circular cylinder with the aim of reducing the\ncylinder drag force or maximizing the power gain efficiency by properly\nselecting the rotational speed of two small diameter cylinders, parallel to and\nlocated downstream of the larger cylinder. Given properly designed rewards and\nnoise reduction techniques, after tens of towing experiments, the RL agent\ncould discover the optimal control strategy, comparable to the optimal static\ncontrol. While RL has been found to be effective in recent computer flow\nsimulation studies, this is the first time that its effectiveness is\ndemonstrated experimentally, paving the way for exploring new optimal active\nflow control strategies in complex fluid mechanics applications.", "category": "physics_flu-dyn" }, { "text": "Transport-Induced-Charge Electroosmosis: We report theoretical analysis of transport-induced-charge electroosmosis\n(TICEO) in a nanopore due to the presence of a local electric field and a\nconductivity gradient. TICEO shares a similar characteristic with classical\ninduced-charge electroosmosis (ICEO) that the mean velocity\n$\\overline{v_\\text{TIC}}$ is proportional to the square of the applied electric\npotential difference $\\Delta\\psi$, $i.e.$, $\\overline{v_\\text{TIC}} \\propto\n(\\Delta\\psi)^2$, appropriate for alternating current (AC) pumping applications.\nIn contrast to ICEO which is primarily used in microfluidics, TICEO does not\nrequire metallic or dielectric patches and is thus suitable for nanopore\npumping, providing new opportunities for AC nanopore applications.", "category": "physics_flu-dyn" }, { "text": "Outer streaming within a two-dimensional channel: Acoustic streaming is the net time-averaged flow that results from the\nnonlinearities in an oscillating flow. Extensive research has sought to\nidentify different physical mechanisms and types of acoustic streaming in\nsystems of various geometries. While streaming in a channel maintains one of\nthe simplest geometries, dimensional analysis of the governing equations\nreveals that multiple regimes of streaming may occur within a channel. In this\nstudy, a framework is developed for investigating and understanding the\nphysical streaming regimes in a two-dimensional channel. By taking different\nlimits of the dimensionless number ratios found within the framework, streaming\nmodels derived in previous works are recovered to demonstrate the different\nstreaming regimes within a channel. The onset of fast streaming is then\nanalyzed with the framework and nonlinear Reynolds numbers, which indicate\nwhether the streaming is slow or fast, are found for the different physical\nstreaming regimes. As a result, the framework provides a base for analyzing\nfast streaming in a channel and streaming in multi-scale systems while\norganizing previous streaming models into a physical spectrum for a channel\ngeometry.", "category": "physics_flu-dyn" }, { "text": "Bifurcations in a Quasi-Two-Dimensional Kolmogorov-Like Flow: We present a combined experimental and theoretical study of the primary and\nsecondary instabilities in a Kolmogorov-like flow. The experiment uses\nelectromagnetic forcing with an approximately sinusoidal spatial profile to\ndrive a quasi-two-dimensional (Q2D) shear flow in a thin layer of electrolyte\nsuspended on a thin lubricating layer of a dielectric fluid. Theoretical\nanalysis is based on a 2D model (Suri ${\\it et al.}$ 2014), derived from first\nprinciples by depth-averaging the full three-dimensional Navier-Stokes\nequations. As the strength of the forcing is increased, the Q2D flow in the\nexperiment undergoes a series of bifurcations, which is compared with results\nfrom direct numerical simulations of the 2D model. The effects of confinement\nand the forcing profile are studied by performing simulations that assume\nspatial periodicity and strictly sinusoidal forcing, as well as simulations\nwith realistic no-slip boundary conditions and an experimentally validated\nforcing profile. We find that only the simulation subject to physical no-slip\nboundary conditions and a realistic forcing profile provides close,\nquantitative agreement with the experiment. Our analysis offers additional\nvalidation of the 2D model as well as a demonstration of the importance of\nproperly modelling the forcing and boundary conditions.", "category": "physics_flu-dyn" }, { "text": "Controlling the stability transfer between oppositely traveling waves\n and standing waves by inversion-symmetry-breaking perturbations: The effect of an externally applied flow on symmetry degenerated waves\npropagating into opposite directions and standing waves that exchange stability\nwith the traveling waves via mixed states is analyzed. Wave structures that\nconsist of spiral vortices in the counter rotating Taylor-Couette system are\ninvestigated by full numerical simulations and explained quantitatively by\namplitude equations containing quintic coupling terms. The latter are\nappropriate to describe the influence of inversion symmetry breaking\nperturbations on many oscillatory instabilities with O(2) symmetry.", "category": "physics_flu-dyn" }, { "text": "Validity of sound-proof approximations for magnetic buoyancy: The presence of acoustic waves in models of compressible flows can present\ncomplications for analytical and numerical analysis. Therefore, several methods\nhave been developed to filter out these waves, leading to various \"sound-proof\"\nmodels, including the Boussinesq, anelastic and pseudo-incompressible models.\nWe assess the validity of each of these approximate models for describing\nmagnetic buoyancy in the context of the solar interior. A general sound-proof\nmodel is introduced and compared to the fully compressible system in a number\nof asymptotic regimes, including both non-rotating and rotating cases. We\nobtain specific constraints that must be satisfied in order that the model\ncaptures the leading-order behaviour of the fully compressible system. We then\ndiscuss which of the existing sound-proof models satisfy these constraints, and\nin what parameter regimes. We also present a variational derivation of the\npseudo-incompressible MHD model, demonstrating its underlying Hamiltonian\nstructure.", "category": "physics_flu-dyn" }, { "text": "Oriented suspension mechanics with applications to flow linear dichrosim\n spectroscopy and pathogen detection: Flow linear dichroism is a biophysical spectroscopic technique that exploits\nthe shear-induced alignment of elongated particles in suspension. Motivated by\nthe broad aim of optimising the sensitivity of this technique, and more\nspecifically by a handheld synthetic biotechnology prototype for\nwaterborne-pathogen detection, a model of steady and oscillating\npressure-driven channel flow and orientation dynamics of a suspension of\nslender microscopic fibres is developed. The model couples the Fokker-Planck\nequation for Brownian suspensions with the narrow channel flow equations, the\nlatter modified to incorporate mechanical anisotropy induced by the particles.\nThe linear dichroism signal is estimated through integrating the perpendicular\ncomponents of the distribution function via an appropriate formula which takes\nthe bi-axial nature of the orientation into account. For the specific\napplication of pathogen detection via binding of M13 bacteriophage, it is found\nthat increases in the channel depth are more significant in improving the\nlinear dichroism signal than increases in the channel width. Increasing channel\ndepth to $2~$mm and pressure gradient to $5 \\times 10^4~$Pa/m essentially\nmaximises the alignment. Oscillating flow can produce nearly equal alignment to\nsteady flow at appropriate frequencies, which has significant potential\npractical value in the analysis of small sample volumes.", "category": "physics_flu-dyn" }, { "text": "Stationary subsonic boundary layer in the regions of local heating of\n surface: The problem of disturbed flow in the laminar boundary layer is investigated\nwhen a heating element is located on the surface of the body. The flow is\nsupposed to be laminar one. It is shown that the problem may be solved in terms\nof free interaction theory. The solution of linear flat problem is constructed.\nThe results of asymptotic analysis are also presented.", "category": "physics_flu-dyn" }, { "text": "Velocity fluctuations in a dilute suspension of viscous vortex rings: We explore the velocity fluctuations in a fluid due to a dilute suspension of\nrandomly-distributed vortex rings at moderate Reynolds number, for instance\nthose generated by a large colony of jellyfish. Unlike previous analysis of\nvelocity fluctuations associated with gravitational sedimentation or\nsuspensions of microswimmers, here the vortices have a finite lifetime and are\nconstantly being produced. We find that the net velocity distribution is\nsimilar to that of a single vortex, except for the smallest velocities which\ninvolve contributions from many distant vortices; the result is a truncated\n$5/3$-stable distribution with variance (and mean energy) linear in the vortex\nvolume fraction $\\phi$. The distribution has an inner core with a width scaling\nas $\\phi^{3/5}$, then long tails with power law $|u|^{-8/3}$, and finally a\nfixed cutoff (independent of $\\phi$) above which the probability density scales\nas $|u|^{-5}$, where $u$ is a component of the velocity. We argue that this\ndistribution is robust in the sense that the distribution of any velocity\nfluctuations caused by random forces localized in space and time has the same\nproperties, except possibly for a different scaling after the cutoff.", "category": "physics_flu-dyn" }, { "text": "Microgravity and Dissipative Granular Gas in a vibrated container: a gas\n with an asymmetric speed distribution in the vibration direction, but with a\n null mean speed everywhere: The main topic of this paper (part 4) is the interpretation of data from\nextended simulations published in previous Poudres & Grains (see P&G 17, #1 to\n#18) concerning the dynamics of N equal-size spheres in a 3d rectangular cell\nexcited along Oz in 0 gravity.(N=100, 500, 1000, 1200, 2000, 3000, 4000, 4500).\nDifferent Oz excitation kinds have been used (symmetric and non symmetric\nbi-parabolic, symmetric and non symmetric saw teeth, thermal wall). No rotation\nis included, dissipation is introduced via a restitution coefficient e=\n-V'n/Vn, where V'n and Vn are the relative ball speed along normal to ball\ncentres after and before collision. It is proved that the local speed\ndistribution along z is fundamentally dissymmetric in most part of the cell\nwhile the mean local speed is 0. This demonstrates the inability of a model\nbased on a thermal bath (with a single local temperature) to describe this\ndissipative-granular-gas-system, even when assuming that this temperature\nvaries in space. The other (1-3) parts sum up few results obtained in the very\nlow density regime.", "category": "physics_flu-dyn" }, { "text": "Self-duality of the asymptotic relaxation states of fluids and plasmas: The states of asymptotic relaxation of 2-dimensional fluids and plasma\npresent a high degree of regularity and obey to the sinh-Poisson equation. We\nfind that embedding the classical fluid description into a field-theoretical\nframework, the same equation appears as a manifestation of the self-duality.", "category": "physics_flu-dyn" }, { "text": "Unified Flow and Thermal Law-of-the-wall for Type-A and B Turbulent\n Boundary Layers: Systematic study of the existing statistical data from direct numerical\nsimulations (DNS) lead to a logical and important classification of generic\nturbulent boundary layers (TBL), namely Type A,B and C TBL, based on the\ndistribution patterns of time averaged wall shear stress. Among these three\ntypes, Type A and B TBLs are investigated in terms of its universal statistical\nlaws of velocity and temperature with independency on both Reynolds (Re) and\nPrandtl (Pr) numbers. With reference to the analysis by von Karman in\ndeveloping the conventional law of the wall for Type A TBL, the current study\nfirst physically distinguishes the time averaged local velocity scale from the\nensemble averaged velocity scale, and properly defines the TBL temperature\nscale based on wall heat flux, similar to the well known frictional velocity\nscale based on wall shear stress. The unified law of the wall formulations are\nthen derived for the inner layer of Type A and B thermal TBLs using these\nscales. The formulations are constructed by introducing the general damping and\nenhancing functions in inner layer, and further, these functions are extended\nto logarithmic coordinate based on the multiscale characteristics of TBL\nvelocity profiles. The validations of the law are conducted both by the DNS\nguided near wall integration of Type B TBL governing equations, which uncovered\nthe Re and Pr independency mechanisms of the law. The research advances the\ncurrent understandings of the conventional TBL theory and develop the complete\nanalytical forms of law of the wall for Type A and B TBLs, including inner\nlayer, transition layer, logarithmic layer and wake layer.", "category": "physics_flu-dyn" }, { "text": "Shear instability of an axisymmetric air-water coaxial jet: We study the destabilization of a round liquid jet by a fast annular gas\nstream. We measure the frequency of the shear instability waves for several\ngeometries and air/water velocities. We then carry out a linear stability\nanalysis, and show that there are three competing mechanisms for the\ndestabilization: a convective instability, an absolute instability driven by\nsurface tension, and an absolute instability driven by confinement. We compare\nthe predictions of this analysis with experimental results, and propose scaling\nlaws for wave frequency in each regime. We finally introduce criteria to\npredict the boundaries between these three regimes.", "category": "physics_flu-dyn" }, { "text": "Experimental and computational investigation of single particle behavior\n in low Reynolds number linear shear flows: Trajectories of a buoyant spherical solid particle in a linear shear flow\nwere investigated at low Reynolds numbers. A two-dimensional CFD analysis was\nperformed to simulate the solid-fluid flows. Our numerical model, the discrete\nphase element method, was used to model and simulate the fluid domain and\nparticle motion as the solid phase. The reliability of the computational\nresults was evaluated for the particle trajectory. The agreement between the\nnumerical results with the experimental data was quantified.", "category": "physics_flu-dyn" }, { "text": "Nonlinear dispersion relation in integrable turbulence: The concept of Nonlinear dispersion relation (NDR) is used in various fields\nof Physics (nonlinear optics, hydrodynamics, hydroelasticity, mechanics,\nquantum optics, plasma physics,...) to characterize fundamental phenomena\ninduced by nonlinearity such as wave frequency shift or turbulence. Nonlinear\nrandom waves described by the onedimensional nonlinear Schrodinger equation\n(1DNLSE) exhibit a remarkable form of turbulence called \"integrable turbulence\"\nwhere solitons play a key role. Surprisingly, little attention has been paid to\nthe NDR of such universal wave systems up to a very recent theoretical study.\nHere, by using an original strategy, we report the accurate measurement of NDR\nof the slowly varying envelop of the waves in one-dimensional deep water waves\nexperiments. We characterize precisely the frequency shift and the broadening\nof the NDR, which interestingly reveals the presence of solitons and of high\norder effects. Our results highlight the relevance of the NDR in the context of\nintegrable turbulence.", "category": "physics_flu-dyn" }, { "text": "Spirographic motion in a vortex: Studies of particle motion in vortical flows have mainly focused on\npoint-like particles, either inertial or self-propelled. This approximation\nassumes that the velocity field that surrounds the particle is linear. We\nconsider an inertialess rigid dumbbell in a two-dimensional steady vortex.\nWhile the system remains analytically tractable, the particle experiences the\nnonlinearity of the surrounding velocity field. By exploiting the rotational\nsymmetry of the flow, we reduce the problem to that of a two-dimensional\ndynamical system, whose fixed points and periodic orbits can be used to explain\nthe motion of the dumbbell. For all vortices in which the fluid angular\nvelocity decreases with radial distance, the center of mass of the dumbbell\nfollows a spirographic trajectory around the vortex center. This results from a\nperiodic oscillation in the radial direction combined with revolution around\nthe center. The shape of the trajectory depends strongly on the initial\nposition and orientation of the dumbbell, but the dynamics is qualitatively the\nsame irrespective of the form of the vortex. If the fluid angular velocity is\nnot monotonic, the spirographic motion is altered by the existence of transport\nbarriers, whose shape is now sensitive to the details of the vortex.", "category": "physics_flu-dyn" }, { "text": "Motion of the vitreous humour in a deforming eye -- fluid-structure\n interaction between a nonlinear elastic solid and a nonlinear viscoleastic\n fluid: We study the motion of vitreous humour in a deforming eyeball. From the\nmechanical and computational perspective this is a task to solve a\nfluid-structure interaction problem between a complex viscoelastic fluid\n(vitreour humour) and a nonlinear elastic solid (sclera and lens). We propose a\nnumerical methodology capable of handling the fluid-structure interaction\nproblem, and we demonstrate its applicability via solving the corresponding\ngoverning equations in a realistic geometrical setting and for realistic\nparameter values. It is shown that the choice of the rheological model for the\nvitreous humour has a negligible influence on the overall flow pattern in the\ndomain of interest, whilst it is has a significant impact on the mechanical\nstress distribution in the domain of interest.", "category": "physics_flu-dyn" }, { "text": "Local flow measurements around flexible filaments under rotating\n magnetic field: Effective mixing of fluids at the microfluidic scale is important for future\napplications in biology, medicine, and chemistry. A promising type of\nmicromixers are magnetic filaments, which can be activated by an external\nmagnetic field. However, there is a lack of research that combines experiments\nand numerical modelling of hydrodynamics around such filaments. Here we use\nmicro-particle image velocimetry to measure flow fields around rotating\nflexible ferromagnetic filaments and compare them to numerical data from\nelastic rod model. We measure that rotating filaments hover above the surface,\nwhereas the resulting fluid velocities are highly dependent on the hovering\ndistance. We also find that the rotating filament causes a 3D flow coming from\nthe rotational plane and being extracted along the axis of rotation. These\nfindings will help develop better micromixers.", "category": "physics_flu-dyn" }, { "text": "Two-dimensional energy spectra in high Reynolds number turbulent\n boundary layers: Here we report the measurements of two-dimensional (2-D) spectra of the\nstreamwise velocity ($u$) in a high Reynolds number turbulent boundary layer. A\nnovel experiment employing multiple hot-wire probes was carried out at friction\nReynolds numbers ranging from 2400 to 26000. Taylor's frozen turbulence\nhypothesis is used to convert temporal-spanwise information into a 2-D spatial\nspectrum which shows the contribution of streamwise ($\\lambda_x$) and spanwise\n($\\lambda_y$) length scales to the streamwise variance at a given wall height\n($z$). At low Reynolds numbers, the shape of the 2-D spectra at a constant\nenergy level shows $\\lambda_y/z \\sim (\\lambda_x/z)^{1/2}$ behaviour at larger\nscales, which is in agreement with the existing literature at a matched\nReynolds number obtained from direct numerical simulations. However, at high\nReynolds numbers, it is observed that the square-root relationship tends\ntowards a linear relationship ($\\lambda_y \\sim \\lambda_x$) as required for\nself-similarity and predicted by the attached eddy hypothesis.", "category": "physics_flu-dyn" }, { "text": "Self-similar geometries within the inertial subrange of scales in\n boundary layer turbulence: The inertial subrange of turbulent scales is commonly reflected by a power\nlaw signature in ensemble statistics such as the energy spectrum and structure\nfunctions - both in theory and from observations. Despite promising findings on\nthe topic of fractal geometries in turbulence, there is no accepted image for\nthe physical flow features corresponding to this statistical signature in the\ninertial subrange. The present study uses boundary layer turbulence\nmeasurements to evaluate the self-similar geometric properties of velocity\nisosurfaces and investigate their influence on statistics for the velocity\nsignal. The fractal dimension of streamwise velocity isosurfaces, indicating\nstatistical self-similarity in the size of \"wrinkles\" along each isosurface, is\nshown to be constant only within the inertial subrange of scales. For the\ntransition between the inertial subrange and production range, it is inferred\nthat the largest wrinkles become increasingly confined by the overall size of\nlarge-scale coherent velocity regions such as uniform momentum zones. The\nself-similarity of isosurfaces yields power law trends in subsequent\none-dimensional statistics. For instance, the theoretical 2/3 power law\nexponent for the structure function can be recovered by considering the\ncollective behavior of numerous isosurface level sets. The results suggest that\nthe physical presence of inertial subrange eddies is manifested in the\nself-similar wrinkles of isosurfaces.", "category": "physics_flu-dyn" }, { "text": "Start-up and cessation of steady shear and extensional flows: Exact\n analytical solutions for the affine linear Phan-Thien--Tanner fluid model: Exact analytical solutions for start-up and cessation flows are obtained for\nthe affine linear Phan-Thien--Tanner fluid model. They include the results for\nstart-up and cessation of steady shear flows, of steady uniaxial and biaxial\nextensional flows, and of steady planar extensional flows. The solutions\nobtained show that at start-up of steady shear flows, the stresses go through\nquasi-periodic exponentially damped oscillations while approaching their\nsteady-flow values (so that stress overshoots are present); at start-up of\nsteady extensional flows, the stresses grow monotonically, while at cessation\nof steady shear and extensional flows, the stresses decay quickly and\nnon-exponentially. The steady-flow rheology of the fluid is also reviewed, the\nexact analytical solutions obtained in this work for steady shear and\nextensional flows being simpler than the alternative formulas found in the\nliterature. The properties of steady and transient solutions, including their\nasymptotic behavior at low and high Weissenberg numbers, are investigated in\ndetail. Generalization to the multimode version of the Phan-Thien--Tanner model\nis also discussed. Thus, this work provides a complete analytical description\nof the rheology of the affine linear Phan-Thien--Tanner fluid in start-up,\ncessation, and steady regimes of shear and extensional flows.", "category": "physics_flu-dyn" }, { "text": "Transition to chaos and modal structure of magnetized Taylor-Couette\n flow: Taylor-Couette flow is often used as a simplified model for complex rotating\nflows in the interior of stars and accretion disks. The flow dynamics in these\nobjects is influenced by magnetic fields. For example, quasi-Keplerian flows in\nTaylor-Couette geometry become unstable to a travelling or standing wave in an\nexternal magnetic field if the fluid is conducting; there is an instability\neven when the flow is hydrodynamically stable. This magnetorotational\ninstability leads to the development of chaotic states and, eventually,\nturbulence, when the cylinder rotation is sufficiently fast. The transition to\nturbulence in this flow can be complex, with the coexistence of parameter\nregions with spatio-temporal chaos and regions with quasi-periodic behaviour,\ninvolving one or two additional modulating frequencies. Although the unstable\nmodes of a periodic flow can be identified with Floquet analysis, here we adopt\na more flexible equation-free data-driven approach. We analyse the data from\nthe transition to chaos in the magnetized Taylor-Couette flow and identify the\nflow structures related to the modulating frequencies with Dynamic Mode\nDecomposition; this method is based on approximating nonlinear dynamics with a\nlinear infinite-dimensional Koopman operator. With the use of these structures,\none can construct a nonlinear reduced model for the transition.", "category": "physics_flu-dyn" }, { "text": "Hydro-dynamical models for the chaotic dripping faucet: We give a hydrodynamical explanation for the chaotic behaviour of a dripping\nfaucet using the results of the stability analysis of a static pendant drop and\na proper orthogonal decomposition (POD) of the complete dynamics. We find that\nthe only relevant modes are the two classical normal forms associated with a\nSaddle-Node-Andronov bifurcation and a Shilnikov homoclinic bifurcation. This\nallows us to construct a hierarchy of reduced order models including maps and\nordinary differential equations which are able to qualitatively explain prior\nexperiments and numerical simulations of the governing partial differential\nequations and provide an explanation for the complexity in dripping. We also\nprovide a new mechanical analogue for the dripping faucet and a simple\nrationale for the transition from dripping to jetting modes in the flow from a\nfaucet.", "category": "physics_flu-dyn" }, { "text": "Large-scale intermittency of liquid-metal channel flow in a magnetic\n field: We predict a novel flow regime in liquid metals under the influence of a\nmagnetic field. It is characterised by long periods of nearly steady,\ntwo-dimensional flow interrupted by violent three-dimensional bursts. Our\nprediction has been obtained from direct numerical simulations in a channel\ngeometry at low magnetic Reynolds number and translates into physical\nparameters which are amenable to experimental verification under laboratory\nconditions. The new regime occurs in a wide range of parameters and may have\nimplications for metallurgical applications.", "category": "physics_flu-dyn" }, { "text": "Electroosmosis in conducting nanofluidic channels: Theoretical modeling of electroosmosis through conducting (ideally\npolarizable) nanochannels is reported. Based on the theory of induced charge\nelectrokinetics, a novel nanofluidic system which possesses both adjustable ion\nselective characteristics and flexible flow control is proposed. Such\nnanofluidic devices operate only with very low gate control voltage applied on\nthe conductive walls of nanochannels, and thus even can be energized by normal\nbatteries. We believe that it is possible to use such metal-electrolyte\nconfigurations to overcome the difficulties met with conventional\nmetal-isolator-electrolyte systems for nanofluidic applications.", "category": "physics_flu-dyn" }, { "text": "Global instability of wing shock buffet: Shock buffet on wings encountered in edge-of-the-envelope transonic flight\nremains an unresolved and disputed flow phenomenon, challenging both\nfundamental fluid mechanics and applied aircraft aerodynamics. The question of\nglobal instability leading to flow unsteadiness is addressed herein. It is\nshown for the first time that incipient three-dimensional shock buffet is\ngoverned by the dynamics of a single unstable linear eigenmode with its spatial\nstructure describing previously reported buffet cells. An inner-outer Krylov\napproach is proposed to solve the arising large-scale eigenvalue problem\niteratively using shift-and-invert spectral transformation and sparse iterative\nlinear solver. The established numerical strategy with an industrial flow\nsolver means that a practical non-canonical test case at high-$Re$ flow\ncondition can be investigated. A modern wing design with publicly available\ngeometry and experimental data for code validation is studied. The numerical\nfindings can be exploited for routes to flow control and model reduction.", "category": "physics_flu-dyn" }, { "text": "Global and local conservation of mass, momentum and kinetic energy in\n the simulation of compressible flow: The spatial discretization of convective terms in compressible flow equations\nis studied from an abstract viewpoint, for finite-difference methods and\nfinite-volume type formulations with cell-centered numerical fluxes. General\nconditions are sought for the local and global conservation of primary (mass\nand momentum) and secondary (kinetic energy) invariants on Cartesian meshes.\nThe analysis, based on a matrix approach, shows that sharp criteria for global\nand local conservation can be obtained and that in many cases these two\nconcepts are equivalent. Explicit numerical fluxes are derived in all\nfinite-difference formulations for which global conservation is guaranteed,\neven for non-uniform Cartesian meshes. The treatment reveals also an intimate\nrelation between conservative finite-difference formulations and cell-centered\nfinite-volume type approaches. This analogy suggests the design of wider\nclasses of finite-difference discretizations locally preserving primary and\nsecondary invariants.", "category": "physics_flu-dyn" }, { "text": "Competitive electrohydrodynamic and electrosolutal advection arrests\n evaporation kinetics of droplets: The present article reports the hitherto unreported phenomenon of arrested\nevaporation dynamics in pendent droplets in an electric field ambience. The\nevaporation kinetics of pendant droplets of electrically conducting saline\nsolutions in the presence of a transverse, alternating electric field is\ninvestigated experimentally. It has been observed that while increase of field\nstrength arrests the evaporation, increment in field frequency has the opposite\neffect. The same has been explained on the solvation kinetics of the ions in\nthe polar water. Theoretical analysis reveals that change in surface tension\nand diffusion driven evaporation model cannot predict the arrested or\ndecelerated evaporation. With the aid of Particle Image Velocimetry,\nsuppression of internal circulation velocity within the droplet is observed\nunder electric field stimulus, and this affects the evaporation rate directly.\nA mathematical scaling model is proposed to quantify the effects of\nelectrohydrodynamic circulation, electrothermal and electro-solutal advection\non the evaporation kinetics of the droplet. The analysis encompasses major\ngoverning parameters, viz. the thermal and solutal Marangoni numbers, the\nElectrohydrodynamic number, the electro Prandtl and electro Schmidt numbers and\ntheir respective contributions. It has been shown that the electrothermal\nMarangoni effect is supressed by the electric field, leading to deteriorated\nevaporation rates. Additionally, the electrosolutal Marangoni effect further\nsupresses the internal advection, which again arrests the evaporation rate by a\nlarger proportion. Stability analysis reveals that the electric body force\nretards the stable internal circulation within such droplets and arrests\nadvection.", "category": "physics_flu-dyn" }, { "text": "Manipulation of a Micro-object Using Topological Hydrodynamic Tweezers: Manipulating micro-scale object plays paramount roles in a wide range of\nfundamental researches and applications. At micro-scale, various methods have\nbeen developed in the past decades, including optical, electric, magnetic,\naerodynamic and acoustic methods. However, these non-contact forces are\nsusceptible to external disturbance, and so finding a way to make micro-scale\nobject manipulation immune to external perturbations is challenging and remains\nelusive. Here we demonstrate a method based on new trapping mechanism to\nmanipulate micro-scale object in a gas flow at ambient conditions. We first\nshow that the micro-droplet is entrapped into a trapping ring constructed by a\nparticular toroidal vortex. The vortex works as tweezers to control the\nposition of the micro-droplet. We then show that the micro-droplet can be\ntransported along the trapping ring. By virtue of the topological character of\nthe gas flow, the transport path is able to bypass external strong\nperturbations automatically. We further demonstrate a topological transfer\nprocess of the micro-droplet between two hydrodynamic tweezers. Our method\nprovides an integrated toolbox to manipulate a micro-scale object, with an\nintrinsic mechanism that protects the target object from external disturbances.", "category": "physics_flu-dyn" }, { "text": "Two-dimensionally stable self-organization arises in simple schooling\n swimmers through hydrodynamic interactions: We present new constrained and free-swimming experiments and simulations of a\npair of pitching hydrofoils interacting in a minimal school. The hydrofoils\nhave an out-of-phase synchronization and they are varied through in-line,\nstaggered, and side-by-side formations within the two-dimensional interaction\nplane. It is discovered that there is a \\textit{two-dimensionally} stable\nequilibrium point for a side-by-side formation. In fact, this formation is\nsuper-stable, meaning that hydrodynamic forces will passively maintain this\nformation even under external perturbations and the school as a whole has no\nnet forces acting on it that cause it to drift to one side or the other.\nMoreover, previously discovered \\textit{one-dimensionally} stable equilibria\ndriven by wake vortex interactions are shown to be, in fact, two-dimensionally\n\\textit{unstable}, at least for an out-of-phase synchronization. Additionally,\nit is discovered that a trailing-edge vortex mechanism provides the restorative\nforce to stabilize a side-by-side formation. The stable equilibrium is further\nverified by experiments and simulations for freely-swimming foils where dynamic\nrecoil motions are present. When constrained, swimmers in compact side-by-side\nformations experience collective efficiency and thrust increases up to 40\\% and\n100\\%, respectively, whereas slightly staggered formations output an even\nhigher efficiency improvement of 84\\% with a 87\\% increase in thrust.\nFreely-swimming foils in a stable side-by-side formation show an efficiency and\nspeed enhancement of up to 9\\% and 15\\%, respectively. These newfound schooling\nperformance and stability characteristics suggest that fluid-mediated\nequilibria may play a role in the control strategies of schooling fish and\nfish-inspired robots.", "category": "physics_flu-dyn" }, { "text": "Flow structure beneath periodic waves with constant vorticity under\n strong horizontal electric Fields: While several articles have been written on Electrohydrodynamics (EHD) flows\nor flows with constant vorticity separately, little is known about the extent\nto which the combined effects of EHD and constant vorticity affect the flow.\nThis study aims to fill this gap by investigating how a horizontal electric\nfield and constant vorticity jointly influence the free surface and the\nemergence of stagnation points. Using the Euler equations framework, we employ\nconformal mapping and pseudo-spectral numerical methods. Our findings reveal\nthat increasing the electric field intensity eliminates stagnation points and\nsmoothen the wave profile. This implies that a horizontal electric field acts\nas a mechanism for the elimination of stagnation points within the fluid body.", "category": "physics_flu-dyn" }, { "text": "IB2d: a Python and MATLAB implementation of the immersed boundary method: The development of fluid-structure interaction (FSI) software involves\ntrade-offs between ease of use, generality, performance, and cost. Typically\nthere are large learning curves when using low-level software to model the\ninteraction of an elastic structure immersed in a uniform density fluid. Many\nexisting codes are not publicly available, and the commercial software that\nexists usually requires expensive licenses and may not be as robust or allow\nthe necessary flexibility that in house codes can provide. We present an open\nsource immersed boundary software package, IB2d, with full implementations in\nboth MATLAB and Python, that is capable of running a vast range of biomechanics\nmodels and is accessible to scientists who have experience in high-level\nprogramming environments. IB2d contains multiple options for constructing\nmaterial properties of the fiber structure, as well as the advection-diffusion\nof a chemical gradient, muscle mechanics models, and artificial forcing to\ndrive boundaries with a preferred motion.", "category": "physics_flu-dyn" }, { "text": "Wavy regimes of film flow down a fibre: We consider axisymmetric traveling waves propagating on the gravity-driven\nflow of a liquid down a vertical fibre. Our starting point is the two-equation\nmodel for the flow derived in the study by Ruyer-Quil \\emph{et al.} [\\emph{J.\nFluid Mech.} {\\bf 603}, 431 (2008)]. The speed, amplitude and shape of the\ntraveling waves are obtained for a wide range of parameters by using asymptotic\nanalysis and elements from dynamical systems theory. Four different regimes are\nidentified corresponding to the predominance of four different physical\neffects: Advection by the flow, azimuthal curvature, inertia and viscous\ndispersion. Construction of the traveling-wave branches of solutions reveals\ncomplex transitions from one regime to another. A phase diagram of the\ndifferent regimes in the parameter space is oferred.", "category": "physics_flu-dyn" }, { "text": "Spectral analysis for elastica 3-dimensional dynamics in a shear flow: We present the spectral analysis of three-dimensional dynamics of an elastic\nfilament in a shear flow of a viscous fluid at a low Reynolds number in the\nabsence of Brownian motion. The elastica model is used. The fiber initially is\nalmost straight at an arbitrary orientation, with small perpendicular\nperturbations in the shear plane and out-of-plane. To analyze the stability of\nboth perturbations, equations for the eigenvalues and eigenfunctions are\nderived and solved by the Chebyshev spectral collocation method. It is shown\nthat their crucial features are the same as in the case of the two-dimensional\nelastica dynamics in shear flow [Becker and Shelley, Phys. Rev. Lett. 2001] and\nthe three-dimensional elastica dynamics in the compressional flow [Chakrabarti\net al., Nat. Phys., 2020]. We find a similar dependence of the buckled shapes\non the ratio of bending to hydrodynamic forces as in the simulations for\nelastic fibers of a nonzero thickness [Slowicka et al., New J. Phys., 2022].", "category": "physics_flu-dyn" }, { "text": "Partially-averaged Navier-Stokes (PANS) Method for Turbulence\n Simulations: Near-wall Modeling and Smooth-surface Separation Computations: The goal of this dissertation is to investigate the PANS model capabilities\nin providing significant improvement over RANS predictions at slightly higher\ncomputational expense and producing LES quality results at significantly lower\ncomputational cost. The objectives of this study are: (i) investigate the model\nfidelity at a fixed level of scale resolution (Generation1-PANS/G1-PANS) for\nsmooth surface separation, (ii) Derive the PANS closure model in regions of\nresolution variation (Generation2-PANS/G2-PANS), and (iii) Validate G2-PANS\nmodel for attached and separated flows. The separated flows considered in this\nstudy have been designated as critical benchmark flows by NASA CFD study group.\n The key contributions of this dissertation are summarized as follows. The\nturbulence closure model of varying resolution, G2-PANS, is developed by\nderiving mathematically-consistent commutation residues and using energy\nconservation principles. The log-layer recovery and accurate computation of\nReynolds stress anisotropy is accomplished by transitioning from steady RANS to\nscaled resolved simulations using the G2-PANS model. Finally, several\nsmooth-separation flows on the NASA turbulence website have been computed with\nhigh degree of accuracy at a significantly reduced computational effort over\nLES using the G1-PANS and G2-PANS models.", "category": "physics_flu-dyn" }, { "text": "Towards indirect assessment of surface anomalies on wind turbine rotor\n blades: We present results from novel field, lab and computer studies, that pave the\nway towards non-invasive classification of localised surface defects on running\nwind turbine rotors using infrared thermography (IRT). In particular, we first\nparametrise the problem from a fluid dynamical point of view using the\nroughness Reynolds number ($Re_k$) and demonstrate how the parameter regime\nrelevant for modern wind turbines translate to parameter values that are\ncurrently feasible in typical wind tunnel and computer experiments. Second, we\ndiscuss preparatory wind tunnel and field measurements, that demonstrate a\npromising degree of sensitivity of the recorded IRT data w.r.t. the key control\nparameter ($Re_k$), which is a minimum requirement for the proposed\nclassification technique to work. Third, we introduce and validate a local\ndomain ansatz for future computer experiments, that enables well-resolved\nNavier-Stokes simulations for the target parameter regime at reasonable\ncomputational costs.", "category": "physics_flu-dyn" }, { "text": "Role of correlation integrals in helical isotropic turbulence: It is shown that the correlation integrals (invariants) play an important,\nand at certain conditions a dominant, role in helical isotropic homogeneous\nturbulence both in the inertial and near-dissipation ranges. Results of direct\nnumerical simulations and recent laboratory experiments with multiscale grids\nhave been used for this purpose. A possibility of spontaneous breaking of\nreflection symmetry has been also briefly discussed in this context.", "category": "physics_flu-dyn" }, { "text": "Exact Solutions of the Equations of Motion of Liquid Helium with a\n Charged Free Surface: The dynamics of the development of instability of the free surface of liquid\nhelium, which is charged by electrons localized above it, is studied. It is\nshown that, if the charge completely screens the electric field above the\nsurface and its magnitude is much larger then the instability threshold, the\nasymptotic behavior of the system can be described by the well-known 3D\nLaplacian growth equations. The integrability of these equations in 2D geometry\nmakes it possible to described the evolution of the surface up to the formation\nof singularities, viz., cuspidal point at which the electric field strength,\nthe velocity of the liquid, and the curvature of its surface assume infinitely\nlarge values. The exact solutions obtained for the problem of the\nelectrocapillary wave profile at the boundary of liquid helium indicate the\ntendency to a charge in the surface topology as a result of formation of\ncharged bubbles.", "category": "physics_flu-dyn" }, { "text": "A constitutive model for volume-based descriptions of gas flows: We derive the pressure tensor and the heat flux to accompany the new\nmacroscopic conservation equations that we developed previously in a\nvolume-based kinetic framework for gas flows. This kinetic description allows\nfor expansion or compression of a fluid element, and is characterized by a flux\nof volume that conventional modelling does not account for. A new convective\nform of transport arises from this new approach, which is due purely to\nmacroscopic expansion or compression. It supplements the conventional transport\nprocesses, which are transport due purely to mass motion (classical convective\ntransport) and transport due purely to random motion (diffusive transport). In\naddition, we show that the thermodynamic parameters of the fluid (temperature\nand pressure) appear with new non-equilibrium contributions that depend on\ndensity variations in the gas. Our new model may be useful for describing gas\nflows that display non-local-thermodynamic-equilibrium (rarefied gas flows),\nflows with relatively large variations of macroscopic properties, and/or highly\ncompressible fluids and flows.", "category": "physics_flu-dyn" }, { "text": "Contribution of a time-dependent metric on the dynamics of an interface\n between two immiscible electro-magnetically controllable Fluids: We consider the case of a deformable material interface between two\nimmiscible moving media, both of them being magnetiable. The time dependence of\nthe metric at the interface introduces a non linear term, proportional to the\nmean curvature, in the surface dynamical equations of mass momentum and angular\nmomentum. We take into account the effects of that term also in the singular\nmagnetic and electric fields inside the interface which lead to the existence\nof currents and charges densities through the interface, from the derivation of\nthe Maxwell equations inside both bulks and the interface. Also, we give the\nexpression for the entropy production and of the different thermo-dynamical\nfluxes. Our results enlarge previous results from other theories where the\nspecific role of the time dependent surface metric was insufficiently stressed.", "category": "physics_flu-dyn" }, { "text": "Turbulent Friction in Rough Pipes and the Energy Spectrum of the\n Phenomenological Theory: The classical experiments on turbulent friction in rough pipes were performed\nby J. Nikuradse in the 1930's. Seventy years later, they continue to defy\ntheory. Here we model Nikuradse's experiments using the phenomenological theory\nof Kolmog\\'orov, a theory that is widely thought to be applicable only to\nhighly idealized flows. Our results include both the empirical scalings of\nBlasius and Strickler, and are otherwise in minute qualitative agreement with\nthe experiments; they suggest that the phenomenological theory may be relevant\nto other flows of practical interest; and they unveil the existence of close\nties between two milestones of experimental and theoretical turbulence.", "category": "physics_flu-dyn" }, { "text": "On angled bounce-off impact of a drop impinging on a flowing soap film: Small drops impinging angularly on thin flowing soap films frequently\ndemonstrate the rare emergence of bulk elastic effects working in-tandem with\nthe more common-place hydrodynamic interactions. Three collision regimes are\nobservable: (a) drop piercing through the film, (b) it coalescing with the\nflow, and (c) it bouncing off the film surface. During impact, the drop deforms\nalong with a bulk elastic deformation of the film. For impacts that are\nclose-to-tangential, the bounce-off regime predominates. We outline a reduced\norder analytical framework assuming a deformable drop and a deformable\nthree-dimensional film, and the idealization invokes a phase-based parametric\nstudy. Angular inclination of the film and the ratio of post and pre impact\ndrop sizes entail the phase parameters. We also perform experiments with\nvertically descending droplets impacting against an inclined soap film, flowing\nunder constant pressure head. Model predicted phase domain for bounce-off\ncompares well to our experimental findings. Additionally, the experiments\nexhibit momentum transfer to the film in the form of shed vortex dipole, along\nwith propagation of free surface waves. On consulting prior published work, we\nnote that for locomotion of water-walking insects using an impulsive action,\nthe momentum distribution to the shed vortices and waves are both significant,\ntaking up respectively 2/3-rd and 1/3-rd of the imparted streamwise momentum.\nIn view of the potentially similar impulse actions, this theory is applied to\nthe bounce-off examples in our experiments, and the resultant shed vortex\ndipole momenta are compared to the momenta computed from particle imaging\nvelocimetry data. The magnitudes reveal identical order ($10^{-7}$ N$\\cdot$s),\nsuggesting that the bounce-off regime can be tapped as a simple analogue for\ninterfacial bio-locomotion relying on impulse reactions.", "category": "physics_flu-dyn" }, { "text": "Data-driven spectral turbulence modeling for Rayleigh-B\u00e9nard\n convection: A data-driven turbulence model for coarse-grained numerical simulations of\ntwo-dimensional Rayleigh-B\\'enard convection is proposed. The model starts from\nhigh-fidelity data and is based on adjusting the Fourier coefficients of the\nnumerical solution, with the aim of accurately reproducing the kinetic energy\nspectra as seen in the high-fidelity reference findings. No assumptions about\nthe underlying PDE or numerical discretization are used in the formulation of\nthe model. We also develop a constraint on the heat flux to guarantee accurate\nNusselt number estimates on coarse computational grids and high Rayleigh\nnumbers. Model performance is assessed in coarse numerical simulations at\n$Ra=10^{10}$. We focus on key features including kinetic energy spectra,\nwall-normal flow statistics, and global flow statistics. The method of\ndata-driven modeling of flow dynamics is found to reproduce the reference\nkinetic energy spectra well across all scales and yields good results for flow\nstatistics and average heat transfer, leading to computationally cheap\nsurrogate models. Large-scale forcing extracted from the high-fidelity\nsimulation leads to accurate Nusselt number predictions across two decades of\nRayleigh numbers, centered around the targeted reference at $Ra=10^{10}$.", "category": "physics_flu-dyn" }, { "text": "Flow force and torque on submerged bodies in lattice-Boltzmann via\n momentum exchange: We present a new derivation of the momentum exchange method to compute the\nflow force and torque on a submerged body in lattice Boltzmann methods. Our\nderivation does not depend on a particular implementation of the boundary\nconditions at the body surface and relies on general principles. We recover\nsome well known expressions, in some cases with slight corrections, to treat\nthe cases of static and moving bodies. We also present some numerical tests\nthat support the correctness of the formulas derived.", "category": "physics_flu-dyn" }, { "text": "Assessment of flamelet manifolds for turbulent flame-wall interactions\n in Large-Eddy Simulations: A turbulent side-wall quenching (SWQ) flame in a fully developed channel flow\nis studied using Large-Eddy Simulation (LES) with a tabulated chemistry\napproach. Three different flamelet manifolds with increasing levels of\ncomplexity are applied: the Flamelet-Generated Manifold (FGM) considering\nvarying enthalpy levels, the Quenching Flamelet-Generated Manifold (QFM), and\nthe recently proposed Quenching Flamelet-Generated Manifold with Exhaust Gas\nRecirculation (QFM-EGR), with the purpose being to assess their capability to\npredict turbulent flame-wall interactions (FWIs), which are highly relevant to\nnumerical simulations of real devices such as gas turbines and internal\ncombustion engines.\n The accuracy of the three manifolds is evaluated and compared a posteriori,\nusing the data from a previously published flame-resolved simulation with\ndetailed chemistry for reference. For LES with the FGM, the main\ncharacteristics such as the mean flow field, temperature, and major species can\nbe captured well, while notable deviations from the reference results are\nobserved for the near-wall region, especially for pollutant species such as\n\\ce{CO}. In accordance with the findings from laminar FWI, improvement is also\nobserved in the simulation with QFM under turbulent flow conditions. Although\nLES with the QFM-EGR shows a similar performance in the prediction of mean\nquantities as LES with QFM, it presents significantly better agreement with the\nreference data regarding instantaneous thermo-chemical states near the\nquenching point. This indicates the necessity to take into account the mixing\neffects in the flamelet manifold to correctly capture the flame-vortex\ninteraction near the flame tip in turbulent configurations. Based on the\nfindings from this study, suitable flamelet manifolds can be chosen depending\non the aspects of interest in practical applications.", "category": "physics_flu-dyn" }, { "text": "Dynamics of Vapor Layer Under a Leidenfrost Drop: In the Leidenfrost effect a small drop of fluid is levitated above a\nsufficiently hot surface, on a persistent vapor layer generated by evaporation\nfrom the drop. The vapor layer thermally insulates the drop from the surface\nleading to extraordinarily long drop lifetimes. The top-view shape of the\nlevitated drops can exhibit persistent star-like vibrations. I extend recent\nwork [Burton et al. PRL 2012] to study the bottom surface of the drop using\ninterference-imaging. In this work I use a high-speed camera and automated\nimage analysis to image, locate and classify the interference fringes. From the\ninterference fringes I reconstruct the shape and height profile of the rim\nwhere the drop is closest to the surface. I measure the drop-size dependence of\nthe planar vibrational mode frequencies, which agree well with previous work. I\nobserve a distinct breathing mode in the average radius of the drop, the\nfrequency of which scales differently with drop size than the other modes. This\nbreathing mode can be tightly coupled to a vertical motion of the drop. I\nfurther observe a qualitative difference in the structure and dynamics of the\nvertical profile of the rim between large and small drops.", "category": "physics_flu-dyn" }, { "text": "Numerical simulation of two-dimensional detonation propagation in\n partially pre-vaporized n-heptane sprays: In this paper, two dimensional detonation propagation in partially\nprevaporized n-heptane sprays is studied by using Eulerian/Lagrangian methods.\nThe effects of droplet preevaporation on the detonation propagation are\ninvestigated. The general features and detailed structures of two-phase\ndetonations are well captured with the present numerical methods. The results\nshow that the detonation propagation speed and detonation structures are\nsignificantly affected by the preevaporated gas equivalence ratio. The\nnumerical soot foils are used to characterize the influence of preevaporated\ngas equivalence ratio on the detonation propagation. Regular detonation\ncellular structures are observed for large preevaporated gas equivalence\nratios, but when decreasing the preevaporated gas equivalence ratio, the\ndetonation cellular structures become much more unstable and the average cell\nwidth also increases. It is also found that the preevaporated gas equivalence\nratio has little effects on the volume averaged heat release when the\ndetonation propagates stably. Moreover, the results also suggest that the\ndetonation can propagate in the two-phase heptane and air mixture without\npreevaporation, but the detonation would be first quenched and then re-ignited\nwhen the preevaporated gas equivalence ratio is small or equal to zero.", "category": "physics_flu-dyn" }, { "text": "Evaporation of dilute droplets in a turbulent jet: clustering and\n entrainment effects: Droplet evaporation in turbulent sprays involves unsteady, multiscale and\nmultiphase processes which make its comprehension and model capabilities still\nlimited. The present work aims to investigate droplet vaporization dynamics\nwithin a turbulent spatial developing jet in dilute, non-reacting conditions.\nWe address the problem using a Direct Numerical Simulation of jet laden with\nacetone droplets using an hybrid Eulerian/Lagrangian approach based on the\npoint droplet approximation. A detailed statistical analysis of both phases is\npresented. In particular, we show how crucial is the preferential sampling of\nthe vapour phase induced by the inhomogeneous localization of the droplets\nthrough the flow. The preferential segregation of droplets develops suddenly\ndownstream the inlet both within the turbulent core and in the mixing layer.\nTwo distinct mechanisms have been found to drive these phenomena, the inertial\nsmall-scale clustering in the jet core and the intermittent dynamics of\ndroplets across the turbulent/non-turbulent interface in the mixing layer where\ndry air entrainment occurs. These phenomenologies strongly affect the overall\nvaporization process and lead to a spectacular widening of droplets size and\nvaporization rate distributions in the downstream evolution of the turbulent\nspray.", "category": "physics_flu-dyn" }, { "text": "Stochastic Dynamical Model of Intermittency in Fully Developed\n Turbulence: A novel model of intermittency is presented in which the dynamics of the\nrates of energy transfer between successive steps in the energy cascade is\ndescribed by a hierarchy of stochastic differential equations. The probability\ndistribution of velocity increments is calculated explicitly and expressed in\nterms of generalized hypergeometric functions of the type ${_n}F_0$, which\nexhibit power-law tails. The model predictions are found to be in good\nagreement with experiments on a low temperature gaseous helium jet. It is\nargued that distributions based on the functions ${_n}F_0$ might be relevant\nalso for other physical systems with multiscale dynamics.", "category": "physics_flu-dyn" }, { "text": "Generation of gravity waves by pedal-wavemakers: Experimental wave generation in channels is usually achieved through\nwavemakers (moving paddles) acting on the surface of the water. Although\npractical for engineering purposes, wavemakers have issues: they perform poorly\nin the generation of long waves and create evanescent waves in their vicinity.\nIn this article, we introduce a framework for wave generation through the\naction of an underwater multipoint mechanism: the pedal-wavemaking method. Our\nmultipoint action makes each point of the bottom move with a prescribed\npedalling-like motion. We analyse the linear response of waves in a uniform\nchannel in terms of the wavelength of the bottom action. The framework\nnaturally solves the problem of the performance for long waves and replaces\nevanescent waves by thin boundary layer at the bottom of the channel. We also\nshow that a proper synchronisation of orbital motion on the bottom can produce\nwaves that mimic deep water waves. This last feature has been proved to be\nuseful to study fluid-structure interaction in simulations based on smoothed\nparticle hydrodynamics.", "category": "physics_flu-dyn" }, { "text": "Turbulent energy cascade through equivalence of Euler and Lagrange\n motion descriptions and bifurcation rates: This work analyses the homogeneous isotropic turbulence by means of the\nequivalence between Euler and Lagrange representations of motion, adopting the\nbifurcation rates associated with Navier--Stokes and kinematic equations, and\nan appropriate hypothesis of fully developed chaos. The equivalence of these\nmotion descriptions allows to show that kinetic and thermal energy cascade\narise both from the convective term of Liouville equation. Accordingly, these\nphenomena, of nondiffusive nature, correspond to a transport in physical space\nlinked to the trajectories divergence. Both the bifurcation rates are properly\ndefined, where the kinematic bifurcation rate is shown to be much greater than\nNavier--Stokes bifurcation rate. This justifies the proposed hypothesis of\nfully developed chaos where velocity field and particles trajectories\nfluctuations are statistically uncorrelated. Thereafter, a specific ergodic\nproperty is presented, which relates the statistics of fluid displacement to\nthat of velocity and temperature fields. A detailed analysis of separation rate\nis proposed which studies the statistics of radial velocity component along the\nmaterial separation vector. Based on previous elements, the closure formulas of\nvon K\\'arm\\'an--Howarth and Corrsin equations are finally achieved. These\nclosures, of nondiffusive kind, represent a propagation phenomenon, and\ncoincide with those just presented by the author in previous works,\ncorroborating the results of these latter.", "category": "physics_flu-dyn" }, { "text": "Sinusoidal shaped hollow fibers for enhanced mass transfer: Inducing secondary flows and vortices is known to enhance mass transport.\nThey can be imposed by structured flow channels for instance. In particular,\nthese vortices reduce fouling and concentration polarization. In this work we\npresent a new method of producing hollow fiber membranes with a sinusoidal\nchange in diameter over the fiber length. We engineered a pulsation module that\nimposes a sinusoidally fluctuating bore liquid flow rate. Harmonic bore flow\nconditions can be varied over a wide range. The fluctuating bore liquid flow\nrate translates into axial membrane properties varying with respect to inner\nbore diameter and wall thickness. The resulting narrowing and widening of the\nmembrane lumen channel are hypothesized to induce secondary vortices to the\nliquid inside the membrane lumen known as the Bellhouse effect. Improved oxygen\ntransport from shell-to-lumen side prove superiority over straight hollow fiber\nmembranes in G/L absorption process. We anticipate the dynamic flow module to\nbe easily integrated into currently existing hollow fiber membrane spinning\nprocesses.", "category": "physics_flu-dyn" }, { "text": "A canonical Hamiltonian formulation of the Navier-Stokes problem: This paper presents a novel Hamiltonian formulation of the isotropic\nNavier-Stokes problem based on a minimum-action principle derived from the\nprinciple of least squares. This formulation uses the velocities\n$u_{i}(x_{j},t)$ and pressure $p(x_{j},t)$ as the field quantities to be\nvaried, along with canonically conjugate momenta deduced from the analysis.\nFrom these, a conserved Hamiltonian functional $H^{*}$ satisfying Hamilton's\ncanonical equations is constructed, and the associated Hamilton-Jacobi equation\nis formulated for both compressible and incompressible flows. This\nHamilton-Jacobi equation reduces the problem of finding four separate field\nquantities ($u_{i}$,$p$) to that of finding a single scalar functional in those\nfields--Hamilton's principal functional $\\text{S}^{*}[t,u_{i},p]$. Moreover,\nthe transformation theory of Hamilton and Jacobi now provides a prescribed\nrecipe for solving the Navier-Stokes problem: Find $\\text{S}^{*}$. If an\nanalytical expression for $\\text{S}^{*}$ can be obtained, it will lead via\ncanonical transformation to a new set of fields which are simply equal to their\ninitial values, giving analytical expressions for the original velocity and\npressure fields. Failing that, if one can only show that a complete solution to\nthis Hamilton-Jacobi equation does or does not exist, that will also resolve\nthe question of existence of solutions. The method employed here is not\nspecific to the Navier-Stokes problem or even to classical mechanics, and can\nbe applied to any traditionally non-Hamiltonian problem.", "category": "physics_flu-dyn" }, { "text": "Preferential imbibition in a dual-permeability pore network: A deep understanding of two-phase displacement in porous media with\npermeability contrast is essential for the design and optimisation of enhanced\noil recovery processes. In this paper, we investigate the forced imbibition\nbehaviour in two dual-permeability geometries that are of equal permeability\ncontrast. First, a mathematical model is developed for the imbibition in a pore\ndoublet, which shows that the imbibition dynamics can be fully described by the\nviscosity ratio $\\lambda$ and capillary number $Ca_m$ which creatively\nincorporates the influence of channel width and length. Through the finite\ndifference solution of the mathematical model, a $\\lambda-Ca_m$ phase diagram\nis established to characterise the imbibition preference in the pore doublet.\nWe then investigate the imbibition process in a dual-permeability pore network\nusing a well-established lattice Boltzmann method, focusing on the competition\nbetween the viscous and capillary forces. Like in the pore doublet, the\npreferential imbibition occurs in high permeability zone at high $Ca_{m}$ but\nin low permeability zone at low $Ca_{m}$. When $Ca_m$ is not sufficiently high,\nan oblique advancing pattern is observed which is attributed to non-trivial\ninterfacial tension. Thanks to the newly defined capillary number, the critical\n$Ca_{m}$ curve on which the breakthrough simultaneously occurs in both\npermeability zones, is found to match perfectly with that from the pore doublet\nand it is the optimal condition for maximising the imbibition efficiency in the\nentire pore network.", "category": "physics_flu-dyn" }, { "text": "Inertial waves and mean velocity profiles in a rotating pipe and a\n circular annulus with axial flow: In this paper we solve the inviscid inertial wave solutions in a circular\npipe or annulus rotating constantly about its axis with moderate angular speed.\nThe solutions are constructed by the so-called helical wave functions. We\nreveal that the mean velocity profiles must satisfy certain conditions to\naccommodate the inertial waves at the bulk region away from boundary. These\nconditions require the axial and azimuthal components of the mean velocity take\nthe shapes of the zeroth and first order Bessel functions of the first kind,\nrespectively. The theory is then verified by data obtained from direct\nnumerical simulations for both rotating pipe and circular annulus, and\nexcellent agreement is found between theory and numerical results. Large scale\nvortex clusters are found in the bulk region where the mean velocity profiles\nmatch the theoretical predictions. The success of the theory in rotating pipe,\ncircular annulus, and streamwise rotating channel suggests that such inertial\nwaves are quite common in wall bounded flow with background rotation.", "category": "physics_flu-dyn" }, { "text": "Transport due to Transient Progressive Waves: We describe and analyze the mean transport due to transient progressive\nwaves, including breaking waves. The waves are packets and are generated with a\nboundary-forced air-water two-phase Navier Stokes solver. The analysis is done\nin the Lagrangian frame. We show that the transport generated by these waves is\nsignificantly larger than the transport generated by steady waves. The\nnumerically generated parcel paths suggest a model for the transport that is\nprimarily driven by an irrotational approximation of the velocity. Wave\nbreaking is shown to increase the variability in the transport. Breaking is\naccounted for in the transport model via an additive stochastic velocity term.", "category": "physics_flu-dyn" }, { "text": "A unified stochastic particle method based on the Bhatnagar-Gross-Krook\n model for polyatomic gases and its combination with DSMC: Simulating hypersonic flow around a space vehicle is challenging because of\nthe multiscale and nonequilibrium nature inherent in these flows. To\neffectively deal with such flows, a novel particle particle hybrid scheme\ncombining the stochastic particle Bhatnagar-Gross-Krook (BGK) method with\nDirect Simulation Monte Carlo (DSMC) was developed recently, but only for\nmonatomic gases [Fei et. al., J. Comput. Phys. 2021]. Here this work is\nextended to the particle-particle hybrid method for polyatomic gases. In the\nnear continuum regime, employing the Ellipsoidal Statistical BGK model proposed\nby Y. Dauvois, et. al. [Eur. J. Mech. B Fluids, 2021] with discrete levels of\nvibrational energy, the stochastic particle BGK method is first established\nfollowing the idea of the unified stochastic particle BGK (USPBGK) scheme. In\nthe fluid limit, it has been proven to be of second order temporal and spatial\naccuracy. Then, the USPBGK scheme with rotational and vibrational energies is\ncombined with DSMC to construct a hybrid method for polyatomic gases. The\npresent hybrid scheme is validated with numerical tests of homogenous\nrelaxation, 1D shock structure and 2D hypersonic flows past a wedge and a\ncylinder. Compared to traditional stochastic particle methods, the proposed\nhybrid method can achieve higher accuracy at a much lower computational cost.\nTherefore, it is a more efficient tool to study multiscale hypersonic flows.", "category": "physics_flu-dyn" }, { "text": "Exploiting self-organized criticality in strongly stratified turbulence: A multiscale reduced description of turbulent free shear flows in the\npresence of strong stabilizing density stratification is derived via asymptotic\nanalysis of the Boussinesq equations in the simultaneous limits of small Froude\nand large Reynolds numbers. The analysis explicitly recognizes the occurrence\nof dynamics on disparate spatiotemporal scales, yielding simplified partial\ndifferential equations governing the coupled evolution of slow large-scale\nhydrostatic flows and fast small-scale isotropic instabilities and internal\nwaves. The dynamics captured by the coupled reduced equations is illustrated in\nthe context of two-dimensional strongly stratified Kolmogorov flow. A\nnoteworthy feature of the reduced model is that the fluctuations are\nconstrained to satisfy quasilinear (QL) dynamics about the comparably\nslowly-varying large-scale fields. Crucially, this QL reduction is not invoked\nas an ad hoc closure approximation, but rather is derived in a physically\nrelevant and mathematically consistent distinguished limit. Further analysis of\nthe resulting slow-fast QL system shows how the amplitude of the fast\nstratified-shear instabilities is slaved to the slowly-evolving mean fields to\nensure the marginal stability of the latter. Physically, this marginal\nstability condition appears to be compatible with recent evidence of\nself-organized criticality in both observations and simulations of stratified\nturbulence. lgorithmically, the slaving of the fluctuation fields enables\nnumerical simulations to be time-evolved strictly on the slow time scale of the\nhydrostatic flow. The reduced equations thus provide a solid mathematical\nfoundation for future studies of three-dimensional strongly stratified\nturbulence in extreme parameter regimes of geophysical relevance and suggest\navenues for new sub-grid-scale parameterizations.", "category": "physics_flu-dyn" }, { "text": "Linear regularized 13-moment equations with Onsager boundary conditions\n for general gas molecules: We develop the steady-state regularized 13-moment equations in the linear\nregime for rarefied gas dynamics with general collision models. For small\nKnudsen numbers, the model is accurate up to the super-Burnett order, and the\nresulting system of moment equations is shown to have a symmetric structure. We\nalso propose Onsager boundary conditions for the moment equations that\nguarantees the stability of the equations. The validity of our model is\nverified by benchmark examples for the one-dimensional channel flows.", "category": "physics_flu-dyn" }, { "text": "Collision of two breathers at surface of deep water: We applied canonical transformation to water wave equation not only to remove\ncubic nonlinear terms but to simplify drastically fourth order terms in\nHamiltonian. This transformation explicitly uses the fact of vanishing exact\nfour waves interaction for water gravity waves for 2D potential fluid. After\nthe transformation well-known but cumbersome Zakharov equation is drastically\nsimplified and can be written in X-space in compact way. This new equation is\nvery suitable as for analytic study as for numerical simulation. Localized in\nspace breather-type solution was found. Numerical simulation of collision of\ntwo such breathers strongly supports hypothesis of integrability of 2-D free\nsurface hydrodynamics.", "category": "physics_flu-dyn" }, { "text": "Rayleigh-Taylor turbulence with singular nonuniform initial conditions: We perform direct numerical simulations of three dimensional Rayleigh-Taylor\nturbulence with a nonuniform singular initial temperature background. In such\nconditions, the mixing layer evolves under the driving of a varying effective\nAtwood number; the long-time growth is still self-similar, but not anymore\nproportional to $t^2$ and depends on the singularity exponent $c$ of the\ninitial profile $\\Delta T \\propto z^c$. We show that the universality is\nrecovered when looking at the efficiency, defined as the ratio of the variation\nrates of the kinetic energy over the heat flux. A closure model is proposed\nthat is able to reproduce analytically the time evolution of the mean\ntemperature profiles, in excellent agreement with the numerical results.\nFinally, we reinterpret our findings in the light of spontaneous stochasticity\nwhere the growth of the mixing layer is mapped into the propagation of a wave\nof turbulent fluctuations on a rough background.", "category": "physics_flu-dyn" }, { "text": "Turbulence modulations and drag reduction by inertialess spheroids in\n turbulent channel flow: Previous studies on nonspherical particle-fluid interaction were mostly\nconfined to elongated fiber-like particles, which were observed to induce\nturbulence drag reduction. However, with the presence of tiny disk-like\nparticles how wall turbulence is modulated and whether drag reduction occurs\nare still unknown. Motivated by those open questions, we performed two-way\ncoupled direct numerical simulations of inertialess spheroids in turbulent\nchannel flow by an Eulerian-Lagrangian approach. The additional stress accounts\nfor the feedback from inertialess spheroids on the fluid phase. The results\ndemonstrate that both rigid elongated fibers (prolate spheroids) and thin disks\n(oblate spheroids) can lead to significant turbulence modulations and drag\nreduction. However, the disk-induced drag reduction is less pronounced than\nthat of rigid fibers with the same volume fraction. Typical features of\ndrag-reduced flows by additives are observed in both flow statistics and\nturbulence coherent structures. Moreover, in contrast to one-way simulations,\nthe two-way coupled results of spheroidal particles exhibit stronger\npreferential alignments and lower rotation rates. At the end we propose a drag\nreduction mechanism by inertialess spheroids and explain the different\nperformance for drag reduction by fibers and disks. We find that the spheroidal\nparticles weaken the quasistreamwise vortices through negative work and,\ntherefore, the Reynolds shear stress is reduced. However, the mean shear stress\ngenerated by particles, which is shape-dependent, partly compensates for the\nreduction of Reynolds shear stress and thus affects the efficiency of drag\nreduction. The present study implies that tiny disk-like particles can be an\nalternative drag reduction agent in wall turbulence.", "category": "physics_flu-dyn" }, { "text": "An oscillating motion of a red blood cell and a neutrally buoyant\n particle in Poiseuille flow in a narrow channel: Two motions of oscillation and vacillating breathing (swing) of a red blood\ncell have been observed in bounded Poiseuille flows (Phys. Rev. E 85, 16307\n(2012)). To understand such motions, we have studied the oscillating motion of\na neutrally buoyant rigid particle of the same shape in Poiseuille flow in a\nnarrow channel and obtained that the crucial point is to have the particle\ninteracting with Poiseuille flow with its mass center moving up and down in the\nchannel central region. Since the mass center of the cell migrates toward the\nchannel central region, its oscillating motion of the inclination angle is\nsimilar to the aforementioned motion as long as the cell keeps the shape of\nlong body. But as the up-and-down oscillation of the cell mass center damps\nout, the oscillating motion of the inclination angle also damps out and the\ncell inclination angle approaches to a fixed angle.", "category": "physics_flu-dyn" }, { "text": "Interplay of vortex rings and streaks in the near field of turbulent\n jets: In light of recent renewed interest in streaks in turbulent jets, the current\nwork explores their coexistence with vortex rings in the near field of\nturbulent round jets. A Reynolds number, $Re=1.5 \\times 10^5$ jet is studied at\ntwo diameters downstream of the nozzle, using high-speed stereo particle image\nvelocimetry. The spectra of individual velocity components reveal radially\nlocalized signatures of the large scale structures. The radial shapes of the\nspectral proper orthogonal decomposition modes corresponding to the signatures,\nconfirm the presence of high speed streaks in the outer edge of the jet at low\nStrouhal number, $St \\rightarrow 0$. The vortex rings and streamwise vortices\nare found to occur in the shear layer at $St \\approx 0.49$, where they convect\ntogether as a system and feed the streaks. The role of vortex rings in the\nexistence of streaks is then studied by strengthening the rings through\naxisymmetric excitation. The streaks are observed to persist, retaining their\nshapes and show only slight changes in their energies.", "category": "physics_flu-dyn" }, { "text": "Investigation of high-pressure turbulent jets using direct numerical\n simulation: Direct numerical simulations of free round jets at a Reynolds number\n($Re_{D}$) of $5000$, based on jet diameter ($D$) and jet-exit bulk velocity\n($U_{e}$), are performed to study jet turbulence characteristics at\nsupercritical pressures. The jet consists of $\\mathrm{N_{2}}$ that is injected\ninto $\\mathrm{N_{2}}$ at same temperature. To understand turbulent mixing, a\npassive scalar is transported with the flow at unity Schmidt number. Two sets\nof inflow conditions that model jets issuing from either a smooth contraction\nnozzle (laminar inflow) or a long pipe nozzle (turbulent inflow) are\nconsidered. By changing one parameter at a time, the simulations examine the\njet-flow sensitivity to the thermodynamic condition (characterized in terms of\nthe compressibility factor ($Z$) and the normalized isothermal\ncompressibility), inflow condition, and ambient pressure ($p_{\\infty}$)\nspanning perfect- to real-gas conditions. The inflow affects flow statistics in\nthe near-field (containing the potential core closure and the transition\nregion) as well as further downstream (containing fully-developed flow with\nself-similar statistics) at both atmospheric and supercritical $p_{\\infty}$.\nThe sensitivity to inflow is larger in the transition region, where the\nlaminar-inflow jets exhibit dominant coherent structures that produce higher\nmean strain rates and higher turbulent kinetic energy than in turbulent-inflow\njets. Decreasing $Z$ at a fixed supercritical $p_{\\infty}$ enhances pressure\nand density fluctuations (normalized by local mean pressure and density,\nrespectively), but the effect on velocity fluctuations depends also on local\nflow dynamics. When $Z$ is reduced, large mean strain rates in the transition\nregion of laminar-inflow jets significantly enhance velocity fluctuations\n(normalized by local mean velocity) and scalar mixing, whereas the effects are\nminimal in jets from turbulent inflow.", "category": "physics_flu-dyn" }, { "text": "The influence of porous media microstructure on filtration: We investigate how a filter media microstructure influences filtration\nperformance. We derive a theory that generalizes classical multiscale models\nfor regular structures to account for filter media with more realistic\nmicrostructures, comprising random microstructures with polydisperse\nunidirectional fibres. Our multiscale model accounts for the fluid flow and\ncontaminant transport at the microscale (over which the media structure is\nfully resolved) and allows us to obtain macroscopic properties such as the\neffective permeability, diffusivity, and fibre surface area. As the fibres grow\ndue to contaminant adsorption this leads to contact of neighbouring fibres. We\npropose an agglomeration algorithm that describes the resulting behaviour of\nthe fibres upon contact, allowing us to explore the subsequent time evolution\nof the filter media in a simple and robust way. We perform a comprehensive\ninvestigation of the influence of the filter-media microstructure on filter\nperformance in a spectrum of possible filtration scenarios.", "category": "physics_flu-dyn" }, { "text": "Importance of fluid inertia for the orientation of spheroids settling in\n turbulent flow: How non-spherical particles orient as they settle in a flow has important\npractical implications in a number of scientific and engineering problems. In a\nquiescent fluid, a slowly settling particle orients so that it settles with its\nbroad side first. This is an effect of the torque due to convective inertia of\nthe fluid set in motion by the settling particle, which maximises the drag\nexperienced by the particle. Turbulent flows tend to randomise the particle\norientation. Recently the settling of non-spherical particles in turbulence was\nanalysed neglecting the effect of convective fluid inertia, but taking into\naccount the effect of the turbulent fluid-velocity gradients on the particle\norientation. These studies reached the opposite conclusion, namely that a rod\nsettles preferentially with its tip first, wheras a disk settles with its edge\nfirst, therefore minimizing the drag on the particle. Here, we consider both\neffects, the convective inertial torque as well as the torque due to\nfluctuating velocity gradients, and ask under which circumstances either one or\nthe other dominate. To this end we estimate the ratio of the magnitudes of the\ntwo torques. Our estimates suggest that the fluid-inertia torque prevails in\nhigh-Reynolds number flows. In this case non-spherical particles are expected\nto settle with a maximal drag. But when the Reynolds number is small then the\ntorque due to fluid-velocity gradients may dominate, causing the particle to\nsettle with its broad side first.", "category": "physics_flu-dyn" }, { "text": "How micropatterns and air pressure affect splashing on surfaces: We experimentally investigate the splashing mechanism of a millimeter-sized\nethanol drop impinging on a structured solid surface, comprised of\nmicro-pillars, through side-view and top-view high speed imaging. By increasing\nthe impact velocity we can tune the impact outcome from a gentle deposition to\na violent splash, at which tiny droplets are emitted as the liquid sheet\nspreads laterally. We measure the splashing threshold for different\nmicropatterns and find that the arrangement of the pillars significantly\naffects the splashing outcome. In particular, directional splashing in\ndirection in which air flow through pattern is possible. Our top-view\nobservations of impact dynamics reveal that an trapped air is responsible for\nthe splashing. Indeed by lowering the pressure of the surrounding air we show\nthat we can suppress the splashing in the explored parameter regime.", "category": "physics_flu-dyn" }, { "text": "Non-Newtonian fluid-structure interaction: Flow of a viscoelastic\n Oldroyd-B fluid in a deformable channel: We analyze the steady non-Newtonian fluid-structure interaction between the\nflow of an Oldroyd-B fluid and a deformable channel. Specifically, we provide a\ntheoretical framework for calculating the leading-order effect of the fluid's\nviscoelasticity on the flow rate-pressure drop relation and on the deformation\nof the channel's elastic wall. We first identify the characteristic scales and\ndimensionless parameters governing the fluid-structure interaction in slender\nand shallow channels. Applying the lubrication approximation for the flow and\nemploying a perturbation expansion in powers of the Deborah number $De$, we\nderive a closed-form expression for the pressure as a function of the\nnon-uniform shape of the channel in the weakly viscoelastic limit up to\n$\\mathrm{O}(De)$. Coupling the hydrodynamic pressure to the elastic\ndeformation, we provide the leading-order effect of the interplay between the\nviscoelasticity of the fluid and the compliance of the channel on the pressure\nand deformation fields, as well as on the flow rate-pressure drop relation. For\nthe flow-rate-controlled regime and in the weakly viscoelastic limit, we show\nanalytically that both the compliance of the deforming top wall and the\nviscoelasticity of the fluid decrease the pressure drop. Furthermore, we reveal\na trade-off between the influence of compliance of the channel and the fluid's\nviscoelasticity on the deformation. While the channel's compliance increases\nthe deformation, the fluid's viscoelasticity decreases it.", "category": "physics_flu-dyn" }, { "text": "Hybrid time series from PIV for characterization of turbulent flow\n fields: ASTRA -- Approach using Spatially and Temporally Resolved Advection: Particle Image Velocimetry (PIV) has become increasingly popular to study\nstructures in turbulent flows. PIV allows direct extraction and investigation\nof spatial structures in the given flow field. Increasing temporal resolution\nof PIV systems allows a more accurate capture of the flow evolution. Despite\nthe very good spatial resolution of PIV, current systems can only match the\nmultiple $kHz$ sampling rates of hot-wire or Laser Doppler Anemometer (LDA)\nmeasurements for a very short period in temporal analyses of flow. In this\nstudy, an advection-based approach is presented which uses Taylor's hypothesis\nof \"frozen turbulence\" for small scale turbulent patterns. Compared to the\nunderlying raw data a major increase of the temporal resolution for extracted\ntime series is shown. The quality of the presented approach is shown for\ntwo-point analyses, which would not be possible with presently known methods.\nTo demonstrate this, different turbulent flow cases behind a fractal grid are\nstudied. For the validation of the results corresponding hot-wire measurements\nat various positions along the centerline were used.", "category": "physics_flu-dyn" }, { "text": "RONAALP: Reduced-Order Nonlinear Approximation with Active Learning\n Procedure: Many engineering applications rely on the evaluation of expensive, non-linear\nhigh-dimensional functions. In this paper, we propose the RONAALP algorithm\n(Reduced Order Nonlinear Approximation with Active Learning Procedure) to\nincrementally learn a fast and accurate reduced-order surrogate model of a\ntarget function on-the-fly as the application progresses. First, the\ncombination of nonlinear auto-encoder, community clustering and radial basis\nfunction networks allows to learn an efficient and compact surrogate model with\nlimited training data. Secondly, the active learning procedure overcome any\nextrapolation issue when evaluating the surrogate model outside of its initial\ntraining range during the online stage. This results in generalizable, fast and\naccurate reduced-order models of high-dimensional functions. The method is\ndemonstrated on three direct numerical simulations of hypersonic flows in\nchemical nonequilibrium. Accurate simulations of these flows rely on detailed\nthermochemical gas models that dramatically increase the cost of such\ncalculations. Using RONAALP to learn a reduced-order thermodynamic model\nsurrogate on-the-fly, the cost of such simulation was reduced by up to 75%\nwhile maintaining an error of less than 10% on relevant quantities of interest.", "category": "physics_flu-dyn" }, { "text": "Effect of Patterned Slip on Micro and Nanofluidic Flows: We consider the flow of a Newtonian fluid in a nano or microchannel with\nwalls that have patterned variations in slip length. We formulate a set of\nequations to describe the effects on an incompressible Newtonian flow of small\nvariations in slip, and solve these equations for slow flows. We test these\nequations using molecular dynamics simulations of flow between two walls which\nhave patterned variations in wettability. Good qualitative agreement and a\nreasonable degree of quantitative agreement is found between the theory and the\nmolecular dynamics simulations. The results of both analyses show that\npatterned wettability can be used to induce complex variations in flow. Finally\nwe discuss the implications of our results for the design of microfluidic\nmixers using slip.", "category": "physics_flu-dyn" }, { "text": "Coalescence of elastic blisters filled with a viscous fluid: Pockets of viscous fluid coalescing beneath an elastic plate are encountered\nin a wide range of natural phenomena and engineering processes, spanning across\nscales. As the pockets merge, a bridge is formed with a height increasing as\nthe plate relaxes. We study the spatiotemporal dynamics of such an\nelasto-hydrodynamic coalescence process by combining experiments, lubrication\ntheory and numerical simulations. The bridge height exhibits an exponential\ngrowth with time, which corresponds to a self-similar solution of the\nbending-driven thin-film equation. We address this unique self-similarity and\nthe self-similar shape of the bridge, both of which are corroborated in\nnumerical simulations and experiments.", "category": "physics_flu-dyn" }, { "text": "Wave turbulence and intermittency in directional sea states: The evolution of surface gravity waves is driven by nonlinear interactions\nthat trigger an energy cascade similarly to the one observed in hydrodynamic\nturbulence. This process, known as wave turbulence, has been found to display\nanomalous scaling with deviation from classical turbulent predictions due to\nthe emergence of coherent and intermittent structures on the water surface. In\nrealistic oceanic sea states, waves are spread over a wide range of directions,\nwith a consequent attenuation of the nonlinear properties. A laboratory\nexperiment in a large wave facility is presented to discuss the effect of wave\ndirectionality on wave turbulence. Results show that the occurrence of coherent\nand intermitted structures become less likely with the broadening of the wave\ndirectional spreading. There is no evidence, however, that intermittency\ncompletely vanishes.", "category": "physics_flu-dyn" }, { "text": "Turbulent spot growth in plane Couette flow: statistical study and\n formation of spanwise vorticity: This article presents direct numerical simulations of the growth of turbulent\nspots in the transitional regime of plane Couette flow. A quantitative\ndescription of the growth process and of the detail of the quadrupolar flow\naround the spot is given. Focus is put on formation of spanwise vorticity in\nthe velocity streaks that resembles a secondary shear instability. The main\nfeatures of the instability (coherence lengths, advection velocity) are studied\nin the context of the turbulent spot, below and above the threshold Reynolds\nnumber of appearance of the oblique turbulent bands of plane Couette flow.", "category": "physics_flu-dyn" }, { "text": "Transient inverse energy cascade in free surface turbulence: We study the statistics of free-surface turbulence at large Reynolds numbers\nproduced by direct numerical simulations in a fluid layer at different\nthickness with fixed characteristic forcing scale. We observe the production of\na transient inverse cascade, with a duration which depends on the thickness of\nthe layer, followed by a transition to three-dimensional turbulence initially\nproduced close to the bottom, no-slip boundary. By switching off the forcing,\nwe study the decaying turbulent regime and we find that it cannot be described\nby an exponential law. Our results show that boundary conditions play a\nfundamental role in the nature of turbulence produced in thin layers and give\nlimits on the conditions to produce a two-dimensional phenomenology.", "category": "physics_flu-dyn" }, { "text": "Completeness of the Lattice-Boltzmann IKT approach for classical\n incompressible fluids: Despite the abundant literature on the subject appeared in the last few\nyears, the lattice Boltzmann method (LBM) is probably the one for which a\ncomplete understanding is not yet available. As an example, an unsolved\ntheoretical issue is related to the construction of a discrete kinetic theory\nwhich yields \\textit{exactly} the fluid equations, i.e., is non-asymptotic\n(here denoted as \\textit{LB inverse kinetic theory}). The purpose of this paper\naims at investigating discrete inverse kinetic theories (IKT) for\nincompressible fluids. We intend to show that the discrete IKT can be defined\nin such a way to satisfy, in particular, the requirement of\n\\emph{completeness}, i.e., {\\it all} fluid fields are expressed as moments of\nthe kinetic distribution function and {\\it all} hydrodynamic equations can be\nidentified with suitable moment equations of an appropriate inverse kinetic\nequation IKE.", "category": "physics_flu-dyn" }, { "text": "Fluid flow analysis in a rough fracture (type II) using complex networks\n and lattice Boltzmann method: Complexity of fluid flow in a rough fracture is induced by the complex\nconfigurations of opening areas between the fracture planes. In this study, we\nmodel fluid flow in an evolvable real rock joint structure, which under certain\nnormal load is sheared. In an experimental study, information regarding about\napertures of the rock joint during consecutive 20 mm displacements and fluid\nflow (permeability) in different pressure heads have been recorded by a scanner\nlaser. Our aim in this study is to simulate the fluid flow in the mentioned\ncomplex geometries using the lattice Boltzmann method (LBM), while the\ncharacteristics of the aperture field will be compared with the modeled fluid\nflow permeability To characterize the aperture, we use a new concept in the\ngraph theory, namely: complex networks and motif analysis of the corresponding\nnetworks. In this approach, the similar aperture profile along the fluid flow\ndirection is mapped in to a network space. The modeled permeability using the\nLBM shows good correlation with the experimental measured values. Furthermore,\nthe two main characters of the obtained networks, i.e., characteristic length\nand number of edges show the same evolutionary trend with the modeled\npermeability values. Analysis of motifs through the obtained networks showed\nthe most transient sub-graphs are much more frequent in residual stages. This\ncoincides with nearly stable fluid flow and high permeability values.", "category": "physics_flu-dyn" }, { "text": "A Control Forced Concurrent Precursor Method for LES Inflow: With the increased application of large eddy simulation techniques, the\ngeneration of realistic turbulence at inflow boundaries is crucial for the\naccuracy of a simulation. The Control Forced Concurrent Precursor Method\n(CFCPM) proposed in this work combines an existing concurrent precursor method\nand a mean flow forcing method with a new extension of the controlled forcing\nmethod to impose turbulent inflow boundary conditions primarily, although not\nexclusively, for domains that require periodic boundary conditions. Turbulent\ninflow boundary conditions are imposed through a region of body forces added to\nthe momentum equations of the main simulation that transfers the precursor\nsimulation into the main domain. Controlled forcing planes, which come into\nplay as body forces added to the momentum equations on planes perpendicular to\nthe flow, located in the precursor simulation, allow for specific Reynolds\nstress tensors and mean velocities to be imposed. The mean flow controlled\nforcing method only modifies the mean velocity profiles, leaving the\nfluctuating velocity field untouched. The proposed fluctuating flow controlled\nforcing methods extends the application of the original controlled forcing\nmethod to multiple fluctuating velocity components and couples their\ncalculation in order to amplify the existing fluctuations present in the\nprecursor flow field so that prescribed anisotropic Reynolds stress tensors can\nbe reproduced. The new method was tested on high Reynolds number turbulent\nboundary layer flow over a wall-mounted cube and low Reynolds number turbulent\nboundary layer flow over a backward-facing step. It was found that the new\nextension of the controlled forcing method reduced the development time for\nboth test cases considered here when compared to not using controlled forcing\nand only using the original controlled forcing method.", "category": "physics_flu-dyn" }, { "text": "Surface tension profiles in vertical soap films: Surface tension profiles in vertical soap films are experimentally\ninvestigated. Measurements are performed introducing deformable elastic objets\nin the films. The shape adopted by those objects set in the film can be related\nto the surface tension value at a given vertical position by numerical solving\nof adapted elasticity equations. We show that the observed dependency of the\nsurface tension versus the vertical position in the soap film can be reproduced\nby simple modeling taking into account film thickness measurements.", "category": "physics_flu-dyn" }, { "text": "Bridging Polymeric Turbulence at different Reynolds numbers: From\n Multiscaling to Multifractality: The addition of polymers modifies a flow in a non-trivial way that depends on\nfluid inertia (given by the Reynolds number Re) and polymer elasticity\n(quantified by the Deborah number De). Using direct numerical simulations, we\nshow that polymeric flows exhibit a Re and De dependent multiscaling energy\nspectrum. The different scaling regimes are tied to various dominant\ncontributions -- fluid, polymer, and dissipation -- to the total energy flux\nacross the scales. At small scales, energy is dissipated away by both polymers\nand the fluid. Fluid energy dissipation, in particular, is shown to be more\nintermittent in the presence of polymers, especially at small Re. The more\nintermittent, singular nature of energy dissipation is revealed clearly by the\nmultifractal spectrum.", "category": "physics_flu-dyn" }, { "text": "Turbulent drag in a rotating frame: What is the turbulent drag force experienced by an object moving in a\nrotating fluid? This open and fundamental question can be addressed by\nmeasuring the torque needed to drive an impeller at constant angular velocity\n$\\omega$ in a water tank mounted on a platform rotating at a rate $\\Omega$. We\nreport a dramatic reduction in drag as $\\Omega$ increases, down to values as\nlow as $12$\\% of the non-rotating drag. At small Rossby number $Ro =\n\\omega/\\Omega$, the decrease in drag coefficient $K$ follows the approximate\nscaling law $K \\sim Ro$, which is predicted in the framework of nonlinear\ninertial wave interactions and weak-turbulence theory. However, stereoscopic\nparticle image velocimetry measurements indicate that this drag reduction\nrather originates from a weakening of the turbulence intensity in line with the\ntwo-dimensionalization of the large-scale flow.", "category": "physics_flu-dyn" }, { "text": "Investigation of transition frequencies of two acoustically coupled\n bubbles using a direct numerical simulation technique: The theoretical results regarding the ``transition frequencies'' of two\nacoustically interacting bubbles have been verified numerically. The theory\nprovided by Ida [Phys. Lett. A 297 (2002) 210] predicted the existence of three\ntransition frequencies per bubble, each of which has the phase difference of\n$\\pi /2$ between a bubble's pulsation and the external sound field, while\nprevious theories predicted only two natural frequencies which cause such phase\nshifts. Namely, two of the three transition frequencies correspond to the\nnatural frequencies, while the remaining does not. In a subsequent paper [M.\nIda, Phys. Rev. E 67 (2003) 056617], it was shown theoretically that transition\nfrequencies other than the natural frequencies may cause the sign reversal of\nthe secondary Bjerknes force acting between pulsating bubbles. In the present\nstudy, we employ a direct numerical simulation technique that uses the\ncompressible Navier-Stokes equations with a surface-tension term as the\ngoverning equations to investigate the transition frequencies of two coupled\nbubbles by observing their pulsation amplitudes and directions of translational\nmotion, both of which change as the driving frequency changes. The numerical\nresults reproduce the recent theoretical predictions, validating the existence\nof the transition frequencies not corresponding to the natural frequency.", "category": "physics_flu-dyn" }, { "text": "Optimizing the method of images for regularized Stokeslets using theory\n and experiments of spheres moving near boundaries: The general system of images for regularized Stokeslets (GSIRS) developed by\nCortez and Varela (2015) is used extensively to model Stokes flow phenomena\nsuch as microorganisms swimming near a boundary. Our collaborative team uses\ndynamically similar scaled macroscopic experiments to test theories for forces\nand torques on spheres moving near a boundary and use these data and the method\nof regularized Stokeslets (MRS) created by Cortez et al. (2015) to calibrate\nthe GSIRS. We find excellent agreement between theory and experiments, which\nprovides the first experimental validation of such series solutions for spheres\nmoving near an infinite plane boundary. We test two surface discretization\nmethods commonly used in the literature: the six-patch method and the spherical\ncentroidal Voronoi tessellation (SCVT) method. Our data show that the SCVT\nmethod provides the most accurate results when the motional symmetry is broken\nby the presence of a boundary. We use theory and the MRS to find optimal values\nfor the regularization parameter in free space and show that the optimal\nregularization parameter values can be fit with simple formulae when using the\nSCVT method, so that other researchers have an easy reference. We also present\na regularization function with higher order accuracy when compared with the\nregularization function previously introduced by Cortez et al. (2005). The\nsimulated force and torque values compare very well with experiments and theory\nfor a wide range of boundary distances. But, we find the simulations lose\naccuracy when the gap between the edge of the sphere and the wall is smaller\nthan the average distance between grid points using the SCVT method. Our\ncomputational parameters and MATLAB and PYTHON implementations of the Lee and\nLeal (1980) theory provide researchers with important resources to optimize the\nnumerical simulations of spheres moving near boundaries.", "category": "physics_flu-dyn" }, { "text": "Effects of polymers on the cavitating flow around a cylinder: A\n Large-scale molecular dynamics analysis: The cavitation flow of linear-polymer solutions around a cylinder is studied\nby performing a large-scale molecular dynamics simulation. The addition of\npolymer chains remarkably suppresses the cavitation. The polymers are stretched\ninto a linear shape near the cylinder and entrained in the vortex behind the\ncylinder. As the polymers stretch, the elongational viscosity increases, which\nsuppresses the vortex formation. Furthermore, the polymers exhibit an entropic\nelasticity owing to the stretching. This elastic energy increases the local\ntemperature, which inhibits the cavitation inception. These effects of polymers\nresult in the dramatic suppression of cavitation.", "category": "physics_flu-dyn" }, { "text": "Comment on Conjugate heat transfer of mixed convection for viscoelastic\n fluid past a stretching sheet, by Hsiao and Chen, Mathematical Problems in\n Engineering: Comment on Conjugate heat transfer of mixed convection for viscoelastic fluid\npast a stretching sheet, by Kai-Long Hsiao and Guan-Bang Chen, Mathematical\nProblems in Engineering, Volume 2007, article 17058, 21 pages", "category": "physics_flu-dyn" }, { "text": "Aperiodic tumbling of microrods advected in a microchannel flow: We report on an experimental investigation of the tumbling of microrods in\nthe shear flow of a microchannel (40 x 2.5 x 0.4 mm). The rods are 20 to 30\nmicrons long and their diameters are of the order of 1 micron. Images of the\ncentre-of-mass motion and the orientational dynamics of the rods are recorded\nusing a microscope equipped with a CCD camera. A motorised microscope stage is\nused to track individual rods as they move along the channel. Automated image\nanalysis determines the position and orientation of a tracked rods in each\nvideo frame. We find different behaviours, depending on the particle shape, its\ninitial position, and orientation. First, we observe periodic as well as\naperiodic tumbling. Second, the data show that different tumbling trajectories\nexhibit different sensitivities to external perturbations. These observations\ncan be explained by slight asymmetries of the rods. Third we observe that after\nsome time, initially periodic trajectories lose their phase. We attribute this\nto drift of the centre of mass of the rod from one to another stream line of\nthe channel flow.", "category": "physics_flu-dyn" }, { "text": "Microscopic laws vs. Macroscopic laws: Perspectives from kinetic theory\n and hydrodynamics: Reductionism is a prevalent viewpoint in science according to which all\nphysical phenomena can be understood from fundamental laws of physics. Anderson\n[Science, 177, 393 (1972)], Laughlin and Pines [PNAS, 97, 28 (2000)], and\nothers have countered this viewpoint and argued in favour hierarchical\nstructure of the universe and laws. In this paper we advance the latter\nperspective by showing that some of the complex flow properties derived using\nhydrodynamic equations (macroscopic laws) are very difficult, if not\nimpossible, to describe in microscopic framework---kinetic theory. These\nproperties include Kolmogorov's theory of turbulence, turbulence dissipation\nand diffusion, and dynamic pressure. We also provide several other examples of\nhierarchical description.", "category": "physics_flu-dyn" }, { "text": "The DRESDYN project: planned experiments and present status: The Dresden sodium facility for dynamo and thermohydraulic studies (DRESDYN)\nis a platform for large-scale liquid sodium experiments devoted to fundamental\ngeo- and astrophysical questions as well as to various applied problems related\nto the conversion and storage of energy. Its most ambitious part is a\nprecession driven dynamo experiment comprising 8 tons of liquid sodium supposed\nto rotate with up to 10 Hz and to precess with up to 1 Hz. Another large-scale\nset-up is a Tayler-Couette experiment with a gap width of 0.2 m and a height of\n2 m, whose inner cylinder rotates with up to 20 Hz. Equipped with a coil system\nfor the generation of an axial field of up to 120 mT and two different axial\ncurrents through the center and the liquid sodium, this experiment aims at\nstudying various versions of the magnetorotational instability and their\ncombinations with the Tayler instability. We discuss the physical background of\nthese two experiments and delineate the present status of their technical\nrealization. Other installations, such as a sodium loop and a test stand for\nIn-Service-Inspection experiments will also be sketched.", "category": "physics_flu-dyn" }, { "text": "NEMD modeling of nanoscale hydrodynamics of clay-water system at\n elevated temperature: The engineering problems involving clay under non-isothermal conditions\n(e.g., geothermal energy harvest, landfill cover system, and nuclear waste\ndisposal) are multiscale and multiphysics by nature. The nanoscale\nhydrodynamics of clay at elevated temperature is essential in developing a\nphysics-based multiscale model for clay under non-isothermal conditions. The\nnonequilibrium molecular dynamics (NEMD) is a useful tool to study the\nnanoscale hydrodyndamics of clay. This article presents an NEMD modeling of\nhydrodynamics of clay nanopores at elevated temperatures. Water flow confined\nin pyrophyllite and montmorillonite clay nanopores is investigated. The\nnonequilibrium state is maintained by uniformly exerting an external force on\neach water molecule. The NEMD simulations have provided a molecular-scale\nperspective of temperature effect on clay-water density, water flow velocity,\nshear viscosity, clay-water slip length, hydraulic conductivity, and clay-water\nfriction coefficient. The numerical results have shown a strong temperature\ndependence of fluid flow velocity, shear viscosity, clay-water slip length, and\nhydraulic conductivity at the nanoscale. We have validated the applicability of\ncubic law in determining hydraulic conductivity at the nanopore scale under at\nelevated temperature. It is found from our numerical results that slip\nclay-water boundary condition is an essential factor in properly determining\nnanoscale fluid flow velocity. By numerical examples, we also study the impact\nof nanopore size and clay layer thickness on the hydrodynamics of the\nclay-water system.", "category": "physics_flu-dyn" }, { "text": "Simultaneous numerical simulation of direct and inverse cascades in wave\n turbulence: Results of direct numerical simulation of isotropic turbulence of surface\ngravity waves in the framework of Hamiltonian equations are presented. For the\nfirst time simultaneous formation of both direct and inverse cascades was\nobserved in the framework of primordial dynamical equations. At the same time,\nstrong long waves background was developed. It was shown, that obtained\nKolmogorov spectra are very sensitive to the presence of this condensate. Such\nsituation has to be typical for experimental wave tanks, flumes, and small\nlakes.", "category": "physics_flu-dyn" }, { "text": "Surface gravity waves propagating in a rotating frame: the Ekman-Stokes\n instability: We report on an instability arising when surface gravity waves propagate in a\nrotating frame. The Stokes drift associated to the uniform wave field, together\nwith global rotation, drives a mean flow in the form of a horizontally\ninvariant Ekman-Stokes spiral. We show that the latter can be subject to an\ninstability that triggers the appearance of an additional\nhorizontally-structured cellular flow. We determine the instability threshold\nnumerically, in terms of the Rossby number Ro associated to the Stokes drift of\nthe waves and the Ekman number E. We confirm the numerical results through\nasymptotic expansions at both large and low Ekman number. At large E the\ninstability reduces to that of a standard Ekman spiral driven by the\nwave-induced surface stress instead of a wind stress, while at low E the\nStokes-drift profile crucially determines the shape of the unstable mode. In\nboth limits the instability threshold asymptotes to an Ekman-number-independent\ncritical Rossby number, which in both cases also corresponds to a critical\nReynolds number associated to the Lagrangian base-flow velocity profile.\nParameter values typical of ocean swell fall into the low-E unstable regime:\nthe corresponding \"anti-Stokes\" flows are unstable, with possible consequences\nfor particle dispersion and mixing.", "category": "physics_flu-dyn" }, { "text": "Thermodynamically consistent description of the hydrodynamics of free\n surfaces covered by insoluble surfactants of high concentration: In this paper we propose several models that describe the dynamics of liquid\nfilms which are covered by a high concentration layer of insoluble surfactant.\nFirst, we briefly review the 'classical' hydrodynamic form of the coupled\nevolution equations for the film height and surfactant concentration that are\nwell established for small concentrations. Then we re-formulate the basic model\nas a gradient dynamics based on an underlying free energy functional that\naccounts for wettability and capillarity. Based on this re-formulation in the\nframework of nonequilibrium thermodynamics, we propose extensions of the basic\nhydrodynamic model that account for (i) nonlinear equations of state, (ii)\nsurfactant-dependent wettability, (iii) surfactant phase transitions, and (iv)\nsubstrate-mediated condensation. In passing, we discuss important differences\nto most of the models found in the literature.", "category": "physics_flu-dyn" }, { "text": "Velocity of viscous fingers in miscible displacement: The paper investigates the linear growth of the mixing zone during polymer\nslug injection into a water reservoir. The velocities of the slug front and of\nthe boundaries of the mixing zone are analyzed as key parameters. Using two\ndifferent numerical methods (finite volumes and finite elements), the impact of\nthe slug size, reservoir dimensions, Peclet number, and viscosity curve shape\non the corresponding velocities is examined. Notwithstanding the realization of\nthe solution by two computational schemes, the simulation results coincide with\nsufficient accuracy. The numerically obtained velocities are compared with\ntheoretical estimates within the transverse flow equilibrium approximation and\nKoval model. Based on the comparison pattern, recommendations are presented on\nthe use of specific analytical methods for estimating the growth rate of the\nmixing zone depending on the characteristics of the polymer.", "category": "physics_flu-dyn" }, { "text": "Data-driven unsteady aeroelastic modeling for control: Aeroelastic structures, from insect wings to wind turbine blades, experience\ntransient unsteady aerodynamic loads that are coupled to their motion.\nEffective real-time control of flexible structures relies on accurate and\nefficient predictions of both the unsteady aeroelastic forces and airfoil\ndeformation. For rigid wings, classical unsteady aerodynamic models have\nrecently been reformulated in state-space for control and extended to include\nviscous effects. Here we further extend this modeling framework to include the\ndeformation of a flexible wing in addition to the quasi-steady, added mass, and\nunsteady viscous forces. We develop low-order linear models based on data from\ndirect numerical simulations of flow past a flexible wing at low Reynolds\nnumber. We demonstrate the effectiveness of these models to track aggressive\nmaneuvers with model predictive control while constraining maximum wing\ndeformation. This system identification approach provides an interpretable,\naccurate, and low-dimensional representation of an aeroelastic system that can\naid in system and controller design for applications where transients play an\nimportant role.", "category": "physics_flu-dyn" }, { "text": "A Unified Strouhal-Reynolds Number Relationship for Laminar Vortex\n Streets Generated by Different Shaped Obstacles: A new Strouhal-Reynolds number relationship, $St=1/(A+B/Re)$, has been\nrecently proposed based on observations of laminar vortex shedding from\ncircular cylinders in a flowing soap film. Since the new $St$-$Re$ relation was\nderived from a general physical consideration, it raises the possibility that\nit may be applicable to vortex shedding from bodies other than circular ones.\nThe work presented herein provides experimental evidence that this is the case.\nOur measurements also show that in the asymptotic limit\n($Re\\rightarrow\\infty$), $St_{\\infty}=1/A\\simeq0.21$ is constant independent of\nrod shapes, leaving $B$ the only parameter that is shape dependent.", "category": "physics_flu-dyn" }, { "text": "Symmetry Breaking in Induced-Charge Electrophoresis: The electrophoretic motion of a conducting particle, driven by an induced\ncharge mechanism, is analyzed. The dependence of the motion upon particle shape\nis embodied in four tensorial coefficients that relate the particle velocities\nto the externally-applied electric field. Several families of particle shapes,\nwhose members are unaffected by the applied field, are identified via use of\nsymmetry arguments. Other particles translate and/or rotate in response to the\nimposed field, even if their net electric charge vanishes. The coefficients are\nrepresented as surface integrals of the electric potential over the particle\nboundary, thereby eliminating the need to solve the flow field.", "category": "physics_flu-dyn" }, { "text": "Uptake of water droplets by nonwetting capillaries: We present direct experimental evidence that water droplets can spontaneously\npenetrate non-wetting capillaries, driven by the action of Laplace pressure due\nto high droplet curvature. Using high-speed optical imaging, microcapillaries\nof radius 50 to 150 micron, and water microdroplets of average radius between\n100 and 1900 micron, we demonstrate that there is a critical droplet radius\nbelow which water droplets can be taken up by hydrophobised glass and\npolytetrafluoroethylene (PTFE) capillaries. The rate of capillary uptake is\nshown to depend strongly on droplet size, with smaller droplets being absorbed\nmore quickly. Droplet size is also shown to influence meniscus motion in a\npre-filled non-wetting capillary, and quantitative measurements of this effect\nresult in a derived water-PTFE static contact angle between 96 degrees and 114\ndegrees. Our measurements confirm recent theoretical predictions and\nsimulations for metal nanodroplets penetrating carbon nanotubes (CNTs). The\nresults are relevant to a wide range of technological applications, such as\nmicrofluidic devices, ink-jet printing, and the penetration of fluids in porous\nmaterials.", "category": "physics_flu-dyn" }, { "text": "Homogenized scattering model of water wave attenuation in marginal ice\n zone: A theoretical model to explain the scattering process of wave attenuation in\na marginal ice zone is developed. Many field observations offer wave energy\ndecay in the form of exponential function with distance, and this is justified\nthrough the complex wave number for the dissipation process. On the other hand,\nsuch a mechanism is not explicitly proven for the scattering process. To\nexplain this, we consider a periodic array of ice floes, where the floe is\nmodeled by a vertical rigid cylinder. Using a homogenization technique, a\nhomogenized free surface equivalent to the array is obtained. Then, we show\nthat a dispersion relation of the homogenized free surface waves makes all wave\nnumbers complex. As a result, the exponential energy decay in the scattering\nprocess is demonstrated. Although our model is obtained using many\nsimplifications, it reproduces consistent tendencies with both existing field\nobservations and numerical simulations; the wave attenuation coefficient for\nthe deep sea is proportional to the ice concentration and the wave number for\nopen water waves, and the coefficient is bigger as the radius and draft of the\nfloe become larger or the wave period is smaller.", "category": "physics_flu-dyn" }, { "text": "Extreme statistics and extreme events in dynamical models of turbulence: We present a study of the intermittent properties of a shell model of\nturbulence with unprecedented statistics, about $\\sim 10^7$ eddy turn over\ntime, achieved thanks to an implementation on a large-scale parallel GPU\nfactory. This allows us to quantify the inertial range anomalous scaling\nproperties of the velocity fluctuations up to the $24$-th order moment. Through\na careful assessment of the statistical and systematic uncertainties, we show\nthat none of the phenomenological and theoretical models previously proposed in\nthe literature to predict the anomalous power-law exponents in the inertial\nrange is in agreement with our high-precision numerical measurements. We find\nthat at asymptotically large moments, the anomalous exponents tend towards a\nlinear scaling, suggesting that extreme turbulent events are dominated by one\nleading singularity. We found that systematic corrections to scaling induced by\nthe infrared and ultraviolet (viscous) cut-offs are the main limitations to\nprecision for low-order moments, while large orders are mainly affected by the\nfinite statistical samples. The unprecedentedly high fidelity numerical results\nreported in this work offer an ideal benchmark for the development of future\ntheoretical models of intermittency for either extreme events (high-order\nmoments) or typical fluctuations (low-order moments). For the latter, we show\nthat we achieve a precision in the determination of the inertial range scaling\nproperties of the order of one part over ten thousand (5th significant digit),\nwhich must be considered a record for out-of-equilibrium fluid-mechanics\nsystems and models.", "category": "physics_flu-dyn" }, { "text": "Lattice Boltzmann Simulations of Droplet formation in confined Channels\n with Thermocapillary flows: Based on mesoscale lattice Boltzmann simulations with the \"Shan-Chen\" model,\nwe explore the influence of thermocapillarity on the break-up properties of\nfluid threads in a microfluidic T-junction, where a dispersed phase is injected\nperpendicularly into a main channel containing a continuous phase, and the\nlatter induces periodic break-up of droplets due to the cross-flowing.\nTemperature effects are investigated by switching on/off both positive/negative\ntemperature gradients along the main channel direction, thus promoting a\ndifferent thread dynamics with anticipated/delayed break-up. Numerical\nsimulations are performed at changing the flow-rates of both the continuous and\ndispersed phases, as well as the relative importance of viscous forces, surface\ntension forces and thermocapillary stresses. The range of parameters is broad\nenough to characterize the effects of thermocapillarity on different mechanisms\nof break-up in the confined T-junction, including the so-called \"squeezing\" and\n\"dripping\" regimes, previously identified in the literature. Some simple\nscaling arguments are proposed to rationalize the observed behaviour, and to\nprovide quantitative guidelines on how to predict the droplet size after\nbreak-up.", "category": "physics_flu-dyn" }, { "text": "Phase Slip Solutions in Magnetically Modulated Taylor-Couette Flow: We numerically investigate Taylor-Couette flow in a wide-gap configuration,\nwith $r_i/r_o=1/2$, the inner cylinder rotating, and the outer cylinder\nstationary. The fluid is taken to be electrically conducting, and a magnetic\nfield of the form $B_z\\approx(1 + \\cos(2\\pi z/z_0))/2$ is externally imposed,\nwhere the wavelength $z_0=50(r_o-r_i)$. Taylor vortices form where the field is\nweak, but not where it is strong. As the Reynolds number measuring the rotation\nrate is increased, the initial onset of vortices involves phase slip events,\nwhereby pairs of Taylor vortices are periodically formed and then drift\noutward, away from the midplane where $B_z=0$. Subsequent bifurcations lead to\na variety of other solutions, including ones both symmetric and asymmetric\nabout the midplane. For even larger Reynolds numbers a different type of phase\nslip arises, in which vortices form at the outer edges of the pattern and drift\ninward, disappearing abruptly at a certain point. These solutions can also be\nsymmetric or asymmetric about the midplane, and co-exist at the same Reynolds\nnumber. Many of the dynamics of these phase slip solutions are qualitatively\nsimilar to previous results in geometrically ramped Taylor-Couette flows.", "category": "physics_flu-dyn" }, { "text": "Levitating Drop in a Tilted Rotating Tank - Gallery of Fluid Motion\n Entry V044: A cylindrical acrylic tank with inner diameter D = 4 in. is mounted such that\nits axis of symmetry is at some angle measured from the vertical plane. The\nmixing tank is identical to that described in [1] The tank is filled with 200\nmL of 1000 cSt silicone oil and a 5 mL drop of de-ionized water is placed in\nthe oil volume. The water drop is allowed to come to rest and then a motor\nrotates the tank about its axis of symmetry at a fixed frequency = 0.3 Hz.\nTherefore the Reynolds number is fixed at about Re ~ 5 yielding laminar flow\nconditions. A CCD camera (PixeLink) is used to capture video of each\nexperiment.", "category": "physics_flu-dyn" }, { "text": "Unbalanced instabilities of rapidly rotating stratified shear flows: The linear stability of a rotating, stratified, inviscid horizontal plane\nCouette flow in a channel is studied in the limit of strong rotation and\nstratification. An energy argument is used to show that unstable perturbations\nmust have large wavenumbers. This motivates the use of a WKB-approach which, in\nthe first instance, provides an approximation for the dispersion relation of\nthe various waves that can propagate in the flow. These are Kelvin waves,\ntrapped near the channel walls, and inertia-gravity waves with or without\nturning points.\n Although, the wave phase speeds are found to be real to all algebraic orders\nin the Rossby number, we establish that the flow, whether cyclonic or\nanticyclonic, is unconditionally unstable. This is the result of linear\nresonances between waves with oppositely signed wave momenta. We derive\nasymptotic estimates for the instability growth rates, which are exponentially\nsmall in the Rossby number, and confirm them by numerical computations. Our\nresults, which extend those of Kushner et al (1998) and Yavneh et al (2001),\nhighlight the limitations of the so-called balanced models, widely used in\ngeophysical fluid dynamics, which filter out Kelvin and inertia-gravity waves\nand hence predict the stability of the Couette flow. They are also relevant to\nthe stability of Taylor-Couette flows and of astrophysical accretion discs.", "category": "physics_flu-dyn" }, { "text": "Controlled and impulsive compression of an entrapped air bubble during\n impact: Wave slamming onto a structure is often accompanied by the entrapment of an\nair pocket. A large scale impact typically has a rapidly evolving and disturbed\nliquid-gas interface, such that several bubbles are entrapped upon impact.\nWhile it is largely understood how the peak pressure is created by liquid\ncoming into contact with the solid structure, it is more challenging to\nascertain how an isolated air pocket is pressurised by an impulsive impact, and\nhow the maximum impact pressure inside this bubble evolves. We study such a\nBagnold-type impulsive compression of an air bubble by performing\nwell-controlled experiments, where we use an inverted, hollow cone as an\nimpactor. The cone is kept immersed throughout in a water bath, such that it\nencloses an air bubble of known and controlled volume. A high-sensitivity\nsensor measures pressures at the vertex of the cone. Using high-speed imaging\nwe show how incoming liquid deforms the air bubble enclosed in such a geometry,\nand how an impact peak is registered inside the bubble, which can be traced\nback to the impact of a liquid jet onto the pressure sensor. We compare the\nmeasured pressures to a Bagnold model, and discuss the dominant resonances in\nthe bubble. From visualisations of the deforming bubble, we also discuss the\nair-pocket's deformations, resulting from the presence of surrounding rigid\ngeometry (such as corrugations in an LNG containment membrane).", "category": "physics_flu-dyn" }, { "text": "A Physics-based Scaling of the Charging Rate in Latent Heat Thermal\n Energy Storage Devices: Thermal energy storage (TES) is increasingly recognized as an essential\ncomponent of efficient Combined Heat and Power (CHP), Concentrated Solar Power\n(CSP), Heating Ventilation and Air Conditioning (HVAC), and refrigeration as it\nreduces peak demand while helping to manage intermittent availability of energy\n(e.g., from solar or wind). Latent Heat Thermal Energy Storage (LHTES) is a\nviable option because of its high energy storage density. Parametric analysis\nof LHTES heat exchangers have been focused on obtaining data with laminar flow\nin the phase changing fluid and then fitting a functional form, such as a power\nlaw or polynomial, to those data. Alternatively, in this paper we present a\nparametric framework to analyze LHTES devices by identifying all relevant fluid\nparameters and corresponding dimensionless numbers. We present 64 simulations\nof an LHTES device using the finite volume method at four values of the\nGrashof, Prandtl and Reynolds numbers in the phase change material (PCM) and\nheat transfer fluid (HTF). We observe that with sufficient energy available in\nthe HTF, the effects of the HTF Reynolds number and Prandtl number on the heat\ntransfer rate are negligible. Under these conditions, we propose a time scale\nfor the variation of energy stored (or melt fraction) of the LHTES device based\non the Fourier number($Fo$), Grashof number($Gr_p$) and Prandtl number($Pr_p$)\nand observe a $Gr_p^1$ and $Pr_p^{(1/3)}$ dependency. We also identify two\ndistinct regions in the variation of the melt fraction with time, namely, the\nlinear and the asymptotic region. We also predict the critical value of the\nmelt fraction at the transition between the two regions. From these analyses,\nwe draw some conclusions regarding the design procedure for LHTES devices.", "category": "physics_flu-dyn" }, { "text": "A partially wetting film of water and surfactant under the influence of\n a propagating MHz surface acoustic wave: We use both theory and experiment to study the response of {\\it partially\nwetting} films of water and surfactant solutions to a propagating MHz vibration\nin the solid substrate in the form of a Rayleigh surface acoustic wave (SAW).\nThe SAW invokes a drift of mass in the liquid film, which is associated with\nthe Schlichting boundary layer flow (also known as the Schlichting streaming).\nWe study thin films that are governed by a balance between the drift and\ncapillary stress alone. We demonstrate weak capillary contributions, such as\nfor silicon oil films, support dynamic wetting and lead to the spreading of the\nliquid over the solid substrate along the path of the SAW. Strong capillary\ncontributions, such as for water films, support however a concurrent dynamic\nwetting and dewetting along the path of the SAW, such that the film displace\nalong the solid substrate. In addition, such films may support the formation of\na capillary train-wave that propagate along the the same path. We further note\nthe mechanism for film dynamics we discuss here is different to the more\nfamiliar Eckart streaming mechanism, which is associated with a film thickness\nthat is greater than the wavelength of the sound leakage off the SAW and\nusually observed to support the motion of drops. The thickness of the films we\ndiscuss here is small in respect to the wavelength of the sound leakage,\nrendering contributions from the Eckart streaming, acoustic radiation pressure,\nand the attenuation of the SAW small.", "category": "physics_flu-dyn" }, { "text": "Large scale three-dimensional topology optimisation of heat sinks cooled\n by natural convection: This work presents the application of density-based topology optimisation to\nthe design of three-dimensional heat sinks cooled by natural convection. The\ngoverning equations are the steady-state incompressible Navier-Stokes equations\ncoupled to the thermal convection-diffusion equation through the Bousinessq\napproximation. The fully coupled non-linear multiphysics system is solved using\nstabilised trilinear equal-order finite elements in a parallel framework\nallowing for the optimisation of large scale problems with order of 40-330\nmillion state degrees of freedom. The flow is assumed to be laminar and several\noptimised designs are presented for Grashof numbers between $10^3$ and $10^6$.\nInterestingly, it is observed that the number of branches in the optimised\ndesign increases with increasing Grashof numbers, which is opposite to\ntwo-dimensional optimised designs.", "category": "physics_flu-dyn" }, { "text": "Similarity and singularity in adhesive elastohydrodynamic touchdown: We consider the touchdown of an elastic sheet as it adheres to a wall, which\nhas a dynamics that is limited by the viscous resistance provided by the\nsqueeze flow of the intervening liquid trapped between the two solid surfaces.\nThe dynamics of the sheet is described mathematically by elastohydrodynamic\nlubrication theory, coupling the elastic deformation of the sheet, the\nmicroscopic van der Waals adhesion and the viscous thin film flow. We use a\ncombination of numerical simulations of the governing partial differential\nequation and a scaling analysis to describe the self-similar solution of the\ntouchdown of the sheet as it approaches the wall. An analysis of the equation\nsatisfied by the similarity variables in the vicinity of the touchdown event\nshows that an entire sequence of solutions are allowed. However, a comparison\nof these shows that only the fundamental similarity solution is observed in the\ntime-dependent numerical simulations, consistent with the fact that it alone is\nstable. Our analysis generalizes similar approaches for rupture in capillary\nthin film hydrodynamics and suggests experimentally verifiable predictions for\na new class of singular flows linking elasticity, hydrodynamics and adhesion.", "category": "physics_flu-dyn" }, { "text": "A statistical state dynamics approach to wall-turbulence: This paper reviews results from the study of wall-bounded turbulent flows\nusing statistical state dynamics (SSD) that demonstrate the benefits of\nadopting this perspective for understanding turbulence in wall-bounded shear\nflows. The SSD approach used in this work employs a second-order closure which\nisolates the interaction between the streamwise mean and the equivalent of the\nperturbation covariance. This closure restricts nonlinearity in the SSD to that\nexplicitly retained in the streamwise constant mean together with nonlinear\ninteractions between the mean and the perturbation covariance. This dynamical\nrestriction, in which explicit perturbation-perturbation nonlinearity is\nremoved from the perturbation equation, results in a simplified dynamics\nreferred to as the restricted nonlinear (RNL) dynamics. RNL systems in which an\nensemble of a finite number of realizations of the perturbation equation share\nthe same mean flow provide tractable approximations to the equivalently\ninfinite ensemble RNL system. The infinite ensemble system, referred to as the\nS3T, introduces new analysis tools for studying turbulence. The RNL with a\nsingle ensemble member can be alternatively viewed as a realization of RNL\ndynamics. RNL systems provide computationally efficient means to approximate\nthe SSD, producing self-sustaining turbulence exhibiting qualitative features\nsimilar to those observed in direct numerical simulations (DNS) despite its\ngreatly simplified dynamics. Finally, we show that RNL turbulence can be\nsupported by as few as a single streamwise varying component interacting with\nthe streamwise constant mean flow and that judicious selection of this\ntruncated support, or \"band-limiting\", can be used to improve quantitative\naccuracy of RNL turbulence. The results suggest that the SSD approach provides\nnew analytical and computational tools allowing new insights into\nwall-turbulence.", "category": "physics_flu-dyn" }, { "text": "Control of growth of local heat release rate fluctuations to suppress\n thermoacoustic instability: This experimental study investigates the dynamical transition from stable\noperation to thermoacoustic instability in a turbulent bluff-body stabilized\ndump combustor. We conduct experiments to characterize the dynamical transition\nutilizing acoustic pressure and local heat release rate fluctuations. We\nobserve the transition to thermoacoustic instability for these experiments as\nwe decrease the equivalence ratio towards a fuel-lean setting. More\nimportantly, we observe significant growth of local heat release rate\nfluctuations near the bluff-body well before the appearance of large-scale\nspatial or temporal patterns during the occurrence of thermoacoustic\ninstability. By strategically positioning slots (perforations) on the\nbluff-body, we ensure the reduction of the growth of local heat release rate\nfluctuations at the stagnation zone near the bluff-body at operating conditions\nfar away from the onset of thermoacoustic instability. This reduction in the\nlocal heat release rate fluctuations ensures that the transition to\nthermoacoustic instability is avoided. We find that modified configurations of\nthe bluff-body that do not quench these local heat release rate fluctuations at\nthe stagnation zone results in the transition to thermoacoustic instability. We\nalso reveal that an effective suppression strategy based on the growth of local\nheat release rate fluctuations requires an optimization of the area ratio of\nthe slots for a given bluff-body position. Further, the suppression strategy\nalso depends on the spatial distribution of perforations on the bluff-body.\nNotably, an inappropriate distribution of the slots which does not quench the\nlocal heat release rate fluctuations at the stagnation zone may even result in\na dramatic increase in amplitudes of pressure oscillations.", "category": "physics_flu-dyn" }, { "text": "Effect of streaks on hypersonic boundary layer instability: Hypersonic boundary layers exhibit diverse transition pathways, influenced by\nvarious flow conditions and environments. Non-modal mechanisms, such as the\nlift-up effect, are recognized as pivotal contributors to transition,\nparticularly over complex geometries like rough walls or blunted forebodies. In\nthis study, we investigate the impact of steady streaks on the transition to\nturbulence in hypersonic boundary layers. Streaky baseflows are generated using\noptimal disturbances at the inlet, forming the basis for our analysis. We\nconduct linearized direct numerical simulations on these baseflows at Mach\nnumber $M_\\infty=6.0$, using white noise forcing to trigger instabilities. An\nefficient extension of the SPOD method is applied to identify and track the\ninstabilities originating from the broadband forcing on the 3D non-homogeneous\nbaseflows. The results highlight the dominant influence of low-frequency\nfirst-mode instabilities and streak-associated instabilities in the linear\nregime. These findings emphasize the critical role of streaks in hypersonic\nboundary layer transition and provide valuable insights into this complex\nphenomenon.", "category": "physics_flu-dyn" }, { "text": "Applying Laser Doppler Anemometry inside a Taylor-Couette geometry -\n Using a ray-tracer to correct for curvature effects: In the present work it will be shown how the curvature of the outer cylinder\naffects Laser Doppler anemometry measurements inside a Taylor-Couette\napparatus. The measurement position and the measured velocity are altered by\ncurved surfaces. Conventional methods for curvature correction are not\napplicable to our setup, and it will be shown how a ray-tracer can be used to\nsolve this complication.\n By using a ray-tracer the focal position can be calculated, and the velocity\ncan be corrected. The results of the ray-tracer are verified by measuring an a\npriori known velocity field, and after applying refractive corrections good\nagreement with theoretical predictions are found. The methods described in this\npaper are applied to measure the azimuthal velocity profiles in high Reynolds\nnumber Taylor-Couette flow for the case of outer cylinder rotation.", "category": "physics_flu-dyn" }, { "text": "Injectable Spontaneous Generation of Tremendous Self-Fueled Liquid Metal\n Droplet Motors in a Moment: Micro motors that could run in liquid environment is very important for a\nvariety of practices such as serving as pipeline robot, soft machine, drug\ndelivery, or microfluidics system etc. However, fabrication of such tiny motors\nis generally rather time and cost consumptive and has been a tough issue due to\ninvolve too many complicated procedures and tools. Here, we show a\nstraightforward injectable way for spontaneously generating autonomously\nrunning soft motors in large quantity. A basic fabrication strategy thus\nenabled is established and illustrated. It was found that, injecting the GaIn\nalloy pre-fueled with aluminum into electrolyte would automatically split in\nseconds into tremendous droplet motors swiftly running here and there. The\ndriving force originated from the galvanic cell reaction among alloy, aluminum\nand surrounding electrolyte which offers interior electricity and hydrogen gas\nas motion power. This finding opens the possibility to develop injectable\ntiny-robots, droplet machines or microfluidic elements. It also raised\nimportant scientific issues regarding characterizing the complicated fluid\nmechanics stimulated by the quick running of the soft metal droplet and the\ngases it generated during the traveling.", "category": "physics_flu-dyn" }, { "text": "Drops bouncing off macro-textured superhydrophobic surfaces: Recent experiments with droplets impacting a macro-textured superhydrophobic\nsurfaces revealed new regimes of bouncing with a remarkable reduction of the\ncontact time. We present here a comprehensive numerical study that reveals the\nphysics behind these new bouncing regimes and quantify the role played by\nvarious external and internal forces that effect the dynamics of a drop\nimpacting a complex surface. For the first time, three-dimensional simulations\ninvolving macro-textured surfaces are performed. Aside from demonstrating that\nsimulations reproduce experiments in a quantitative manner, the study is\nfocused on analyzing the flow situations beyond current experiments. We show\nthat the experimentally observed reduction of contact time extends to higher\nWeber numbers, and analyze the role played by the texture density. Moreover, we\nreport a non-linear behavior of the contact time with the increase of the Weber\nnumber for application relevant imperfectly coated textures, and also study the\nimpact on tilted surfaces in a wide range of Weber numbers. Finally, we present\nnovel energy analysis techniques that elaborate and quantify the interplay\nbetween the kinetic and surface energy, and the role played by the dissipation\nfor various Weber numbers.", "category": "physics_flu-dyn" }, { "text": "The effect of phase change on stability of convective flow in a layer of\n volatile liquid driven by a horizontal temperature gradient: Buoyancy-thermocapillary convection in a layer of volatile liquid driven by a\nhorizontal temperature gradient arises in a variety of situations. Recent\nstudies have shown that the composition of the gas phase, which is typically a\nmixture of vapour and air, has a noticeable effect on the critical Marangoni\nnumber describing the onset of convection as well as on the observed convection\npattern. Specifically, as the total pressure or, equivalently, the average\nconcentration of air is decreased, the threshold of the instability leading to\nthe emergence of convective rolls is found to increase rather significantly. We\npresent a linear stability analysis of the problem which shows that this trend\ncan be readily understood by considering the transport of heat and vapour\nthrough the gas phase. In particular, we show that transport in the gas phase\nhas a noticeable effect even at atmospheric conditions, when phase change is\ngreatly suppressed.", "category": "physics_flu-dyn" }, { "text": "Surface texture and pulsation due to balloon bursting in different\n liquids: We study the instabilities occurring during the burst of an air balloon in a\nliquid. These instabilities are typical for the deformation of an interface\nbetween two fluids of different densities, similar to fingering in\nRayleigh-Taylor instability (see e.g. Sharp, 1984). In the Video a series of\nbursts are shown for air balloons in different liquids. When the balloon tears\nit tracks the surface, generating wrinkles and releasing the pressure inside.\nApparently, the texture of the surface during the burst becomes smoother as the\nviscosity increases. During the burst the surface breaks and generates several\nsmall bubbles. Furthermore, the pressure inside the balloon is higher than the\nexternal pressure before the burst; once the balloon tears the pressure is\nreleased and the generated bubbles pulsate several times (see e.g. Brennen,\n1995). Such oscillations are more evident for higher internal pressures.", "category": "physics_flu-dyn" }, { "text": "Laboratory Experiments on Wave Turbulence: This review paper is devoted to a presentation of recent progress in wave\nturbulence. I first present the context and state of the art of this field of\nresearch both experimentally and theoretically. I then focus on the case of\nwave turbulence on the surface of a fluid, and I discuss the main results\nobtained by our group: caracterization of the gravity and capillary wave\nturbulence regimes, the first observation of intermittency in wave turbulence,\nthe occurrence of strong fluctuations of injected power in the fluid, the\nobservation of a pure capillary wave turbulence in low gravity environment and\nthe observation of magnetic wave turbulence on the surface of a ferrofluid.\nFinally, open questions in wave turbulence are discussed.", "category": "physics_flu-dyn" }, { "text": "Nonlinear waves on the surface of a fluid covered by an elastic sheet: We experimentally study linear and nonlinear waves on the surface of a fluid\ncovered by an elastic sheet where both tension and flexural waves take place.\nAn optical method is used to obtain the full space-time wave field, and the\ndispersion relation of waves. When the forcing is increased, a significant\nnonlinear shift of the dispersion relation is observed. We show that this shift\nis due to an additional tension of the sheet induced by the transverse motion\nof a fundamental mode of the sheet. When the system is subjected to a random\nnoise forcing at large scale, a regime of hydro-elastic wave turbulence is\nobserved with a power-law spectrum of the scale in disagreement with the wave\nturbulence prediction. We show that the separation between relevant time scales\nis well satisfied at each scale of the turbulent cascade as expected\ntheoretically. The wave field anisotropy, and finite size effects are also\nquantified and are not at the origin of the discrepancy. Finally, the\ndissipation is found to occur at all scales of the cascade contrary to the\ntheoretical hypothesis, and could thus explain this disagreement.", "category": "physics_flu-dyn" }, { "text": "Multiscale properties of Large Eddy Simulations: correlations between\n resolved-scale velocity-field increments and subgrid-scale quantities: We provide analytical and numerical results concerning multi-scale\ncorrelations between the resolved velocity field and the subgrid-scale (SGS)\nstress-tensor in large eddy simulations (LES). Following previous studies for\nNavier-Stokes equations (NSE), we derive the exact hierarchy of LES equations\ngoverning the spatio-temporal evolution of velocity structure functions of any\norder. The aim is to assess the influence of the sub-grid model on the inertial\nrange intermittency. We provide a series of predictions, within the\nmultifractal theory, for the scaling of correlation involving the SGS stress\nand we compare them against numerical results from high-resolution Smagorinsky\nLES and from a-priori filtered data generated from direct numerical simulations\n(DNS). We find that LES data generally agree very well with filtered DNS\nresults and with the multifractal prediction for all leading terms in the\nbalance equations. Discrepancies are measured for some of the subleading terms\ninvolving cross-correlation between resolved velocity increments and the SGS\ntensor or the SGS energy transfer, suggesting that there must be room to\nimprove the SGS modelisation to further extend the inertial range properties\nfor any fixed LES resolution.", "category": "physics_flu-dyn" }, { "text": "Separable unsteady nonparallel flow stability problems: The so-called 'direct' approach to separation of variables in linear PDEs is\napplied to the hydrodynamic stability problem. Calculations are made for the\ncomplete linear stability equations in cylindrical coordinates. Several classes\nof the exact solutions of the Navier-Stokes equations describing spatially\ndeveloping and unsteady flows, for which the linear stability problems can be\nrigorously reduced to eigenvalue problems of ordinary differential equations,\nare defined. Those exactly solvable nonparallel and unsteady flow stability\nproblems can be used for testing approximate approaches and the methods based\non direct numerical simulations of the (linearized) Navier-Stokes equations.\nThe exact solutions of the viscous incompressible Navier-Stokes equations\ndetermined as the basic states, for which the linear stability problem is\nexactly separable, may be themselves of interest from theoretical and\nengineering points of view.", "category": "physics_flu-dyn" }, { "text": "Coupling kinetic and continuum using data-driven maximum entropy\n distribution: An important class of multi-scale flow scenarios deals with an interplay\nbetween kinetic and continuum phenomena. While hybrid solvers provide a natural\nway to cope with these settings, two issues restrict their performance.\nForemost, the inverse problem implied by estimating distributions has to be\naddressed, to provide boundary conditions for the kinetic solver. The next\nissue comes from defining a robust yet accurate switching criterion between the\ntwo solvers. This study introduces a data-driven kinetic-continuum coupling,\nwhere the Maximum-Entropy-Distribution (MED) is employed to parametrize\ndistributions arising from continuum field variables. Two regression\nmethodologies of Gaussian-Processes (GPs) and Artificial-Neural-Networks (ANNs)\nare utilized to predict MEDs efficiently. Hence the MED estimates are employed\nto carry out the coupling, besides providing a switching criterion. To achieve\nthe latter, a continuum breakdown parameter is defined by means of the Fisher\ninformation distance computed from the MED estimates. We test the performance\nof our devised MED estimators by recovering bi-modal densities. Next, MED\nestimates are integrated into a hybrid kinetic-continuum solution algorithm.\nHere Direct Simulation Monte-Carlo (DSMC) and Smoothed-Particle Hydrodynamics\n(SPH) are chosen as kinetic and continuum solvers, respectively. The problem of\nmonatomic gas inside Sod's shock tube is investigated, where DSMC-SPH coupling\nis realized by applying the devised MED estimates. Very good agreements with\nrespect to benchmark solutions along with a promising speed-up are observed in\nour reported test cases.", "category": "physics_flu-dyn" }, { "text": "A volume-based description of gas flows with localised mass-density\n variations: We reconsider some fundamental aspects of the fluid mechanics model, and the\nderivation of continuum flow equations from gas kinetic theory. Two topologies\nfor fluid representation are presented, and a set of macroscopic equations are\nderived through a modified version of the classical Boltzmann kinetic equation\nfor monatomic gases. The free volumes around the gaseous molecules are\nintroduced into the set of kinetic microscopic parameters. Our new description\ncomprises four, rather than three, conservation equations; the classical\ncontinuity equation, which conflates actual mass-density and number-density in\na single equation, has been split into a conservation equation of mass (which\ninvolves only the classical number-density of the gaseous particles) and an\nevolution equation purely of the mass-density (mass divided by the actual\nvolume of the fluid). We propose this model as a better description of gas\nflows displaying non-local-thermodynamic-equilibrium (rarefied flows), flows\nwith relatively large variations of macroscopic properties, and/or highly\ncompressible fluids/flows.", "category": "physics_flu-dyn" }, { "text": "Ewald summation for the rotlet singularity of Stokes flow: Ewald summation is an efficient method for computing the periodic sums that\nappear when considering the Green's functions of Stokes flow together with\nperiodic boundary conditions. We show how Ewald summation, and accompanying\ntruncation error estimates, can be easily derived for the rotlet, by\nconsidering it as a superposition of electrostatic force calculations.", "category": "physics_flu-dyn" }, { "text": "A simple model of ultrasound propagation in a cavitating liquid. Part\n II: Primary Bjerknes force and bubble structures: In a companion paper, a reduced model for propagation of acoustic waves in a\ncloud of inertial cavitation bubbles was proposed. The wave attenuation was\ncalculated directly from the energy dissipated by a single bubble, the latter\nbeing estimated directly from the fully nonlinear radial dynamics. The use of\nthis model in a mono-dimensional configuration has shown that the attenuation\nnear the vibrating emitter was much higher than predictions obtained from\nlinear theory, and that this strong attenuation creates a large traveling wave\ncontribution, even for closed domain where standing waves are normally\nexpected. In this paper, we show that, owing to the appearance of traveling\nwaves, the primary Bjerknes force near the emitter becomes very large and tends\nto expel the bubbles up to a stagnation point. Two-dimensional axi-symmetric\ncomputations of the acoustic field created by a large area immersed sonotrode\nare also performed, and the paths of the bubbles in the resulting Bjerknes\nforce field are sketched. Cone bubble structures are recovered and compare\nreasonably well to reported experimental results. The underlying mechanisms\nyielding such structures is examined, and it is found that the conical\nstructure is generic and results from the appearance a sound velocity gradient\nalong the transducer area. Finally, a more complex system, similar to an\nultrasonic bath, in which the sound field results from the flexural vibrations\nof a thin plate, is also simulated. The calculated bubble paths reveal the\nappearance of other commonly observed structures in such configurations, such\nas streamers and flare structures.", "category": "physics_flu-dyn" }, { "text": "Monitoring the orientation of rare-earth-doped nanorods for flow shear\n tomography: Rare-earth phosphors exhibit unique luminescence polarization features\noriginating from the anisotropic symmetry of the emitter ion's chemical\nenvironment. However, to take advantage of this peculiar property, it is\nnecessary to control and measure the ensemble orientation of the host particles\nwith a high degree of precision. Here, we show a methodology to obtain the\nphotoluminescence polarization of Eu-doped LaPO4 nano rods assembled in an\nelectrically modulated liquid-crystalline phase. We measure Eu3+ emission\nspectra for the three main optimal configurations ({\\sigma}, {\\pi} and\n{\\alpha}, depending on the direction of observation and the polarization axes)\nand use them as a reference for the nano rod orientation analysis. Based on the\nfact that flowing nano rods tend to orient along the shear strain profile, we\nuse this orientation analysis to measure the local shear rate in a flowing\nliquid. The potential of this approach is then demonstrated through tomographic\nimaging of the shear rate distribution in a microfluidic system.", "category": "physics_flu-dyn" }, { "text": "Eulerian-Lagrangian method for simulation of cloud cavitation: We present a coupled Eulerian-Lagrangian method to simulate cloud cavitation\nin a compressible liquid. The method is designed to capture the strong,\nvolumetric oscillations of each bubble and the bubble-scattered acoustics. The\ndynamics of the bubbly mixture is formulated using volume-averaged equations of\nmotion. The continuous phase is discretized on an Eulerian grid and integrated\nusing a high-order, finite-volume weighted essentially non-oscillatory (WENO)\nscheme, while the gas phase is modeled as spherical, Lagrangian point-bubbles\nat the sub-grid scale, each of whose radial evolution is tracked by solving the\nKeller-Miksis equation. The volume of bubbles is mapped onto the Eulerian grid\nas the void fraction by using a regularization (smearing) kernel. In the most\ngeneral case, where the bubble distribution is arbitrary, three-dimensional\nCartesian grids are used for spatial discretization. In order to reduce the\ncomputational cost for problems possessing translational or rotational\nhomogeneities, we spatially average the governing equations along the direction\nof symmetry and discretize the continuous phase on two-dimensional or\naxi-symmetric grids, respectively. We specify a regularization kernel that maps\nthe three-dimensional distribution of bubbles onto the field of an averaged\ntwo-dimensional or axi-symmetric void fraction. A closure is developed to model\nthe pressure fluctuations at the sub-grid scale as synthetic noise. For the\nexamples considered here, modeling the sub-grid pressure fluctuations as white\nnoise agrees a priori with computed distributions from three-dimensional\nsimulations, and suffices, a posteriori, to accurately reproduce the statistics\nof the bubble dynamics. The numerical method and its verification are described\nby considering test cases of the dynamics of a single bubble and cloud\ncavitaiton induced by ultrasound fields.", "category": "physics_flu-dyn" }, { "text": "Study of Instability of Liquid Jets Under Gravity: Breakup of water jets under gravity is a common-place phenomenon. The role of\nsurface tension in the instability of water jets was recognized by Rayleigh and\nthe theory propounded goes by the name of Plateau-Rayleigh theory. The necks\nand bulges down along the jet-length that are created by perturbation waves of\nwavelengths larger than a certain value keep growing with time and ultimately\ncause the jet to breakup into drops. The effect of perturbation waves have been\ninvestigated experimentally and found to confirm the essentials of the theory.\nHowever, there is no unanimity about the origin of these perturbation waves.\nRecently, the idea of recoil capillary waves as an important source of the\nperturbation waves has been emphasized. The recoil of the end point of the\nremaining continuous jet at its breakup point is considered to travel upward as\na recoil capillary wave which gets reflected at the mouth of the nozzle from\nwhich the jet originates. The reflected capillary wave travels along the jet\ndownward with its Doppler shifted wavelength as a perturbation wave. We set up\nan experiment to directly verify the existence and effect of the recoil\ncapillary waves and present some preliminary results of our experiment.", "category": "physics_flu-dyn" }, { "text": "A review of turbulent skin-friction drag reduction by near-wall\n transverse forcing: The quest for reductions in fuel consumption and CO2 emissions in transport\nhas been a powerful driving force for scientific research into methods that\nmight underpin drag-reducing technologies for a variety of vehicular transport\non roads, by rail, in the air, and on or in the water. In civil aviation,\nskin-friction drag accounts for around 50% of the total drag in cruise\nconditions, thus being a preferential target for research. With laminar\nconditions excluded, skin friction is intimately linked to the turbulence\nphysics in the fluid layer closest to the skin. Thus, research into drag\nreduction has focused on methods to depress the turbulence activity near the\nsurface. The most effective method of doing so is to subject the drag-producing\nflow in the near-wall layer to an unsteady and/or spatially varying cross-flow\ncomponent, either by the action of transverse wall oscillations, by embedding\nrotating discs into the surface or by plasma-producing electrodes that\naccelerate the near-wall fluid in the transverse direction. In ideal\nconditions, drag-reduction margins of order of 50% can be achieved. The present\narticle provides a review of research into the response of turbulent near-wall\nlayers to the imposition of unsteady and wavy transverse motion, encompassing\nexperiments, simulation, analysis and modelling. It covers issues such as the\ndrag-reduction margin in a variety of actuation scenarios, the underlying\nphysical phenomena that contribute to the interpretation of the origin of the\ndrag reduction, the dependence of the drag reduction on the Reynolds number,\npassive control methods that are inspired by active control, and a forward look\ntowards possible future research and practical realizations. The authors hope\nthat this review, by far the most extensive of its kind for this subject, will\nbe judged as a useful foundation for future research targeting friction-drag\nreduction", "category": "physics_flu-dyn" }, { "text": "$\u03bb$-Navier-Stokes turbulence: We investigate numerically the model proposed in Sahoo et al [Phys. Rev.\nLett. 118, 164501, (2017)] where a parameter $\\lambda$ is introduced in the\nNavier-Stokes equations such that the weight of homochiral to heterochiral\ninteractions is varied while preserving all original scaling symmetries and\ninviscid invariants. Decreasing the value of $\\lambda$ leads to a change in the\ndirection of the energy cascade at a critical value $\\lambda_c \\sim 0.3$. In\nthis work, we perform numerical simulations at varying $\\lambda$ in the forward\nenergy cascade range and at changing the Reynolds number $\\mathrm{Re}$. We show\nthat for a fixed injection rate, as $\\lambda \\to \\lambda_c$, the kinetic energy\ndiverges with a scaling law $\\mathcal{E} \\propto (\\lambda-\\lambda_c)^{-2/3}$.\nThe energy spectrum is shown to display a larger bottleneck as $\\lambda$ is\ndecreased. The forward heterochiral flux and the inverse homochiral flux both\nincrease in amplitude as $\\lambda_c$ is approached while keeping their\ndifference fixed and equal to the injection rate. As a result, very close to\n$\\lambda_c$ a stationary state is reached where the two opposite fluxes are of\nmuch higher amplitude than the mean flux and large fluctuations are observed.\nFurthermore, we show that intermittency as $\\lambda_c$ is approached is\nreduced. The possibility of obtaining a statistical description of regular\nNavier-Stokes turbulence as an expansion around this newly found critical point\nis discussed.", "category": "physics_flu-dyn" }, { "text": "A method to estimate the size of particles using the open source\n software OpenPTV: A method to obtain particle sizes from images that are used for particle\ntracking velocimetry is proposed. This is an open source method, developed to\nuse together with the open source software OpenPTV. First, the analysis of\ndifferent factors that affect the estimation of the particle size is made.\nThen, a transformation is proposed to estimate the sizes from information\nobtained in OpenPTV. The method requires an extra calibration with a flat\ntarget containing circles of different sizes, but if the calibration is\nsuccessful, the sizes of particles can be estimated reliable within certain\nlimits. For example, for solid of non-translucent particles, sizes can be\nestimated with an error of maximum 10 %, as long as the particle diameter\ncorrespond to at least three pixels width in the obtained images. For\ntranslucent particles some extra assumptions will be needed in the\ntransformation.", "category": "physics_flu-dyn" }, { "text": "Trajectory of particles exposed to a tilted-angle washboard potential:\n application to high-throughput acoustophoresis in microchannels: A wide variety of tilted washboard potentials based on acoustic waves,\nelectric fields, optical patterns and so on have been applied to sort particles\nin microchannels. In this paper, we present a theoretical analysis of the\nparticles trajectory in a washboard potential making a tilt angle $\\theta$ with\nthe flow. Depending on the sorting to drag force ratio $\\epsilon$, we\nidentified a transition threshold $\\epsilon = \\cos(\\theta)$ between two\ndistinct regimes of particles motion: drift and locked modes. In drift mode,\nthe particles follow an oscillating trajectory which slope is given by\n$\\frac{\\epsilon^2}{2}\\tan(\\theta)$, while in locked mode the trajectory slope\nis given by $1/\\tan(\\theta)$. These theoretical predictions agree\nquantitatively well with previously published experimental data and may help in\nthe design of high-performance microfluidic sorting devices.", "category": "physics_flu-dyn" }, { "text": "Bees with attitude: the effect of gusts on flight dynamics: Flight is a complicated task at small scales in part due to the ubiquitous\nunsteady air which contains it. Flying organisms deal with these difficulties\nusing active and passive control mechanisms to steer their body motion. Body\nattitudes of flapping organisms are linked with their resultant flight\ntrajectories and performance, yet little is understood about how discrete\nunsteady aerodynamic phenomena affect the interlaced dynamics of such systems.\nIn this study, we examined freely flying bumblebees subject to a single\ndiscrete gust to emulate aerodynamic disturbances encountered in nature.\nBumblebees are expert commanders of the aerial domain as they persistently\nforage within complex terrain elements. Physical obstacles such as flowers\nproduce local effects representative of a typified gust which threatens the\nprecise control of intricate maneuvers. By tracking the 3D dynamics of bees\nflying through gusts, we determined the sequences of motion that permit flight\nin three disturbance conditions. Bees repetitively executed a series of passive\nimpulsive maneuvers followed by active recovery maneuvers. Impulsive motion was\nunique in each gust direction, maintaining control purely by passive\nmanipulation of the body. Bees pitched up and slowed-down at the beginning of\nrecovery in every disturbance, followed by corrective maneuvers which brought\nattitudes back to their original state. Bees were displaced the most by the\nsideward gust, displaying large lateral translations and roll deviations.\nUpward gusts were easier for bees to fly through, causing only minor flight\nchanges and minimal recovery times. Downward gusts severely impaired the\ncontrol response of bees, inflicting strong adverse forces which sharply upset\ntrajectories. Bees used interesting control strategies when flying in each\ndisturbance, offering new insights into insect-scale flapping flight and\nbio-inspired robotic systems.", "category": "physics_flu-dyn" }, { "text": "UnDiFi-2D: an Unstructured Discontinuity Fitting code for 2D grids: UnDiFi-2D, an open source (free software) Unstructured-grid, Discontinuity\nFitting code, is presented. The aim of UnDiFi-2D is to model gas-dynamic\ndiscontinuities in two-dimensional (2D) flows as if they were true\ndiscontinuities of null thickness that bound regions of the flow-field where a\nsmooth solution to the governing PDEs exists. UnDiFi-2D therefore needs to be\ncoupled with an unstructured CFD solver that is used to discretize the\ngoverning PDEs within the smooth regions of the flow-field. Two different,\nin-house developed, CFD solvers are also included in the current distribution.\nThe main features of the UnDiFi-2D software can be summarized as follows:\n Programming Language: UnDiFi-2D is written in standard Fortran 77/95; its\ndesign is highly modular in order to enhance simplicity of use, maintenance and\nallow coupling with virtually any existing CFD solver;\n Usability, Maintenance and Enhancement: In order to improve the usability,\nmaintenance and enhancement of the code also the documentation has been\ncarefully taken into account. The git distributed versioning system has been\nadopted to facilitate collaborative maintenance and code development;\n Copyrights: UnDiFi-2D is a free software that anyone can use, copy,\ndistribute, change and improve under the GNU Public License version 3.\n The present paper is a manifesto of the first public release of the UnDiFi-2D\ncode. It describes the currently implemented features, which are the result of\nmore than a decade of still ongoing CFD developments. This work is focused on\nthe computational techniques adopted and a detailed description of the main\ncharacteristics is reported. UnDiFi-2D capabilities are demonstrated by means\nof examples test cases. The design of the code allows to easily include\nexisting CFD codes and is aimed at ease code reuse and readability.", "category": "physics_flu-dyn" }, { "text": "Instantons in a Lagrangian model of turbulence: The role of instantons is investigated in the Lagrangian model for the\nvelocity gradient evolution known as the Recent Fluid Deformation\napproximation. After recasting the model into the path-integral formalism, the\nprobability distribution function is computed along with the most probable path\nin the weak noise limit through the saddle-point approximation. Evaluation of\nthe instanton solution is implemented numerically by means of the iteratively\nChernykh-Stepanov method. In the case of the longitudinal velocity gradient\nstatistics, due to symmetry reasons, the number of degrees of freedom can be\nreduced to one, allowing the pdf to be evaluated analytically as well, thereby\nenabling a prediction of the scaling of the moments as a function of Reynolds\nnumber. It is also shown that the instanton solution lies on the Vieillefosse\nline concerning the RQ-plane. We illustrate how instantons can be unveiled in\nthe stochastic dynamics performing a conditional statistics.", "category": "physics_flu-dyn" }, { "text": "Flow characterisation and power consumption in an inline high shear\n rotor-stator mixer using CFD: The aim of this paper is two-fold: (1) to provide a detailed investigation of\nthe turbulent flow in an inline high-shear rotor stator mixer; (2) to provide a\ncomparison of two different classes of turbulence models and solution methods\ncurrently available. The widely used multiple reference frame (MRF) method is\ncontrasted against a more recently developed sliding mesh method. The sliding\nmesh algorithm accounts for rotation of the blades and is able to capture the\ntransient effects arising from the rotor-stator interaction. The choice of\nturbulence model is shown to have a significant impact, with second moment\nclosures able to capture best the hydrodynamics. With an appropriate choice of\nturbulence model and solution algorithm, we thus demonstrate the capacity of\nCFD to provide accurate and computationally cost effective characteristic power\ncurve predictions.", "category": "physics_flu-dyn" }, { "text": "MHD turbulence and distributed chaos: It is shown, using results of recent direct numerical simulations, that\nspectral properties of distributed chaos in MHD turbulence with zero mean\nmagnetic field are similar to those of hydrodynamic turbulence. An exception is\nMHD spontaneous breaking of space translational symmetry, when the stretched\nexponential spectrum $\\exp(-k/k_{\\beta})^{\\beta}$ has $\\beta=4/7$.", "category": "physics_flu-dyn" }, { "text": "Bio-inspired variable-stiffness flaps for hybrid flow control, tuned via\n reinforcement learning: A bio-inspired, passively deployable flap attached to an airfoil by a\ntorsional spring of fixed stiffness can provide significant lift improvements\nat post-stall angles of attack. In this work, we describe a hybrid\nactive-passive variant to this purely passive flow control paradigm, where the\nstiffness of the hinge is actively varied in time to yield passive\nfluid-structure interaction (FSI) of greater aerodynamic benefit than the\nfixed-stiffness case. This hybrid active-passive flow control strategy could\npotentially be implemented using variable stiffness actuators with less expense\ncompared with actively prescribing the flap motion. The hinge stiffness is\nvaried via a reinforcement learning (RL)-trained closed-loop feedback\ncontroller. A physics-based penalty and a long-short-term training strategy for\nenabling fast training of the hybrid controller are introduced. The hybrid\ncontroller is shown to provide lift improvements as high as 136\\% and 85\\% with\nrespect to the flap-less airfoil and the best fixed-stiffness case,\nrespectively. These lift improvements are achieved due to large-amplitude flap\noscillations as the stiffness varies over four orders of magnitude, whose\ninterplay with the flow is analyzed in detail.", "category": "physics_flu-dyn" }, { "text": "Effects of Syringe Pump Fluctuations On Cell-Free Layer in Hydrodynamic\n Separation Microfluidic Devices: Syringe pumps are widely used biomedical equipment which offer low-cost\nsolutions to drive and control flow through microfluidic chips. However, they\nhave been shown to transmit mechanical oscillations resulting from their\nstepper motors, into the flow, perturbing device performance. In this work,\nunlike previous studies at lower flow rates, we have uncovered that the\nrelative pressure fluctuation plateau from 5mL/h onwards to approximately 2% of\nthe average pressure. Furthermore, we find that absolute pressure fluctuations\nincrease as a non-linear monotonic function of kinematic viscosity at flow\nrates in the 5-25 mL/h range, while the relative pressure fluctuations peak at\n1.25 cSt. Using a novel low-cost coded compressive rotating mirror (CCRM)\ncamera, we investigated the effect of fluctuations in a hydrodynamic\nmicrofluidic separation device based on a cell-free layer concept. Using this\nhigh-speed imaging set-up, we quantified the cell-free zone width fluctuations\nat bifurcations. We demonstrated that these fluctuations have the same\nfrequency and amplitude than the syringe pump induced pressure oscillations.\nFinally, to illustrate that pressure fluctuations degrade the separation\nefficiency in such devices, we demonstrate using milk samples, instances of\nparticles diverted to undesired outlets. This work provides an insight into the\neffect of syringe pump fluctuations on microfluidic separation, which will\ninform the design of microfluidic systems and improve their resilience to\npulsating or fluctuating flow conditions.", "category": "physics_flu-dyn" }, { "text": "Dynamical landscape of transitional pipe flow: The transition to turbulence in pipes is characterized by a coexistence of\nlaminar and turbulent states. At the lower end of the transition, localized\nturbulent pulses, called puffs, can be excited. Puffs can decay when rare\nfluctuations drive them close to an edge state lying at the phase-space\nboundary with laminar flow. At higher Reynolds numbers, homogeneous turbulence\ncan be sustained, and dominates over laminar flow. Here we complete this\nlandscape of localized states, placing it within a unified bifurcation picture.\nWe demonstrate our claims within the Barkley model, and motivate them\ngenerally. Specifically, we suggest the existence of an antipuff and a gap-edge\n-- states which mirror the puff and related edge state. Previously observed\nlaminar gaps forming within homogeneous turbulence are then naturally\nidentified as antipuffs nucleating and decaying through the gap edge.", "category": "physics_flu-dyn" }, { "text": "Influence of the porosity pattern on the aerodynamics of a square plate: The evolution of the normal aerodynamic coefficient of 19 configurations of\nsquare plates with various porosity patterns, ranging from solid plate to\nhomogeneous porous plate, is experimentally characterized. The variation of the\nporosity pattern is obtained by partially covering the holes of a commercial\nfly-swatter using adhesive tape. Evolution of the normal aerodynamic\ncoefficient is assessed from the measurement of the angular position of the\nporous plate, placed as a freely rotating pendulum swept by a flow in a wind\ntunnel. These angular measurements are also supported by PIV measurements of\nthe structure of the wake. We show that the porosity pattern determines whether\nor not an abrupt stall occurs. In particular, the details of the porosity\npattern on the edges of the plate are decisive for the existence of abrupt\nstall.", "category": "physics_flu-dyn" }, { "text": "Experimental study of the inverse cascade in gravity wave turbulence: We perform experiments to study the inverse cascade regime of gravity wave\nturbulence on the surface of a fluid. Surface waves are forced at an\nintermediate scale corresponding to the gravity-capillary wavelength. In\nresponse to this forcing, waves at larger scales are observed. The spectrum of\ntheir amplitudes exhibits a frequency-power law at high enough forcing. Both\nobservations are ascribed to the upscale wave action transfers of gravity wave\nturbulence. The spectrum exponent is close to the value predicted by the weak\nturbulence theory. The spectrum amplitude is found to scale linearly with the\nmean injected power. We measure also the distributions of the injected power\nfluctuations in the presence of upscale (inverse) transfers or in the presence\nof a downscale (direct) cascade in gravity wave turbulence.", "category": "physics_flu-dyn" }, { "text": "Direct numerical simulations of Taylor--Couette turbulence: the effect\n of sand grain roughness: Progress in roughness research, mapping any given roughness geometry to its\nfluid dynamic behaviour, has been hampered by the lack of accurate and direct\nmeasurements of skin-friction drag, especially in open systems. The\nTaylor--Couette (TC) system has the benefit of being a closed system, but its\npotential for characterizing irregular, realistic, 3-D roughness has not been\npreviously considered in depth. Here, we present direct numerical simulations\n(DNSs) of TC turbulence with sand grain roughness mounted on the inner\ncylinder. The model proposed by Scotti (\\textit{Phys. Fluids}, vol. 18, 031701,\n2006) has been improved to simulate a random rough surface of monodisperse sand\ngrains, which is characterized by the equivalent sand grain height $k_s$.\nTaylor numbers range from $Ta = 1.0\\times 10^7$(corresponding to $Re_\\tau =\n82$) to $Ta = 1.0\\times 10^9$($Re_\\tau = 635$). We focus on the influence of\nthe roughness height $k_s^+$ in the transitionally rough regime, through\nsimulations of TC with rough surfaces, ranging from $k_s^+=5$ up to $k_s^+ =\n92$, where the superscript `$+$' indicates non-dimensionalization in viscous\nunits. We find that the downwards shift of the logarithmic layer, due to\ntransitionally rough sand grains exhibits remarkably similar behavior to that\nof the Nikuradse (\\textit{VDI-Forschungsheft} 361, 1933) data of sand grain\nroughness in pipe flow, regardless of the Taylor number dependent constants of\nthe logarithmic layer.", "category": "physics_flu-dyn" }, { "text": "New Property Averaging Scheme for Volume of Fluid Method for Two-Phase\n Flows with Large Viscosity Ratios: To predict liquid-gas two-phase flow phenomena, accurate tracking and\nprediction of the evolving liquid-gas interface is required. Volume-of-Fluid or\nVoF method has been used in the literature for computationally modeling of such\nflows. In the VoF method, a single set of governing equations are solved for\nboth phases along with an advection equation for the volume fraction. The\nproperties in each computational cell are determined by a linear weighted\naverage of the properties of the two fluids based on the phase fraction. While\nthe method predicts water-air flows well, the predictions tend to deviate\nsignificantly from experimental data for liquids with high viscosity. A new\nproperty averaging technique is proposed in this paper, which is shown to\nprovide accurate results for high viscosity liquids. Computational predictions\nusing the open source VoF solver interFoam (available as a part of the OpenFOAM\ncomputational tool), and those obtained using the proposed method are compared\nwith experimental data for multiple two-phase applications. Four different\nproblems, viz., suspended droplet in air, jet breakup, drop impact on thin\nfilms, and air entrapment during drop interaction with liquid pool, are\nconsidered to extensively validate the new method. Experimental data are used\nto cover a range of surface tension and viscosities. For all cases, the\nmodified VoF solver is observed to perform significantly better than original\nVoF method.", "category": "physics_flu-dyn" }, { "text": "Measuring Lagrangian accelerations using an instrumented particle: Accessing and characterizing a flow impose a number of constraints on the\nemployed measurement techniques; in particular optical methods require\ntransparent fluids and windows in the vessel. Whereas one can adapt apparatus,\nfluid and methods in the lab to these constraints, this is hardly possible for\nindustrial mixers. We present in this article a novel measurement technique\nwhich is suitable for opaque or granular flows: an instrumented particle, which\ncontinuously transmits the force/acceleration acting on it as it is advected in\na flow. Its density is adjustable for a wide range of fluids and because of its\nsmall size and its wireless data transmission, the system can be used both in\nindustrial and scientific mixers allowing a better understanding of the flow\nwithin. We demonstrate the capabilities and precision of the particle by\ncomparing its transmitted acceleration to alternative measurements, in\nparticular in the case of a turbulent von K\\'arm\\'an flow. Our technique shows\nto be an efficient and fast tool to characterize flows.", "category": "physics_flu-dyn" }, { "text": "Anomalous scaling of passive scalars in rotating flows: We present results of direct numerical simulations of passive scalar\nadvection and diffusion in turbulent rotating flows. Scaling laws and the\ndevelopment of anisotropy are studied in spectral space, and in real space\nusing an axisymmetric decomposition of velocity and passive scalar structure\nfunctions. The passive scalar is more anisotropic than the velocity field, and\nits power spectrum follows a spectral law consistent with $\\sim\nk_\\perp^{-3/2}$. This scaling is explained with phenomenological arguments that\nconsider the effect of rotation. Intermittency is characterized using scaling\nexponents and probability density functions of velocity and passive scalar\nincrements. In the presence of rotation, intermittency in the velocity field\ndecreases more noticeably than in the passive scalar. The scaling exponents\nshow good agreement with Kraichnan's prediction for passive scalar\nintermittency in two-dimensions, after correcting for the observed scaling of\nthe second order exponent.", "category": "physics_flu-dyn" }, { "text": "Investigation of G\u00f6rtler vortices in high-speed boundary layers via\n an efficient numerical solution to the non-linear boundary region equations: Streamwise vortices and the associated streaks evolve in boundary layers over\nflat or concave surfaces due to disturbances initiated upstream or triggered by\nthe wall surface. Following the transient growth phase, the fully-developed\nvortex structures become susceptible to inviscid secondary instabilities\nresulting in early transition to turbulence via `bursting' processes. In\nhigh-speed boundary layers, more complications arise due to compressibility and\nthermal effects, which become more significant for higher Mach numbers. In this\npaper, we study G\\\"{o}rtler vortices developing in high-speed boundary layers\nusing the boundary region equations (BRE) formalism, which we solve using an\nefficient numerical algorithm. Streaks are excited using a small transpiration\nvelocity at the wall. Our BRE-based algorithm is found to be superior to direct\nnumerical simulation (DNS) and ad-hoc nonlinear parabolized stability equation\n(PSE) models. BRE solutions are less computationally costly than a full DNS and\nhave a more rigorous theoretical foundation than PSE-based models. For example,\nthe full development of a G\\\"{o}rtler vortex system in high-speed boundary\nlayers can be predicted in a matter of minutes using a single processor via the\nBRE approach. This substantial reduction in calculation time is one of the\nmajor achievements of this work. We show, among other things, that it allows\ninvestigation into feedback control in reasonable total computational times. We\ninvestigate the development of the G\\\"{o}rtler vortex system via the BRE\nsolution with feedback control parametrically at various freestream Mach\nnumbers $M_\\infty$ and spanwise separations $\\lambda$ of the inflow\ndisturbances.", "category": "physics_flu-dyn" }, { "text": "On Secondary Tones Arising in Trailing-Edge Noise at Moderate Reynolds\n Numbers: Direct numerical simulations are carried out to investigate the flow features\nresponsible for secondary tones arising in trailing-edge noise at moderate\nReynolds numbers. Simulations are performed for a NACA 0012 airfoil at\nfreestream Mach numbers 0.1, 0.2 and 0.3 for angle of incidence 0 deg. and for\nMach number 0.3 at 3 deg. angle of incidence. The Reynolds number based on the\nairfoil chord is fixed at $Re_c=10^5$. Flow configurations are investigated\nwhere noise generation arises from the scattering of boundary layer\ninstabilities at the trailing edge. Results show that noise emission has a main\ntone with equidistant secondary tones, as discussed in literature. An\ninteresting feature of the present flows at zero incidence is shown; despite\nthe geometric symmetry, the flows become non-symmetric with a separation bubble\nonly on one side of the airfoil. A separation bubble is also observed for the\nnon-zero incidence flow. For both angles of incidence analyzed, it is shown\nthat low-frequency motion of the separation bubbles induce a frequency\nmodulation of the flow instabilities developed along the airfoil boundary\nlayer. When the airfoil is at 0 deg. angle of attack an intense amplitude\nmodulation is also observed in the flow quantities, resulting in a complex\nvortex interaction mechanism at the trailing edge. Both amplitude and frequency\nmodulations directly affect the velocity and pressure fluctuations that are\nscattered at the trailing edge, what leads to secondary tones in the acoustic\nradiation.", "category": "physics_flu-dyn" }, { "text": "A probabilistic, data-driven closure model for RANS simulations with\n aleatoric, model uncertainty: We propose a data-driven, closure model for Reynolds-averaged Navier-Stokes\n(RANS) simulations that incorporates aleatoric, model uncertainty. The proposed\nclosure consists of two parts. A parametric one, which utilizes previously\nproposed, neural-network-based tensor basis functions dependent on the rate of\nstrain and rotation tensor invariants. This is complemented by latent, random\nvariables which account for aleatoric model errors. A fully Bayesian\nformulation is proposed, combined with a sparsity-inducing prior in order to\nidentify regions in the problem domain where the parametric closure is\ninsufficient and where stochastic corrections to the Reynolds stress tensor are\nneeded. Training is performed using sparse, indirect data, such as mean\nvelocities and pressures, in contrast to the majority of alternatives that\nrequire direct Reynolds stress data. For inference and learning, a Stochastic\nVariational Inference scheme is employed, which is based on Monte Carlo\nestimates of the pertinent objective in conjunction with the reparametrization\ntrick. This necessitates derivatives of the output of the RANS solver, for\nwhich we developed an adjoint-based formulation. In this manner, the parametric\nsensitivities from the differentiable solver can be combined with the built-in,\nautomatic differentiation capability of the neural network library in order to\nenable an end-to-end differentiable framework. We demonstrate the capability of\nthe proposed model to produce accurate, probabilistic, predictive estimates for\nall flow quantities, even in regions where model errors are present, on a\nseparated flow in the backward-facing step benchmark problem.", "category": "physics_flu-dyn" }, { "text": "Experimental evidence of conformal invariance in soap film turbulent\n flows: We present experimental evidence of statistical conformal invariance in\nisocontours of fluid thickness in experiments of two-dimensional turbulence\nusing soap films. A Schlieren technique is used to visualize regions of the\nflow with constant film thickness, and association of isocontours with\nSchramm-L\\\"owner evolution (SLE) is used to identify conformal invariance. In\nexperiments where an inverse energy cascade develops, statistical evidence is\nconsistent with such an association. The diffusivity of the associated\none-dimensional Brownian process is close to 8/3, a value previously identified\nin isocontours of vorticity in high-resolution numerical simulations of\ntwo-dimensional turbulence (D. Bernard et al., Nature Phys. 2, 124, 2006). In\nexperiments where the inverse energy cascade is not sufficiently developed, no\nstatistical evidence of conformal invariance is found.", "category": "physics_flu-dyn" }, { "text": "Flow over an espresso cup: Inferring 3D velocity and pressure fields\n from tomographic background oriented schlieren videos via physics-informed\n neural networks: Tomographic background oriented schlieren (Tomo-BOS) imaging measures density\nor temperature fields in 3D using multiple camera BOS projections, and is\nparticularly useful for instantaneous flow visualizations of complex fluid\ndynamics problems. We propose a new method based on physics-informed neural\nnetworks (PINNs) to infer the full continuous 3D velocity and pressure fields\nfrom snapshots of 3D temperature fields obtained by Tomo-BOS imaging. PINNs\nseamlessly integrate the underlying physics of the observed fluid flow and the\nvisualization data, hence enabling the inference of latent quantities using\nlimited experimental data. In this hidden fluid mechanics paradigm, we train\nthe neural network by minimizing a loss function composed of a data mismatch\nterm and residual terms associated with the coupled Navier-Stokes and heat\ntransfer equations. We first quantify the accuracy of the proposed method based\non a 2D synthetic data set for buoyancy-driven flow, and subsequently apply it\nto the Tomo-BOS data set, where we are able to infer the instantaneous velocity\nand pressure fields of the flow over an espresso cup based only on the\ntemperature field provided by the Tomo-BOS imaging. Moreover, we conduct an\nindependent PIV experiment to validate the PINN inference for the unsteady\nvelocity field at a center plane. To explain the observed flow physics, we also\nperform systematic PINN simulations at different Reynolds and Richardson\nnumbers and quantify the variations in velocity and pressure fields. The\nresults in this paper indicate that the proposed deep learning technique can\nbecome a promising direction in experimental fluid mechanics.", "category": "physics_flu-dyn" }, { "text": "Effects of dilute coal char particle suspensions on propagating methane\n detonation wave: Methane/coal dust hybrid explosion is one of the common hazards in process\nand mining industries. In this study, methane detonation propagation in dilute\ncoal char particle suspensions is studied based on Eulerian-Lagrangian method.\nThe effects of char combustion on methane detonation dynamics are focused on.\nThe results show that propagation of the methane detonation wave in coal\nparticle suspensions are considerably affected by particle concentration and\nsize. Detonation extinction occurs when the coal particle size is small and\nconcentration is high. The averaged lead shock speed generally decreases with\nincreased particle concentration and decreased particle size. Mean structure\nand interphase coupling of hybrid detonation are analysed, based on the gas and\nparticle quantities. It is found that char combustion proceeds in the subsonic\nregion behind the detonation wave and heat release is relatively distributed\ncompared to that from gas phase reaction. The mass and energy transfer rates\nincrease rapidly to the maximum near the reaction front in the induction zone.\nMoreover, for 1 {\\mu}m particles, if the particle concentration is beyond a\nthreshold value, detonation re-initiation occurs after it is quenched at the\nbeginning of the coal dust suspensions. This is caused by hot spots from the\nshock focusing along the reaction front in a decoupled detonation and these\nshocks are generated from char combustion behind the lead shock.", "category": "physics_flu-dyn" }, { "text": "Finite volume based film flow and ice accretion models on aircraft wings: The thin runback water films driven by the gas flow, the pressure gradient\nand the gravity on the iced aircraft surface are investigated in this paper. A\nthree-dimensional film flow model based on Finite Volume Method (FVM) and the\nlubrication theory is proposed to describe the flow. The depth-averaged\nvelocity of the film is stored in Cartesian coordinates to avoid the appearance\nof the metric tensors. The governing equations are discretized in the first\nlayer structured grid cell which is selected as the grids for film flow. In\norder to verify this method, comparisons between numerical results and\nexperimental results of ice shapes on NACA 0012 airfoil and GLC-305 swept wing\nare presented, both showing a good agreement for rime and glaze ice condition.\nOverall, this model shows great potential to model ice accretion reasonably\nunder different icing conditions. Besides, the present method doesn't require\nanalytic metric terms, and can be easily coupled to existing finite volume\nsolvers for logically Cartesian meshes.", "category": "physics_flu-dyn" }, { "text": "Model-based deep reinforcement learning for accelerated learning from\n flow simulations: In recent years, deep reinforcement learning has emerged as a technique to\nsolve closed-loop flow control problems. Employing simulation-based\nenvironments in reinforcement learning enables a priori end-to-end optimization\nof the control system, provides a virtual testbed for safety-critical control\napplications, and allows to gain a deep understanding of the control\nmechanisms. While reinforcement learning has been applied successfully in a\nnumber of rather simple flow control benchmarks, a major bottleneck toward\nreal-world applications is the high computational cost and turnaround time of\nflow simulations. In this contribution, we demonstrate the benefits of\nmodel-based reinforcement learning for flow control applications. Specifically,\nwe optimize the policy by alternating between trajectories sampled from flow\nsimulations and trajectories sampled from an ensemble of environment models.\nThe model-based learning reduces the overall training time by up to $85\\%$ for\nthe fluidic pinball test case. Even larger savings are expected for more\ndemanding flow simulations.", "category": "physics_flu-dyn" }, { "text": "Robust learning from noisy, incomplete, high-dimensional experimental\n data via physically constrained symbolic regression: Machine learning offers an intriguing alternative to first-principles\nanalysis for discovering new physics from experimental data. However, to date,\npurely data-driven methods have only proven successful in uncovering physical\nlaws describing simple, low-dimensional systems with low levels of noise. Here\nwe demonstrate that combining a data-driven methodology with some general\nphysical principles enables discovery of a quantitatively accurate model of a\nnon-equilibrium spatially-extended system from high-dimensional data that is\nboth noisy and incomplete. We illustrate this using an experimental weakly\nturbulent fluid flow where only the velocity field is accessible. We also show\nthat this hybrid approach allows reconstruction of the inaccessible variables\n-- the pressure and forcing field driving the flow.", "category": "physics_flu-dyn" }, { "text": "Determination of energy flux rate in homogeneous ferrohydrodynamic\n turbulence using two-point statistics: In ferrofluids the suspended ferromagnetic particles agglomerate due to the\ninteraction between the particle magnetic moment and the external magnetic\nfield, which in turn, influences the turbulence and relaxation time. The\nrelaxation time becomes large when we are considering turbulence drag force\ninstead of viscous drag force in Brownian motion. We investigate that the total\nenergy conservation in ferrofluids taking into interaction between the external\nmagnetic field and the ferrofluid.", "category": "physics_flu-dyn" }, { "text": "Measurements and modeling of induced flow in collective vertical\n migration: Hydrodynamic interactions among swimming or flying organisms can lead to\ncomplex flows on the scale of the group. These emergent fluid dynamics are\noften more complex than a linear superposition of individual organism flows,\nespecially at intermediate Reynolds numbers. This paper presents an approach to\nestimate the flow induced by multiple swimmer wakes in proximity using an\nanalytical model that conserves mass and momentum in the aggregation. This\nanalytical model was informed by and validated with empirical measurements of\ninduced vertical migrations of brine shrimp, $\\textit{Artemia salina}$. The\nresponse of individual swimmers to ambient background flow and light intensity\nwas evaluated. In addition, the time-resolved three-dimensional spatial\nconfiguration of the swimmers was measured using a recently developed laser\nscanning system. Computational experiments using the analytical model found\nthat the induced flow at the front of the aggregation was insensitive to the\npresence of downstream swimmers, with the induced flow reaching an asymptote\nbeyond a threshold aggregation length. Closer swimmer spacing led to higher\ninduced flow, in some cases leading to model predictions of induced flow\nexceeding swimmer speeds required to maintain a stable spatial configuration.\nThis result was reconciled by comparing two different models for the near-wake\nof each swimmer. Our results demonstrate that aggregation-scale flows result\nfrom a complex, yet predictable interplay amongst organism-scale wake\nstructure, swimmer spacing and configuration, and aggregation size.", "category": "physics_flu-dyn" }, { "text": "Simultaneous measurements of deforming Hinze-scale bubbles with\n surrounding turbulence: We experimentally investigate the breakup mechanisms and probability of\nHinze-scale bubbles in turbulence. The Hinze scale is defined as the critical\nbubble size based on the critical mean Weber number, across which the bubble\nbreakup probability was believed to have an abrupt transition from being\ndominated by turbulence stresses to being suppressed completely by the surface\ntension. In this work, to quantify the breakup probability of bubbles with\nsizes close to the Hinze scale and to examine different breakup mechanisms,\nboth bubbles and their surrounding tracer particles were simultaneously\ntracked. From the experimental results, two Weber numbers, one calculated from\nthe slip velocity between the two phases and the other one acquired from local\nvelocity gradients, are separated and fitted with models that can be linked\nback to turbulence characteristics. Moreover, we also provide an empirical\nmodel to link bubble deformation to the two Weber numbers by extending the\nrelationship obtained from potential flow theory. The proposed relationship\nbetween bubble aspect ratio and the Weber numbers seems to work consistently\nwell for a range of bubble sizes. Furthermore, the time traces of the bubble\naspect ratio and the two Weber numbers are connected using the linear forced\noscillator model. Finally, having access to the distributions of these two\nWeber numbers provides a unique way to extract the breakup probability of\nbubbles with sizes close to the Hinze scale.", "category": "physics_flu-dyn" }, { "text": "A Primer on the Dynamical Systems Approach to Transport in Porous Media: Historically, the dominant conceptual paradigm of porous media flow, solute\nmixing and transport was based on steady two-dimensional flows in heterogeneous\nporous media. Although it is now well recognised that novel transport phenomena\ncan arise in unsteady and/or three-dimensional flows at both the pore- or\nDarcy-scales, appropriate methods for analysis and understanding of these more\ncomplex flows have not been widely employed. In this primer we advocate for\nmethods borrowed from dynamical systems (chaos) theory, which aim to uncover\nthe \\emph{Lagrangian kinematics} of these flows: namely how fluid particle\ntrajectories (which form a dynamical system) are organized and interact and the\nassociated impacts on solute transport and mixing. This dynamical systems\napproach to transport is inherently Lagrangian, and the Lagrangian kinematics\nform Lagrangian coherent structures (LCSs), special sets of trajectories that\ndivide the Lagrangian frame into chaotic mixing regions, poorly mixing hold-up\nregions (and in some cases non-mixing ``islands'') and the transport barriers\nthat organise these regions. Hence the dynamical systems approach provides\ninsights into flows that may exhibit chaotic, regular (non-chaotic) or mixed\nLagrangian kinematics, and also into how LCSs organize solute transport and\nmixing. Novel experimental methods are only recently permitting visualization\nof LCSs are in porous media flows. In this primer we review the dynamical\nsystems approach to porous media flow and transport and connect the associated\ntools and techniques with the latest research findings from pore to Darcy\nscales. This primer provides an introduction to the methods and tools of\ndynamical systems theory. Once familiar with these approaches, porous media\nresearchers will be better positioned to know when to expect complex Lagrangian\nkinematics, how to uncover and understand LCSs and their impacts ...", "category": "physics_flu-dyn" }, { "text": "Wake interaction of two disks falling in tandem: The fluid dynamics video illustrates the interaction of two disks falling in\ntandem at Reynolds number close to 100. Two fluorescent dyes were used to\nvisualize the wake of each body. We can observe that the trailing body\naccelerates thanks to the entrainment provided by the wake of the leading body\nand eventually catches up the leadind body. Then, thick disks\n(diameter/thickness = 3) lose their initial wakes, separate laterally and fall\nside by side. On the contrary, the wakes of thinner disks (d/t = 10) merge in a\nsingle wake and the bodies continue their fall together adopting a stable\nY-configuration.", "category": "physics_flu-dyn" }, { "text": "Coating-free Underwater Breathing via Biomimicry: Numerous natural and engineering scenarios necessitate entrapment of air\npockets or bubbles on submerged surfaces, e.g., aquatic insects, smartphones,\nand membranes for separation and purification. Current technologies for bubble\nentrapment rely heavily on perfluorocarbon coatings, which limits their\nsustainability and applications. Here, we investigate doubly reentrant\ncavities, a biomimetic microtexture capable of entrapping air under wetting\nliquids, under static and dynamic pressure cycling. The effects of positive,\nnegative, and positive-negative cycles are studied across a range of pressure\namplitudes, ramp rates, intercycle intervals, and water column heights.\nRemarkably, the fate of the trapped air under pressure cycling falls into the\nfollowing three distinct regimes: the bubble (i) monotonically depletes, (ii)\nremains indefinitely stable, or (iii) starts growing. This hitherto unrealized\nrichness of underwater bubble dynamics will guide the development of\ncoating-free underwater technologies and provide clues into the curious lives\nof air-breather aquatic/marine insects.", "category": "physics_flu-dyn" }, { "text": "Temperature-gradient induced massive augmentation of solute dispersion\n in viscoelastic micro-flows: Enhancing solute dispersion in electrically actuated flows has always been a\nchallenging proposition, as attributed to the inherent uniformity of the flow\nfield in absence of surface patterns. Over the years, researchers have focused\ntheir attention towards circumventing this limitation, by employing several\nfluidic and geometric modulations. However, the corresponding improvements in\nsolute dispersion often turn out to be inconsequential. Here we unveil that by\nexploiting the interplay between an externally imposed temperature gradient,\nsubsequent electrical charge redistribution and ionic motion, coupled with the\nrheological complexities of the fluid, one can achieve up to one order of\nmagnitude enhancement of solute dispersion in a pressure-driven flow of an\nelectrolyte solution. Our results demonstrate that the complex coupling between\nthermal, electrical, hydro-dynamic and rheological parameters over small\nscales, responsible for such exclusive phenomenon, can be utilitarian in\ndesigning novel thermally-actuated micro and bio-microfluidic devices with\nfavorable solute separation and dispersion characteristics.", "category": "physics_flu-dyn" }, { "text": "Spatio-temporal detection of Kelvin waves in quantum turbulence\n simulations: We present evidence of Kelvin excitations in space-time resolved spectra of\nnumerical simulations of quantum turbulence. Kelvin waves are transverse and\ncircularly polarized waves that propagate along quantized vortices, for which\nthe restitutive force is the tension of the vortex line, and which play an\nimportant role in theories of superfluid turbulence. We use the\nGross-Pitaevskii equation to model quantum flows, letting an initial array of\nwell-organized vortices develop into a turbulent bundle of intertwined vortex\nfilaments. By achieving high spatial and temporal resolution we are able to\ncalculate space-time resolved mass density and kinetic energy spectra. Evidence\nof Kelvin and sound waves is clear in both spectra. Identification of the waves\nallows us to extract the spatial spectrum of Kelvin waves, clarifying their\nrole in the transfer of energy", "category": "physics_flu-dyn" }, { "text": "Heating of the fuel mixture due to viscous stress ahead of accelerating\n flames in deflagration-to-detonation transition: The role of viscous stress in heating of the fuel mixture in\ndeflagration-to-detonation transition in tubes is studied both analytically and\nnumerically. The analytical theory is developed in the limit of low Mach\nnumber; it determines temperature distribution ahead of an accelerating flame\nwith maximum achieved at the walls. The heating effects of viscous stress and\nthe compression wave become comparable at sufficiently high values of the Mach\nnumber. In the case of relatively large Mach number, viscous heating is\ninvestigated by direct numerical simulations. The simulations were performed on\nthe basis of compressible Navier-Stokes gas-dynamic equations taking into\naccount chemical kinetics. In agreement with the theory, viscous stress makes\nheating and explosion of the fuel mixture preferential at the walls. The\nexplosion develops in an essentially multi-dimensional way, with fast\nspontaneous reaction spreading along the walls and pushing inclined shocks.\nEventually, the combination of explosive reaction and shocks evolves into\ndetonation.", "category": "physics_flu-dyn" }, { "text": "On the origin of impinging tones at low supersonic flow: Impinging compressible jets may cause deafness and material fatigue due to\nimmensely loud tonal noise. It is generally accepted that a feedback mechanism\nsimilar to the screech feedback loop is responsible for impinging tones. The\nclose of the loop remained unclear. One hypothesis hold up in the literature\nexplains the emanated sound with the direct interaction of vortices and the\nwall. Other explanations name the standoff shock oscillations as the origin of\nthe tones. Using direct numerical simulations (DNS) we were able to identify\nthe source mechanism for under-expanded impinging jets with a nozzle pressure\nratio (NPR) of 2.15 and a plate distance of 5 diameters. We found two different\ntypes of interactions between vortices and shocks to be responsible for the\ngeneration of the impinging tones. They are not related to screech.", "category": "physics_flu-dyn" }, { "text": "The origins of $k^{-2}$ spectrum in the decaying Taylor-Green\n magnetohydrodynamic turbulent flows: We investigate the origins of $k^{-2}$ spectrum in a decaying Taylor-Green\nmagnetohydrodynamic flow with zero large scale magnetic flux that was reported\nin Lee et al. (2010). A possible candidate for this scaling exponent has been\nthe weak turbulence phenomenology. From our numerical simulations, we observe\nthat current sheets in the magnetic Taylor-Green flow are formed in regions of\nmagnetic discontinuities. Based on this observation and by studying the\ninfluence of the current sheets on the energy spectrum, using a filtering\ntechnique, we argue that the discontinuities are responsible for the $-2$ power\nlaw scaling of the energy spectra of this flow.", "category": "physics_flu-dyn" }, { "text": "Highly parallelisable simulations of time-dependent viscoplastic fluid\n flow simulations with structured adaptive mesh refinement: We present the extension of an efficient and highly parallelisable framework\nfor incompressible fluid flow simulations to viscoplastic fluids. The system is\ngoverned by incompressible conservation of mass, the Cauchy momentum equation\nand a generalised Newtonian constitutive law. In order to simulate a wide range\nof viscoplastic fluids, we employ the Herschel-Bulkley model for yield-stress\nfluids with nonlinear stress-strain dependency above the yield limit. We\nutilise Papanastasiou regularisation in our algorithm to deal with the\nsingularity in apparent viscosity. The resulting system of partial differential\nequations is solved using the IAMR code (Incompressible Adaptive Mesh\nRefinement), which uses second-order Godunov methodology for the advective\nterms and semi-implicit diffusion in the context of an approximate projection\nmethod to solve on adaptively refined meshes. By augmenting the IAMR code with\nthe ability to simulate regularised Herschel-Bulkley fluids, we obtain\nefficient numerical software for time-dependent viscoplastic flow in three\ndimensions, which can be used to investigate systems not considered previously\ndue to computational expense. We validate results from simulations using this\nnew capability against previously published data for Bingham plastics and\npower-law fluids in the two-dimensional lid-driven cavity. In doing so, we\nexpand the range of Bingham and Reynolds numbers which have been considered in\nthe benchmark tests. Moreover, extensions to time-dependent flow of\nHerschel-Bulkley fluids and three spatial dimensions offer new insights into\nthe flow of viscoplastic fluids in this test case, and we provide missing\nbenchmark results for these extensions.", "category": "physics_flu-dyn" }, { "text": "Direct Numerical Simulation of the Moist Stably Stratified Surface\n Layer: Turbulence and Fog Formation: We investigate the effects of condensation and liquid water loading on the\nstably stratified surface layer, with an eye towards understanding the\ninfluence of turbulent mixing on fog formation. Direct numerical simulations\n(DNS) of dry and moist open channel flows are conducted, where in both a\nconstant cooling rate is applied at the ground to mimic longwave radiative\ncooling. Depending on the cooling rate, it can lead to either turbulent (weakly\nstable) or laminar (very stable) flows. Compared to the completely dry case,\nthe condensation of liquid water in the moist case enables slightly higher\ncooling rates to be achieved before leading to turbulence collapse. In the very\nstable cases, runaway cooling leads to the substantial condensation of liquid\nwater close to the ground and fog (visibility less than 1 km) results over much\nof the domain. In the weakly stable cases, turbulent mixing narrowly yields\nvisibilities of 1 km close to the ground over a similar time period. However,\ndespite the idealized nature of the system, the present results suggest that\nturbulence impedes, although will not necessarily inhibit, fog formation. A\npossible mechanism for fog formation within turbulent flows is identified,\nwherein regions of increased liquid water content form within the low-speed\nstreaks of the near-wall cycle. These streaks are energized in the moist cases\ndue to reduced dissipation of turbulence kinetic energy compared to the dry\ncase, although in both cases the streaks are less energetic and persistent than\nin neutrally stratified flow.", "category": "physics_flu-dyn" }, { "text": "Statistical and geometrical properties of thermal plumes in turbulent\n Rayleigh-B\u00e9nard convection: We present a systematic experimental study of geometric and statistical\nproperties of thermal plumes in turbulent Rayleigh-B\\'{e}nard convection using\nthe thermochromic-liquid-crystal (TLC) technique. The experiments were\nperformed in three water-filled cylindrical convection cells with aspect ratios\n2, 1, and 0.5 and over the Rayleigh-number range $5\\times10^7 \\leq Ra \\leq\n10^{11}$. TLC thermal images of horizontal plane cuts at various depths below\nthe top plate were acquired. Three-dimensional images of thermal plumes were\nthen reconstructed from the two-dimensional slices of the temperature field.\nThe results show that the often-called sheetlike plumes are really\none-dimensional structures and may be called rodlike plumes. We find that the\nnumber densities for both sheetlike/rodlike and mushroomlike plumes have\npower-law dependence on $Ra$ with scaling exponents of $\\sim 0.3$, which is\nclose to that between the Nusselt number $Nu$ and $Ra$. This result suggests\nthat it is the plume number that primarily d ermines the scaling exponent of\nthe $Nu$-$Ra$ scaling relation. The evolution of the aspect ratio of\nsheetlike/rodlike plumes reveals that as $Ra$ increases the plume geometry\nchanges from more-elongated to less-elongated. Our study of the plume area\nfraction (fraction of coverage over the surface of the plate) further reveals\nthat the increased plume numbers with $Ra$ mainly comes from increased plume\nemission, rather than fragmentation of plumes. In addition, the area,\nperimeter, and the shape complexity of the two-dimensional horizontal cuts of\nsheetlike/rodlike plumes were studied and all are found to obey log-normal\ndistributions.", "category": "physics_flu-dyn" }, { "text": "Turbulence suppression by streamwise-varying wall rotation in pipe flow: Direct numerical simulations (DNSs) of turbulent pipe flow subjected to\nstreamwise-varying wall rotation are performed. This control method is able to\nachieve drag reduction and even relaminarize the flow under certain control\nparameters at low Reynolds number. Two control parameters, which are velocity\namplitude and wavelength, are considered. An annular boundary layer, called\nSpatial Stokes Layer (SSL), is formed by the wall rotation. Based on the\nthickness of SSL, two types of drag reduction scenarios are identified. When\nthe thickness is low, the SSL acts as a spacer layer, obstructing the impact of\nfluids against the wall and hence reducing the shear stress. The flow\nstructures in the outer layer are less affected and are stretched in streamwise\ndirection due to the increased velocity gradient. Within the SSL, the\nturbulence intensity diminishes dramatically. When the thickness is large, a\nstreamwise wavy pattern of near-wall low-speed streaks is formed. The\norientation of streaks is found to lag behind the direction of local mean\nvelocity in phase and conform to the direction of local resultant shear stress.\nThe streamwise scale of near-wall flow structures is significantly reduced,\nresulting in the disruption of downstream development of flow structures and\nhence leading to the drag reduction. Besides, it is found that it requires both\nlarge enough thickness of SSL and velocity amplitude to relaminarize the\nturbulence. The relaminarization mechanism is that the annular SSL can\ncontinuously absorb energy from wall-normal stress due to the rotation effect,\nthereby the turbulence self-sustaining process could not be maintained. For the\nrelaminarization cases, the laminar state is stable to even extremely large\nperturbations, which possibly makes laminar state the only fixed point for the\nwhole system.", "category": "physics_flu-dyn" }, { "text": "Large-scale convective flow sustained by thermally active Lagrangian\n tracers: Non-isothermal particles suspended in a fluid lead to complex interactions --\nthe particles respond to changes in the fluid flow, which in turn is modified\nby their temperature anomaly. Here, we perform a novel proof-of-concept\nnumerical study based on tracer particles that are thermally coupled to the\nfluid. We imagine that particles can adjust their internal temperature reacting\nto some local fluid properties and follow simple, hard-wired active control\nprotocols. We study the case where instabilities are induced by switching the\nparticle temperature from hot to cold depending on whether it is ascending or\ndescending in the flow. A macroscopic transition from a stable to unstable\nconvective flow is achieved, depending on the number of active particles and\ntheir excess negative/positive temperature. The stable state is characterized\nby a flow with low turbulent kinetic energy, strongly stable temperature\ngradient, and no large-scale features. The convective state is characterized by\nhigher turbulent kinetic energy, self-sustaining large-scale convection, and\nweakly stable temperature gradients. The particles individually promote the\nformation of stable temperature gradients, while their aggregated effect\ninduces large-scale convection. When the Lagrangian temperature scale is small,\na weakly convective laminar system forms. The Lagrangian approach is also\ncompared to a uniform Eulerian bulk heating with the same mean injection\nprofile and no such transition is observed. Our empirical approach shows that\nthermal convection can be controlled by pure Lagrangian forcing and opens the\nway for other data-driven particle-based protocols to enhance or deplete\nlarge-scale motion in thermal flows.", "category": "physics_flu-dyn" }, { "text": "Corona splashing triggered by a loose monolayer of particles: In nature, high-speed raindrops often impact and spread on particulate\nsurfaces (e.g., soil, plant leaves with spores or pollen). We study the\ndynamics of droplet impact on a loosely packed monolayer of particles by\ncombining experimental and mathematical approaches. We find that the presence\nof mobile particles lowers the critical impact velocity at which the droplet\nexhibits corona splashing, as the particle area fraction is systematically\nincreased. We rationalize this experimental observation by considering the\njamming of frictional particles at the spreading rim. Elucidating the splashing\ntransition of the drop on a particulate bed can lead to a better understanding\nof soil loss and erosion from falling raindrops.", "category": "physics_flu-dyn" }, { "text": "Large eddy simulation of a low pressure turbine cascade with turbulent\n end wall boundary layers: We present first results of an implicit large eddy simulation of the MTU T161\nlow pressure turbine at a Reynolds number of 90,000 and Mach number of 0.6,\nboth based on isentropic exit conditions, using a high order discontinuous\nGalerkin method. The aim is to validate the numerical setup with respect to\navailable experimental data. We discuss the steps taken to create realistic\ninflow boundary conditions in terms of end wall boundary layer thickness and\nfree stream turbulence intensity. This is achieved by tailoring the input\ndistribution of Reynolds stresses and turbulent length scale to a Fourier\nseries based synthetic turbulence generator. Both blade loading and total\npressure losses at midspan show excellent agreement with the measurements.\nFollowing a short discussion of the secondary flow structures emerging due to\nthe interaction of the incoming boundary layer and the turbine blade, we show\nthat this simulation is also able to reproduce loss distribution behind the\nblade over the whole channel height.", "category": "physics_flu-dyn" }, { "text": "Theoretical study on the interfacial instability of a spherical droplet\n subject to vertical vibration: Interfacial instability would be aroused on a spherical liquid droplet when\nit is subject to external vertical vibration. In this paper, a linear analysis\nwas conducted on this instability problem. The polar-angle dependent\nacceleration in the spherical coordinate is strongly coupled with the temporal\nand spatial component of the surface deformation displacement, which gives a\nrecursion equation that implicitly expresses the dispersion relation between\nthe growth rate and spherical mode numbers. The unstable regions (or unstable\ntongues) for the inviscid fluids considering latitudinal mode (longitudinal\nmode number m = 0) were derived and presented in the parameter plane. Compared\nwith the solution of the spherical Faraday instability under radial vibration\nacceleration, the regions of harmonic unstable tongues for the mono-directional\nvibration case is much narrowed and the subharmonic unstable tongues almost\nbecome straight lines. The analysis shows that the latitudinal waves emerging\non the spherical droplet surface ought to oscillate harmonically instead of\nsubharmonically, which is opposite to the results for the case under radial\nvibration acceleration. A corresponding experiment of a liquid droplet lying on\na vertically vibrating plate was conducted and the observations substantiate\nour theoretical predictions.", "category": "physics_flu-dyn" }, { "text": "On the tuning of a wave-energy driven oscillating-water-column seawater\n pump to polychromatic waves: Performance of wave-energy devices of the oscillating water column (OWC) type\nis greatly enhanced when a resonant condition with the forcing waves is\nmaintained. The natural frequency of such systems can in general be tuned to\nresonate with a given wave forcing frequency. In this paper we address the\ntuning of an OWC sea-water pump to polychromatic waves. We report results of\nwave tank experiments, which were conducted with a scale model of the pump.\nAlso, a numerical solution for the pump equations, which were proven in\nprevious work to successfully describe its behavior when driven by\nmonochromatic waves, is tested with various polychromatic wave spectra. Results\nof the numerical model forced by the wave trains measured in the wave tank\nexperiments are used to develop a tuning criterion for the sea-water pump.", "category": "physics_flu-dyn" }, { "text": "A brief introduction to bulk viscosity of fluids: Fluid flows are typically studied by solving the Navier--Stokes equation. One\nof the fundamental assumptions of this equation is Stokes' hypothesis. This\nhypothesis assumes bulk viscosity, to be identically zero. The Stokes'\nhypothesis is a reasonable approximation for commonly observed fluid flows;\ntherefore, Navier--Stokes equation gives satisfactory results in these\nsituations. However, there are circumstances where this hypothesis does not\nhold good, and therefore, the classical Navier--Stokes equation becomes inapt.\nThese situations include the absorption of sound waves, hypersonic flows,\nturbulent flows, and flow of Martian air etc. Reliable analytical and\ncomputational studies of these flows requires account of bulk viscosity in the\ngoverning equations. In this article, we provide a brief review of the subject\nof bulk viscosity. We start with a brief background of this topic. Then we\ndiscuss the underlying microscopic mechanisms that give rise to bulk viscosity\neffects. It was followed by a review of methods available in the literature for\nestimation of this parameter. Finally, a review of the studies that analyze the\neffects of bulk viscosity in various fluid flows is provided.", "category": "physics_flu-dyn" }, { "text": "Eddy diffusivities of inertial particles in random Gaussian flows: We investigate the large-scale transport of inertial particles. We derive\nexplicit analytic expressions for the eddy diffusivities for generic Stokes\ntimes. These latter expressions are exact for any shear flow while they\ncorrespond to the leading contribution either in the deviation from the shear\nflow geometry or in the P\\'eclet number of general random Gaussian velocity\nfields. Our explicit expressions allow us to investigate the role of inertia\nfor such a class of flows and to make exact links with the analogous transport\nproblem for tracer particles.", "category": "physics_flu-dyn" }, { "text": "Aerothermodynamic Analysis of Faceted Aeroshell at Hypersonic Speed: This study explores the aerothermal behaviour of a rigid mechanically\ndeployable aeroshell developed at Imperial College London for high payload\natmospheric entry missions. The multiphysics CFD software STAR-CCM+ is used to\nperform a Conjugate Heat Transfer analysis on the aeroshell's faceted geometry.\nResults are presented for four different geometry models tested in air at Mach\n5 with angles of attack 0{\\deg}, 5{\\deg} and 10{\\deg}. The predicted surface\nheat transfer reveals areas of elevated heat loads at the ribs between facets\nand at the aeroshell shoulder, due to local boundary layer thinning. The\nincrease in heat transfer at the ribs depends on the sharpness of the rib: more\nrounded shapes result in lower heat fluxes. Comparison with high-speed wind\ntunnel tests shows good agreement with experimental data. Stanton number and\ntemperature profiles agree within 8% and 2%, respectively. The discrepancies\nbetween experiments and simulations are largest at the sharp ribs of the\naeroshell. The sources of error can be associated with three-dimensional\neffects neglected in the heat flux derivations from temperature measurements as\nwell as experimental uncertainties.", "category": "physics_flu-dyn" }, { "text": "Flow through randomly curved manifolds: We have found that the relation between the flow through campylotic\n(generically curved) media, consisting of randomly located curvature\nperturbations, and the average Ricci scalar of the system exhibits two distinct\nfunctional expressions (hysteresis), depending on whether the typical spatial\nextent of the curvature perturbation lies above or below the critical value\nmaximizing the overall Ricci curvature. Furthermore, the flow through such\nsystems as a function of the number of curvature perturbations presents a\nsublinear behavior for large concentrations due to the interference between\ncurvature perturbations that, consequently, produces a less curved space. For\nthe purpose of this study, we have developed and validated a lattice kinetic\nmodel capable of describing fluid flow in arbitrarily curved manifolds, which\nallows to deal with highly complex spaces in a very compact and efficient way.", "category": "physics_flu-dyn" }, { "text": "Modeling and Simulation of the Evaporation and Drying of a Two Component\n Slurry Droplet: In this paper, we present a mathematical model and numerical simulation of\nthe evaporation and drying process of a liquid droplet containing suspended\nsolids. This type of drying is commonly encountered in manufacturing processes\nsuch as spray drying and spray pyrolysis, which have applications in industries\nsuch as food and pharmaceuticals. The proposed model consists of three stages.\nIn the first stage, we consider the evaporation of the liquid in the presence\nof solid particles. The second stage involves the formation of a porous crust\naround a wet core region, with liquid evaporation occurring through the crust\nlayer. Finally, the third stage involves sensible heating of the dry particle\nto reach ambient temperature. To solve the physical models governing these\nprocesses, we use a finite difference method with a moving grid methodology.\nThis allows us to account for the moving interface between the crust and the\nwet core region of the droplet. In this study, we use a non-uniform temperature\nmodel that takes into account the spatial variation of temperature inside the\ndroplet. We also assess the validity of a uniform temperature model. To\nvalidate our model, we compare it with experimental data on the drying of a\nsingle droplet containing colloidal silica particles. We find that our model\nagrees well with the experimental results. We rigorously examine assumptions\nmade in the model, such as the shape of the solid particles and the continuum\nflow of vapor through the porous crust. In addition, we analyze the effects of\ndrying conditions, such as the velocity, temperature, relative humidity, and\nconcentration of solid particles, on the drying rate and the final morphology\nof the particle. Finally, we develop a regime map that can be used to determine\nwhether the final particle will be solid or hollow, based on the operating\nconditions.", "category": "physics_flu-dyn" }, { "text": "Life and Death of a Thin Liquid Film: Thin films, bubbles and membranes are central to numerous natural and\nengineering processes, i.e., in thin-film solar cells, coatings, biosensors,\nelectrowetting displays, foams, and emulsions. Yet, the characterization and an\nadequate understanding of their rupture is limited by the scarcity of atomic\ndetail. We present here the complete life-cycle of freely suspended films using\nnon-equilibrium molecular dynamics simulations of a simple atomic fluid free of\nsurfactants and surface impurities, thus isolating the fundamental rupture\nmechanisms. Counter to the conventional notion that rupture occurs randomly, we\ndiscovered a short-term 'memory' by rewinding in time from a rupture event,\nextracting deterministic behaviors from apparent stochasticity. A comprehensive\ninvestigation of the key rupture-stages including both unrestrained and\nfrustrated propagation is made - characterization of the latter leads to a\nfirst-order correction to the classical film-retraction theory. Furthermore,\nthe highly resolved time window reveals that the different modes of the\nmorphological development, typically characterized as heterogeneous nucleation\nand spinodal decomposition, continuously evolve seamlessly with time from one\ninto the other.", "category": "physics_flu-dyn" }, { "text": "Estimating mean profiles and fluxes in high-speed turbulent boundary\n layers using inner/outer-layer transformations: Accurately predicting drag and heat transfer for compressible high-speed\nflows is of utmost importance for a range of engineering applications. This\nrequires the precise knowledge of the entire velocity and temperature profiles.\nA common approach is to use compressible velocity scaling laws\n(transformation), that inverse transform the velocity profile of an\nincompressible flow, together with a temperature-velocity relation. In this\nNote, we use distinct velocity transformations for the inner and outer layers.\nIn the inner layer, we utilize a recently proposed scaling law that\nappropriately incorporates variable property and intrinsic compressibility\neffects, while the outer layer profile is inverse-transformed with the\nwell-known Van Driest transformation. The result is an analytical expression\nfor the mean shear valid in the entire boundary layer, which combined with a\ntemperature-velocity relationship, provides predictions of mean velocity and\ntemperature profiles at unprecedented accuracy. Using these profiles, drag and\nheat transfer is evaluated with an accuracy of +/-4% and +/-8%, respectively,\nfor a wide range of compressible turbulent boundary layers up to Mach numbers\nof 14.", "category": "physics_flu-dyn" }, { "text": "Electro-osmotic flow through a nanopore: Electroosmotic pumping of fluid through a nanopore that traverses an\ninsulating membrane is considered. The density of surface charge on the\nmembrane is assumed uniform, and sufficiently low for the Poisson-Boltzmann\nequation to be linearized. The reciprocal theorem gives the flow rate generated\nby an applied weak electric field, expressed as an integral over the fluid\nvolume. For a circular hole in a membrane of zero thickness, an analytical\nresult is possible up to quadrature. For a membrane of arbitrary thickness, the\nfull Poisson--Nernst--Planck--Stokes system of equations is solved numerically\nusing a finite volume method. The numerical solution agrees with the standard\nanalytical result for electro-osmotic flux through a long cylindrical pore when\nthe membrane thickness is large compared to the hole diameter. When the\nmembrane thickness is small, the flow rate agrees with that calculated using\nthe reciprocal theorem.", "category": "physics_flu-dyn" }, { "text": "Analysis of chaotic flow in a 2D multi-turn closed-loop pulsating heat\n pipe: Numerical study has been conducted for the chaotic flow in a multi-turn\nclosed-loop pulsating heat pipe (PHP). Heat flux and constant temperature\nboundary conditions have been applied for heating and cooling sections\nrespectively. Water was used as working fluid. Volume of Fluid (VOF) method has\nbeen employed for two-phase flow simulation. Volume fraction results showed\nformation of perfect vapor and liquid plugs in the fluid flow of PHP.\nNon-linear time series analysis, power spectrum density, correlation dimension\nand autocorrelation function were used to investigate the chaos. Absence of\ndominating peaks in the power spectrum density was a signature of chaos in the\npulsating heat pipe. It was found that by increasing the filling ratio and\nevaporator heating power the correlation dimension increases. Decreasing of the\nautocorrelation function with respect to time showed the prediction ability is\nfinite as a result of chaotic state. An optimal filling ratio of 60% and\nminimum thermal resistance of 1.62 K/W were found for better thermal\nperformance of the pulsating heat pipe.", "category": "physics_flu-dyn" }, { "text": "Dynamics of flags over wide ranges of mass and bending stiffness: There have been many studies of the instability of a flexible plate or flag\nto flapping motions, and of large-amplitude flapping. Here we use inviscid\nsimulations and a linearized model to study more generally how key quantities\n-- mode number (or wavenumber), frequency, and amplitude -- depend on the two\ndimensionless parameters, flag mass and bending stiffness. In the limit of\nsmall flag mass, flags perform traveling wave motions that move at nearly the\nspeed of the oncoming flow. The flag mode number scales as the -1/4 power of\nbending stiffness. The flapping frequency has the same scaling, with an\nadditional slight increase with flag mass in the small-mass regime. The\nflapping amplitude scales approximately as flag mass to the 1/2 power. For\nlarge flag mass, the dominant mode number is low (0 or 1), the flapping\nfrequency tends to zero, and the amplitude saturates in the neighborhood of its\nupper limit (the flag length). In a linearized model, the fastest growing modes\nhave somewhat different power law scalings for wavenumber and frequency. We\ndiscuss how the numerical scalings are consistent with a weakly nonlinear\nmodel.", "category": "physics_flu-dyn" }, { "text": "Influence of initial conditions on coherent structures in a round jet: Coherent structures play an important role in the near-field of shear flows\nsuch as round jets. In the following the near-field of a round jet is\nnumerically investigated , this for different initial conditions. Two\nconfigurations are considered one with a fully developed turbulence at the\ninlet, the other with an under developed turbulence. Calculations are carried\nout with a k-epsilon turbulence model but different sets of values for the\nconstants inducing different level of diffusion within the flow. Finally, the\nresults show that the presence of coherent structures in the near-field depends\nstrongly on the turbulence level at the inlet. For a fully developed turbulence\nthese structures are missing or very weak which is not the case for an under\ndeveloped turbulence.", "category": "physics_flu-dyn" }, { "text": "CoNFiLD: Conditional Neural Field Latent Diffusion Model Generating\n Spatiotemporal Turbulence: This study introduces the Conditional Neural Field Latent Diffusion (CoNFiLD)\nmodel, a novel generative learning framework designed for rapid simulation of\nintricate spatiotemporal dynamics in chaotic and turbulent systems within\nthree-dimensional irregular domains. Traditional eddy-resolved numerical\nsimulations, despite offering detailed flow predictions, encounter significant\nlimitations due to their extensive computational demands, restricting their\napplications in broader engineering contexts. In contrast, deep learning-based\nsurrogate models promise efficient, data-driven solutions. However, their\neffectiveness is often compromised by a reliance on deterministic frameworks,\nwhich fall short in accurately capturing the chaotic and stochastic nature of\nturbulence. The CoNFiLD model addresses these challenges by synergistically\nintegrating conditional neural field encoding with latent diffusion processes,\nenabling the memory-efficient and robust probabilistic generation of\nspatiotemporal turbulence under varied conditions. Leveraging Bayesian\nconditional sampling, the model can seamlessly adapt to a diverse range of\nturbulence generation scenarios without the necessity for retraining, covering\napplications from zero-shot full-field flow reconstruction using sparse sensor\nmeasurements to super-resolution generation and spatiotemporal flow data\nrestoration. Comprehensive numerical experiments across a variety of\ninhomogeneous, anisotropic turbulent flows with irregular geometries have been\nconducted to evaluate the model's versatility and efficacy, showcasing its\ntransformative potential in the domain of turbulence generation and the broader\nmodeling of spatiotemporal dynamics.", "category": "physics_flu-dyn" }, { "text": "Entrainment in dry and moist thermals: We study entrainment in dry thermals in neutrally and unstably stratified\nambients, and moist thermals in dry-neutrally stratified ambients using direct\nnumerical simulations (DNS). We find, in agreement with results of Lecoanet and\nJeevanjee \\tb{[J. Atmos. Sci. 76(12), 3785-3801, (2019)]} that turbulence plays\na minor role in entrainment in dry thermals in a neutral ambient for Reynolds\nnumbers $Re \\lesssim 10^4$. We then show that the net entrainment rate\nincreases when the buoyancy of the thermals increases, either by condensation\nheating or because of an unstably stratified ambient. This is in contrast with\nthe findings of Morrison et al. [J. Atmos. Sci. 78(3), 797-816, (2021)]. We\nalso show that the role of turbulence is greater in these cases than in dry\nthermals and, significantly, that the combined action of condensation heating\nand turbulence creates intense small scale vorticity, destroying the {coherent}\nvortex ring that is seen in dry and moist laminar thermals. These findings\nsuggest that fully resolved simulations at Reynolds numbers significantly\nlarger than the mixing transition Reynolds number $Re=10^4$ are necessary to\nunderstand the role of turbulence in the entrainment in growing cumulus clouds,\nwhich consist of a series of thermals rising and decaying in succession.", "category": "physics_flu-dyn" }, { "text": "Irreversible energy extraction from negative temperature two-dimensional\n turbulence: The formation and transition of patterns of two-dimensional turbulent flows\nobserved in various geophysical systems are commonly explained in terms of\nstatistical mechanics. Different from ordinary systems, for a two-dimensional\nflow, the absolute temperature defined for a statistical equilibrium can take\nnegative values. In a state of negative temperature, the second law of\nthermodynamics predicts that energy in microscopic fluctuations is irreversibly\nconverted to a macroscopic form. This study explores the possibility of this\none-way energy conversion in a two-dimensional flow using a basic conceptual\nmodel. We consider an inviscid incompressible fluid contained in a bounded\ndomain, the shape of which is distorted by an externally imposed force. Unlike\nthe usual fixed boundary cases, the flow energy within the domain is exchanged\nwith the external system via pressure work through the moving lateral boundary.\nConcurrently, the flow field remains constrained by vorticity conservation.\nBeginning from a state of Kraichnan's grand-canonical ensemble, when the domain\nshape is distorted from one shape to another in a finite time, the Jarzynski\nequality is established. This equality states that, on average, the direction\nof a net energy flow through the boundary during a cycle of domain distortion\nchanges with the sign of the initial temperature of the system. Numerical\nexperiments are carried out to verify this theoretical argument and to\ninvestigate the parameter dependence of the energy exchange rate.", "category": "physics_flu-dyn" }, { "text": "Interaction of Shock Train with Cavity Shear Layer in a Scramjet\n Isolator: The interaction between the self-excited shock train flow and the cavity\nshear layer in a scramjet isolator is investigated numerically using\ndetached-eddy simulations (DES). The effect of changing the position of the\nshock train by controlling the back pressure ratio and the effect of changing\nthe cavity front wall angle are analyzed using unsteady statistics and modal\nanalysis. The propagation mechanism of the pressure disturbance was\ninvestigated by spatiotemporal cross-correlation coefficient analysis. In the\npresent numerical study, a constant isolator section with a cavity front wall\nwas considered, followed by a diffuser section simulated at Mach number 2.2\nwith three different back pressure ratios. The change in back pressure provides\nthree different conditions. To understand the unsteady dynamics of the\ninteraction of the shear layer with the shock train, the spatiotemporal\ntrajectory of the wall pressure and the centerline pressure distribution, the\nspatiotemporal cross-correlation coefficient, and the modal analysis by dynamic\nmode decomposition are obtained. The results show that the low-frequency shock\ntrain oscillation dominates the cavity oscillation. The spatiotemporal\ncross-correlation between the wall surface and the cavity bottom wall indicates\nthe propagation of local disturbances originating from the separated boundary\nlayer caused by the shock and the recirculation zone in the corners of the\ncavity. Dynamic mode decomposition analysis shows the shear layer at the\nleading edge of the cavity and the downstream propagation of large eddies from\nthe cavity. It also shows the pairing of coherent structures between the shock\ntrain and the recirculation zone of the cavity.", "category": "physics_flu-dyn" }, { "text": "Inverse kinetic theory for incompressible thermofluids: An interesting issue in fluid dynamics is represented by the possible\nexistence of inverse kinetic theories (IKT) which are able to deliver, in a\nsuitable sense, the complete set of fluid equations which are associated to a\nprescribed fluid. From the mathematical viewpoint this involves the formal\ndescription of a fluid by means of a classical dynamical system which advances\nin time the relevant fluid fields. The possibility of defining an IKT for the\n3D incompressible Navier-Stokes equations (INSE), recently investigated (Ellero\n\\textit{et al}, 2004-2007) raises the interesting question whether the theory\ncan be applied also to thermofluids, in such a way to satisfy also the second\nprinciple of thermodynamics. The goal of this paper is to prove that such a\ngeneralization is actually possible, by means of a suitable \\textit{extended\nphase-space formulation}. We consider, as a reference test, the case of\nnon-isentropic incompressible thermofluids, whose dynamics is described by the\nFourier and the incompressible Navier-Stokes equations, the latter subject to\nthe conditions of validity of the Boussinesq approximation.", "category": "physics_flu-dyn" }, { "text": "Theoretical model of a finite force at the moving contact line: In theoretical analyses of the moving contact line, an infinite force along\nthe solid wall has been reported based off the non-integrable stress along a\nsingle interface. In this investigation we demonstrate that the stress\nsingularity is integrable and results in a finite force at the moving contact\nline if the contact line is treated as a one-dimensional manifold and all three\ninterfaces that make up the moving contact line are taken into consideration.\nThis is due to the dipole nature of the vorticity and pressure distribution\naround the moving contact line. Mathematically, this finite force is determined\nby summing all the forces that act over an infinitesimally small cylindrical\ncontrol volume that encloses the entire moving contact line. With this finite\nforce, we propose a new dynamic Young's equation for microscopic dynamic\ncontact angle that is a function of known parameters only, specifically the\ninterface velocity, surface tension, and fluid viscosity. We combine our model\nwith Cox's model for apparent dynamic contact angle and find good agreement\nwith published dynamic contact angle measurements.", "category": "physics_flu-dyn" }, { "text": "Numerical evaluation of relative permeability using\n Johnson--Koplik--Dashen model: We present a numerical study aimed at comparing two approaches to the\nevaluation of relative permeability curves from 3D binary images of porous\nmedia. One approach hinges on the numerical solution of Stokes equations, while\nthe other is based on the Johnson-Koplik-Dashen (JKD) universal scaling theory\nof viscous frequency-dependent flow [D.~L. Johnson, J.~Koplik, and R.~Dashen,\n\\emph{Theory of dynamic permeability and tortuosity in fluid--saturated porous\nmedia}, Journal of Fluid Mechanics \\textbf{176} (1987), 379--402.] and the\nmethod of maximal inscribed spheres.\n JKD steady-flow simulations only require the solution of a boundary-value\nproblem for the Laplace equation, which is computationally less intensive than\nthe solution of Stokes equations.\n A series of numerical calculations performed on 3D pore-space images of\nnatural rock demonstrate that JKD-based estimates are in good agreement with\nthe corresponding Stokes-flow numerical simulations.", "category": "physics_flu-dyn" }, { "text": "On the stabilizing mechanism of 2D absolute and global instabilities by\n 3D streaks: Global and local absolute instabilities of 2D wakes are known to be\nstabilized by spanwise periodic modulations of the wake profile. The present\nstudy shows that this stabilizing effect is of general nature and can be\nmimicked by enforcing spanwise periodic modulations of the wave advection\nvelocity in the generalized complex Ginzburg-Landau equation. The first order\nsensitivity of the absolute and global growth rate to the enforced modulation\nis zero, exactly as in the Navier-Stokes case. We show that a second order\nsensitivity analysis is effective to quantify and interpret the observed\nstabilizing effect. The global growth rates predicted by the second order\nexpansion closely match those issued from a direct computation of the\neigenvalues. It is shown that, at leading order, the modulation of the wave\nadvection velocity alters the effective wave diffusion coefficient in the\ndispersion relation and that this variation induces a reduction of the absolute\ngrowth rate and the stabilization of the global instability.", "category": "physics_flu-dyn" }, { "text": "Re-orienting the Turbulent flow over an Inclined Cylinder of Finite\n Aspect ratio: Turbulent flow around a swept back circular cylinder was investigated\nexperimentally using one or two small sweeping jets emanating tangentially to\nthe surface and orthogonal to the axis of the cylinder creating a yawing moment\nthat overcomes the natural restoring force to the plane of symmetry. It appears\nfrom the integral force balance data that large yawing moment coefficients can\nbe generated in this manner, and the results could be used in the orientation\nand attitude control of a refueling boom thus avoiding the H-shape control\nsurfaces that are currently used on Air Force tankers. The interaction between\nthe jet and the flow in the lee of the cylinder was mapped using surface oil\nflow visualization and 2D-Particle Image velocimetry technique. The sensitivity\nof the interaction and its outcome to the change in the azimuthal location of\nthe actuators was investigated.", "category": "physics_flu-dyn" }, { "text": "Stationary Statistics of Turbulence as an Attractor: A calculational approach in fluid turbulence is presented. Use is made of the\nattracting nature of the fluid-dynamic dynamical system. An approach is offered\nthat effectively propagates the statistics in time. Loss of sensitivity to an\ninitial probability density functional and generation of stationary statistical\neffects is speculated.", "category": "physics_flu-dyn" }, { "text": "Rough or Wiggly? Membrane Topology and Morphology for Fouling Control: Reverse Osmosis Membrane (ROM) filtration systems are widely applied in\nwastewater recovery, seawater desalination, landfill water treatment, etc.\nDuring filtration, the system performance is dramatically affected by membrane\nfouling which causes a significant decrease in permeate flux as well as an\nincrease in the energy input required to operate the system. Design and\noptimization of ROM filtration systems aim at reducing membrane fouling by\nstudying the coupling between membrane structure, local flow field, local\nsolute concentration and foulant adsorption patterns. Yet, current studies\nfocus exclusively on oversimplified steady-state models that ignore any dynamic\ncoupling between the fluid dynamics and the transport through the membrane,\nwhile membrane design still proceeds through trials and errors. In this work,\nwe develop a model that couples the transient Navier-Stokes and the\nAdvection-Diffusion-Equations, as well as an adsorption-desorption equation for\nthe foulant accumulation, and we validate it against unsteady measurements of\npermeate flux as well as steady-state spatial fouling patterns. Furthermore, we\nanalytically show that, for a straight channel, a universal scaling\nrelationship exists between the Sherwood and Bejam numbers. We then generalize\nthis result to membranes subject to morphological and/or topological\nmodifications. We demonstrate that universal scaling behavior can be identified\nthrough the definition of a modified Reynolds number, that accounts for the\nadditional length scales introduced by the membrane modifications, and a\nmembrane performance index, which represents an aggregate efficiency measure\nwith respect to both clean permeate flux and energy input required to operate\nthe system.", "category": "physics_flu-dyn" }, { "text": "Near-wall approximations to speed up simulations for atmosphere boundary\n layers in the presence of forests using lattice Boltzmann method on GPU: Forests play an important role in influencing the wind resource in\natmospheric boundary layers and the fatigue life of wind turbines. Due to\nturbulence, a difficulty in the simulation of the forest effects is that flow\nstatistical and fluctuating content should be accurately resolved using a\nturbulence-resolved CFD method, which requires a large amount of computing time\nand resources. In this paper, we demonstrate a fast but accurate simulation\nplatform that uses a lattice Boltzmann method with large eddy simulation on\nGraphic Processing Units (GPU). The simulation tool is the open-source program,\nGASCANS, developed at the University of Manchester. The simulation platform is\nvalidated based on canonical wall-bounded turbulent flows. A forest is modelled\nin the form of body forces injected near the wall. Since a uniform cell size is\napplied throughout the computational domain, the averaged first-layer cell\nheight over the wall reaches to $\\langle \\Delta y^+\\rangle = 165$. Simulation\nresults agree well with previous experiments and numerical data obtained from\nfinite volume methods. We demonstrate that good results are possible without\nthe use of a wall-function, since the forest forces overwhelm wall friction.\nThis is shown to hold as long as the forest region is resolved with several\ncells. In addition to the GPU speedup, the approximations also significantly\nbenefit the computation efficiency.", "category": "physics_flu-dyn" }, { "text": "Energy transfer in turbulent channel flows and implications for\n resolvent modelling: We analyse the inter-scale transfer of energy for two types of plane\nPoiseuille flow: the P4U exact coherent state of Park and Graham (2015) and\nturbulent flow in a minimal channel. For both flows, the dominant\nenergy-producing modes are streamwise-constant streaks with a spanwise spacing\nof approximately 100 wall units. Since the viscous dissipation for these scales\nis not sufficient to balance production, the nonlinear terms redistribute the\nexcess energy to other scales. Spanwise-constant scales (that is,\nTollmien-Schlichting-like modes with zero spanwise wavenumber), in particular,\naccount for a significant amount of net energy gain from the nonlinear terms.\nWe compare the energy balance to predictions from resolvent analysis and we\nshow that it does not model energy transfer well. Nevertheless, we find that\nthe energy transferred from the streamwise-constant streaks can be predicted\nreasonably well by a Cess eddy viscosity profile. As such, eddy viscosity is an\neffective model for the nonlinear terms in resolvent analysis and explains good\npredictions for the most energetic streamwise-constant streaks. It also\nimproves resolvent modes as a basis for structures whose streamwise lengths are\ngreater than their spanwise widths by counteracting non-normality of the\nresolvent operator. Eddy viscosity does not respect the conservative nature of\nthe nonlinear energy transfer which must sum to zero over all scales. It is\nless effective, consequently, for scales which receive energy from the\nnonlinear terms.", "category": "physics_flu-dyn" }, { "text": "Wave turbulence: A solvable problem applied to Navier-Stokes: Wave turbulence and eddy turbulence are the two regimes that we may encounter\nin nature. The attention of fluid mechanics being mainly focused on\nincompressible hydrodynamics, it is usually the second regime that is treated\nin books, whereas waves are often present in geophysics and astrophysics. In\nthese lecture notes, I present the theory of wave turbulence which is free from\nthe closure problem encountered in eddy turbulence. Basically, the wave\namplitude is introduced in a multiple time scale method as a small parameter to\nderive the so-called kinetic equations from which exact results can be obtained\n(power-law spectra, direction of the cascade, Kolmogorov's constant) and\ncompared with the data. Two hydrodynamic applications are considered with\ncapillary waves and inertial waves, the first leading to isotropic turbulence\nand the second to anisotropic turbulence.", "category": "physics_flu-dyn" }, { "text": "A visual study on the spray of gas-liquid atomizer: A Visual investigation of spray cone angle for different air-blast atomizers\nand flow conditions is described. Liquid jets are exposed to high gas stream\nwith specific relative angle. Using high speed camera, spray cone angle over a\nrange of Reynolds number 4x104 to 9x104 and Weber number 1 to 140 is studied,\nfollowed by laser-based diagnosis of particle distribution and Sauter mean\ndiameter. The results show that for high Reynolds and Weber number, the cone\nangle is independent of flow condition, and is only dependent on geometry of\natomizer, including orifice diameter with significant effect on cone angle,\nSauter mean diameter and particle distribution", "category": "physics_flu-dyn" }, { "text": "Delaying Leading Edge Vortex Detachment by Plasma Flow Control at\n Topologically Critical Locations: Flapping wing propulsion offers unrivalled manoeuvrability and efficiency at\nlow flight speeds and in hover. These advantages are attributed to the leading\nedge vortex developing on an unsteady wing, which induces additional lift. We\npropose and validate a manipulation hypothesis that allows prolongation of the\nleading edge vortex growth phase, by delaying its detachment with the aid of\nflow control. This approach targets an overall lift increase on unsteady\nairfoils. A dielectric barrier discharge plasma actuator is successfully used\nto compress secondary structures upstream of the main vortex on a pitching and\nplunging flat plate. To determine flow control timing and location, the\ntangential velocity on the airfoil surface is used, which is also used to\nquantify topological effects of flow control. This flow control is then tested\nfor different motion kinematics and on a NACA 0012 airfoil. Significant\nincrease of the peak circulation of the leading edge vortex of about 20% for\nall cases with flow control indicates that this approach is applicable for\nvarious kinematics, dynamics and airfoil types.", "category": "physics_flu-dyn" }, { "text": "Electrical Potential, Mass Transport and Velocity Distribution of\n Electro-osmotic Flow in a Nanochannel by Incorporating the Variation of\n Dielectric Constant of Aqueous Electrolyte Solution: We consider a coupled system of Navier Stokes, Maxwell Stefan and Poisson\nBoltzmann equations by incorporating the variation of dielectric constant,\nwhich governs the electro osmotic flow in nano channel, describing the\nevolution of the velocity, concentration and potential fields of dissolved\nconstituents in an aqueous electrolyte solution. We apply the finite difference\ntechnique to solve one and two dimensional systems of these equations. The\nsolutions give an extremely accurate prediction of the dielectric constant for\na variety of salts and a wide range of concentrations.", "category": "physics_flu-dyn" }, { "text": "Fractional Fourier approximations for potential gravity waves on deep\n water: In the framework of the canonical model of hydrodynamics, where fluid is\nassumed to be ideal and incompressible, waves are potential, two-dimensional,\nand symmetric, the authors have recently reported the existence of a new type\nof gravity waves on deep water besides well studied Stokes waves (Phys. Rev.\nLett., 2002, v. 89, 164502). The distinctive feature of these waves is that\nhorizontal water velocities in the wave crests exceed the speed of the crests\nthemselves. Such waves were found to describe irregular flows with stagnation\npoint inside the flow domain and discontinuous streamlines near the wave\ncrests. Irregular flows produce a simple model for describing the initial stage\nof the formation of spilling breakers when a localized jet is formed at the\ncrest following by generating whitecaps.\n In the present work, a new highly efficient method for computing steady\npotential gravity waves on deep water is proposed to examine the above results\nin more detail. The method is based on the truncated fractional approximations\nfor the velocity potential in terms of the basis functions\n$1/\\bigr(1-\\exp(y_0-y-ix)\\bigl)^n$, $y_0$ being a free parameter. The\nnon-linear transformation of the horizontal scale $x = \\chi - \\gamma \\sin\\chi,\n0<\\gamma<1,$ is additionally applied to concentrate a numerical emphasis on the\ncrest region of a wave for accelerating the convergence of the series.\nFractional approximations were employed for calculating both steep Stokes waves\nand irregular flows. For lesser computational time, the advantage in accuracy\nover ordinary Fourier expansions in terms the basis functions $\\exp\\bigl(n\n(y+ix)\\bigr)$ was found to be from one to ten decimal orders depending on the\nwave steepness and flow parameters.", "category": "physics_flu-dyn" }, { "text": "Saturation of a turbulent mixing layer over a cavity: response to\n harmonic forcing around mean flows: Turbulent mixing layers over cavities can couple with acoustic waves and lead\nto undesired oscillations. To understand the nonlinear aspects of this\nphenomenon, a turbulent mixing layer over a deep cavity at Reynolds number 150\n000 is considered and its response to harmonic forcing is analysed with\nlarge-eddy simulations (LES) and linearised Navier-Stokes equations (LNSE). As\na model of incoming acoustic perturbations, spatially uniform time-harmonic\nforcing is applied at the cavity end, with amplitudes in the wide range\n0.045-8.9% of the bulk velocity. Compressible LES provide reference nonlinear\nresponses of the shear layer, and the associated mean flows. Linear responses\nare calculated with the incompressible LNSE around the LES mean flows; they\npredict well the amplification (both measured with kinetic energy and with a\nproxy for vortex sound production) and capture the nonlinear saturation\nobserved as the forcing amplitude increases and the mixing layer thickens.\nPerhaps surprisingly, LNSE calculations based on a monochromatic (single\nfrequency) assumption yield a good agreement even though higher harmonics and\ntheir nonlinear interaction (Reynolds stresses) are not negligible. However,\nthe leading Reynolds stresses do not force the mixing layer efficiently, as\nshown by a comparison with the optimal volume forcing obtained in a resolvent\nanalysis. Thus, they cannot fully benefit from the potential for amplification\navailable in the flow. Finally, the sensitivity of the optimal harmonic forcing\nat the cavity end is computed with an adjoint method. The sensitivities to mean\nflow modification and to a localised feedback (structural sensitivity) both\nidentify the upstream cavity corner as the region where a small-amplitude\nmodification has the strongest effect. This can guide in a systematic way the\ndesign of strategies for the control of amplification and saturation\nmechanisms.", "category": "physics_flu-dyn" }, { "text": "High-speed turbulent flows towards the exascale: STREAmS-2 porting and\n performance: Exascale High Performance Computing (HPC) represents a tremendous opportunity\nto push the boundaries of Computational Fluid Dynamics (CFD), but despite the\nconsolidated trend towards the use of Graphics Processing Units (GPUs),\nprogrammability is still an issue. STREAmS-2 (Bernardini et al. Comput. Phys.\nCommun. 285 (2023) 108644) is a compressible solver for canonical wall-bounded\nturbulent flows capable of harvesting the potential of NVIDIA GPUs. Here we\nextend the already available CUDA Fortran backend with a novel HIPFort backend\ntargeting AMD GPU architectures. The main implementation strategies are\ndiscussed along with a novel Python tool that can generate the HIPFort and CPU\ncode versions allowing developers to focus their attention only on the CUDA\nFortran backend. Single GPU performance is analysed focusing on NVIDIA A100 and\nAMD MI250x cards which are currently at the core of several HPC clusters. The\ngap between peak GPU performance and STREAmS-2 performance is found to be\ngenerally smaller for NVIDIA cards. Roofline analysis allows tracing this\nbehavior to unexpectedly different computational intensities of the same kernel\nusing the two cards. Parallel performance is measured on the two largest\nEuroHPC pre-exascale systems, LUMI (AMD GPUs) and Leonardo (NVIDIA GPUs).\nStrong scalability reveals more than 80% efficiency up to 16 nodes for Leonardo\nand up to 32 for LUMI. Weak scalability shows an impressive efficiency of over\n95% up to the maximum number of nodes tested (256 for LUMI and 512 for\nLeonardo). This analysis shows that STREAmS-2 is the perfect candidate to fully\nexploit the power of current pre-exascale HPC systems in Europe, allowing users\nto simulate flows with over a trillion mesh points, thus reducing the gap\nbetween the Reynolds numbers achievable in high-fidelity simulations and those\nof real engineering applications.", "category": "physics_flu-dyn" }, { "text": "Relating statistics to dynamics in axisymmetric homogeneous turbulence: The structure and the dynamics of homogeneous turbulence are modified by the\npresence of body forces such that the Coriolis or the buoyancy forces, which\nmay render a wide range of turbulence scales anisotropic. The corresponding\nstatistical characterization of such effects is done in physical space using\nstructure functions, as well as in spectral space with spectra of two-point\ncorrelations, providing two complementary viewpoints. In this framework,\nsecond-order and third-order structure functions are put in parallel with\nspectra of two-point second- and third-order velocity correlation functions,\nusing passage relations. Such relations apply in the isotropic case, or for\nisotropically averaged statistics, which, however, do not reflect the actual\nmore complex structure of anisotropic turbulence submitted to rotation or\nstratification. This complexity is demonstrated in this paper by\norientation-dependent energy and energy transfer spectra produced in both cases\nby means of a two-point statistical model for axisymmetric turbulence. We show\nthat, to date, the anisotropic formalism used in the spectral transfer\nstatistics is especially well-suited to analyze the refined dynamics of\nanisotropic homogeneous turbulence, and that it can help in the analysis of\nisotropically computed third-order structure function statistics often used to\ncharacterize anisotropic contexts.", "category": "physics_flu-dyn" }, { "text": "On high Taylor number Taylor vortices: Axisymmetric steady solutions of Taylor-Couette flow at high Taylor numbers\nare studied numerically and theoretically. As the axial period of the solution\nshortens from about one gap length, the Nusselt number goes through two peaks\nbefore returning to laminar flow. In this process, the asymptotic nature of the\nsolution changes in four stages, as revealed by the asymptotic analysis. When\nthe aspect ratio of the roll cell is about unity, the solution captures\nquantitatively the characteristics of the classical turbulence regime.\nTheoretically, the Nusselt number of the solution is proportional to the\nquarter power of the Taylor number. The maximised Nusselt number obtained by\nshortening the axial period can reach the experimental value around the onset\nof the ultimate turbulence regime, although at higher Taylor numbers the\ntheoretical predictions eventually underestimate the experimental values. An\nimportant consequence of the asymptotic analyses is that the mean angular\nmomentum should become uniform in the core region unless the axial wavelength\nis too short. The theoretical scaling laws deduced for the steady solutions can\nbe carried over to Rayleigh-B\\'enard convection.", "category": "physics_flu-dyn" }, { "text": "Theoretical impulse threshold for particle dislodgement: The problem of determining the threshold of motion of a sediment particle\nresting on the bed of an open channel has historically been dominated by an\napproach based on the time-space-averaged bed shear stress (i.e. Shields\ncriterion). Recently, experimental studies have promoted an alternative\napproach to predict the dislodgement threshold, which is based on the impulse\nof the flow-induced force. Nonetheless, theoretical analyses accompanying these\nstudies result in complex expressions that fail to provide a direct estimate of\nsaid impulse threshold. We employ the work-energy principle to derive a\nprediction of the fundamental impulse threshold that the destabilising\nhydrodynamic force must overcome in order to achieve full particle\ndislodgement. For the bed configuration studied, which is composed of spheres,\nthe proposed expression depends on the mobile particle's size and mass, and\nshows excellent agreement with experimental observations previously published.\nThe derivation presented in this paper may thus represent a robust theoretical\nframework that aids in the re-interpretation of existing data, as well as in\nthe design of future experiments aimed at analysing the importance of\nhydrodynamic impulse as criterion for prediction of particle dislodgement.", "category": "physics_flu-dyn" }, { "text": "Second-order adjoint-based sensitivity for hydrodynamic stability and\n control: Adjoint-based sensitivity analysis is routinely used today to assess\nefficiently the effect of open-loop control on the linear stability properties\nof unstable flows. Sensitivity maps identify regions where small-amplitude\ncontrol is the most effective, i.e. yields the largest first-order (linear)\neigenvalue variation. In this study an adjoint method is proposed for computing\na second-order (quadratic) sensitivity operator, and applied to the flow past a\ncircular cylinder, controlled with a steady body force or a passive device\nmodel. Maps of second-order eigenvalue variations are obtained, without\ncomputing controlled base flows and eigenmodes. For finite control amplitudes,\nthe second-order analysis improves the accuracy of the first-order prediction,\nand informs about its range of validity, and whether it underestimates or\noverestimates the actual eigenvalue variation. Regions are identified where\ncontrol has little or no first-order effect but a second-order effect. In the\ncylinder wake, the effect of a control cylinder tends to be underestimated by\nthe first-order sensitivity, and including second-order effects yields larger\nregions of flow restabilisation. Second-order effects can be decomposed into\ntwo mechanisms: second-order base flow modification, and interaction between\nfirst-order modifications of the base flow and eigenmode. Both contribute\nequally in general in sensitive regions of the cylinder wake. Exploiting the\nsecond-order sensitivity operator, the optimal control maximising the total\nsecond-order stabilisation is computed via a quadratic eigenvalue problem. The\napproach is applicable to other types of control (e.g. wall blowing/suction and\nshape deformation) and other eigenvalue problems (e.g. amplification of\ntime-harmonic perturbations, or resolvent gain, in stable flows).", "category": "physics_flu-dyn" }, { "text": "Lattice Boltzmann simulations of apparent slip in hydrophobic\n microchannels: Various experiments have found a boundary slip in hydrophobic microchannel\nflows, but a consistent understanding of the results is still lacking. While\nMolecular Dynamics (MD) simulations cannot reach the low shear rates and large\nsystem sizes of the experiments, it is often impossible to resolve the needed\ndetails with macroscopic approaches. We model the interaction between\nhydrophobic channel walls and a fluid by means of a multi-phase lattice\nBoltzmann model. Our mesoscopic approach overcomes the limitations of MD\nsimulations and can reach the small flow velocities of known experiments. We\nreproduce results from experiments at small Knudsen numbers and other\nsimulations, namely an increase of slip with increasing liquid-solid\ninteractions, the slip being independent of the flow velocity, and a decreasing\nslip with increasing bulk pressure. Within our model we develop a semi-analytic\napproximation of the dependence of the slip on the pressure.", "category": "physics_flu-dyn" }, { "text": "Effects of Initial Condition Spectral Content on Shock Driven Turbulent\n Mixing: The mixing of materials due to the Richtmyer-Meshkov instability and the\nensuing turbulent behavior is of intense interest in a variety of physical\nsystems including inertial confinement fusion, combustion, and the final stages\nof stellar evolution. Extensive numerical and laboratory studies of\nshock-driven mixing have demonstrated the rich behavior associated with the\nonset of turbulence due to the shocks. Here we report on progress in\nunderstanding shock-driven mixing at interfaces between fluids of differing\ndensities through 3D numerical simulations using the RAGE code in the implicit\nlarge eddy simulation context. We consider a shock tube configuration with a\nband of high density gas (SF$_6$) embedded in low density gas (air). Shocks\nwith a Mach number of 1.26 are passed through SF$_6$ bands, resulting in\ntransition to turbulence driven by the Richtmyer-Meshkov instability. The\nsystem is followed as a rarefaction wave and a reflected secondary shock from\nthe back wall pass through the SF$_6$ band. We apply a variety of initial\nperturbations to the interfaces between the two fluids in which the physical\nstandard deviation, wave number range, and the spectral slope of the\nperturbations are held constant, but the number of modes initially present is\nvaried. By thus decreasing the density of initial spectral modes of the\ninterface, we find that we can achieve as much as 25\\% less total mixing at\nlate times. This has potential direct implications for the treatment of initial\nconditions applied to material interfaces in both 3D and reduced dimensionality\nsimulation models.", "category": "physics_flu-dyn" }, { "text": "Numerical investigation of the formation and stability of homogeneous\n pairs of soft particles in inertial microfluidics: We investigate the formation and stability of a pair of identical soft\ncapsules in channel flow under mild inertia. We employ a combination of the\nlattice Boltzmann, finite element and immersed boundary methods to simulate the\nelastic particles in flow. Validation tests show excellent agreement with\nnumerical results obtained by other research groups. Our results reveal new\ntrajectory types that have not been observed for pairs of rigid particles.\nWhile particle softness increases the likelihood of a stable pair forming, the\npair stability is determined by the lateral position of the particles. A key\nfinding is that stabilisation of the axial distance occurs after lateral\nmigration of the particles. During the later phase of pair formation, particles\nundergo damped oscillations that are independent of initial conditions. These\ndamped oscillations are driven by a strong hydrodynamic coupling of the\nparticle dynamics, particle inertia and viscous dissipation. While the\nfrequency and damping coefficient of the oscillations depend on particle\nsoftness, the pair formation time is largely determined by the initial particle\npositions: the time to form a stable pair grows exponentially with the initial\naxial distance. Our results demonstrate that particle softness has a strong\nimpact on the behaviour of particle pairs. The findings could have significant\nramifications for microfluidic applications where a constant and reliable axial\ndistance between particles is required, such as flow cytometry.", "category": "physics_flu-dyn" }, { "text": "Falling of a viscous jet onto a moving surface: We analyze the stationary flow of a jet of Newtonian fluid that is drawn by\ngravity onto a moving surface. The situation is modeled by a third-order ODE on\na domain of unknown length and with an additional integral condition; by\nsolving part of the equation explicitly we can reformulate the problem as a\nfirst-order ODE, again with an integral constraint. We show that there are two\nflow regimes, and characterize the associated regions in the three-dimensional\nparameter space in terms of an easily calculable quantity. In a qualitative\nsense the results from the model are found to correspond with experimental\nobservations.", "category": "physics_flu-dyn" }, { "text": "Optimizing Flow Control with Deep Reinforcement Learning: Plasma\n Actuator Placement around a Square Cylinder: The present study proposes an active flow control (AFC) approach based on\ndeep reinforcement learning (DRL) to optimize the performance of multiple\nplasma actuators on a square cylinder. The investigation aims to modify the\ncontrol inputs of the plasma actuators to reduce the drag and lift forces\naffecting the cylinder while maintaining a stable flow regime. The environment\nof the proposed model is represented by a two-dimensional direct numerical\nsimulation (DNS) of a flow past a square cylinder. The control strategy is\nbased on the regulation of the supplied alternating current (AC) voltage at\nthree distinct configurations of the plasma actuators. The effectiveness of the\ndesigned strategy is first investigated for Reynolds number, $Re_{D} = 100$,\nand further applied for $Re_{D} = 180$. The applied active flow control\nstrategy is able to reduce the mean drag coefficient by 97\\% at $Re_{D} = 100$\nand by 99\\% at $Re_D=180$. Furthermore, the results from this study show that\nwith the increase in Reynolds number, it becomes more challenging to eliminate\nvortex shedding with plasma actuators located only on the rear surface of the\ncylinder. Nevertheless, the proposed control scheme is able to completely\nsuppress it with an optimized configuration of the plasma actuators.", "category": "physics_flu-dyn" }, { "text": "A unified hyperbolic formulation for viscous fluids and elastoplastic\n solids: We discuss a unified flow theory which in a single system of hyperbolic\npartial differential equations (PDEs) can describe the two main branches of\ncontinuum mechanics, fluid dynamics, and solid dynamics. The fundamental\ndifference from the classical continuum models, such as the Navier-Stokes for\nexample, is that the finite length scale of the continuum particles is not\nignored but kept in the model in order to semi-explicitly describe the essence\nof any flows, that is the process of continuum particles rearrangements. To\nallow the continuum particle rearrangements, we admit the deformability of\nparticle which is described by the distortion field. The ability of media to\nflow is characterized by the strain dissipation time which is a characteristic\ntime necessary for a continuum particle to rearrange with one of its\nneighboring particles. It is shown that the continuum particle length scale is\nintimately connected with the dissipation time. The governing equations are\nrepresented by a system of first order hyperbolic PDEs with source terms\nmodeling the dissipation due to particle rearrangements. Numerical examples\njustifying the reliability of the proposed approach are demonstrated.", "category": "physics_flu-dyn" }, { "text": "Synchronization to big-data: nudging the Navier-Stokes equations for\n data assimilation of turbulent flows: Nudging is an important data assimilation technique where partial field\nmeasurements are used to control the evolution of a dynamical system and/or to\nreconstruct the entire phase-space configuration of the supplied flow. Here, we\napply it to the toughest problem in fluid dynamics: three dimensional\nhomogeneous and isotropic turbulence. By doing numerical experiments we perform\na systematic assessment of how well the technique reconstructs large- and\nsmall-scales features of the flow with respect to the quantity and the\nquality/type of data supplied to it. The types of data used are: (i) field\nvalues on a fixed number of spatial locations (Eulerian nudging), (ii) Fourier\ncoefficients of the fields on a fixed range of wavenumbers (Fourier nudging),\nor (iii) field values along a set of moving probes inside the flow (Lagrangian\nnudging). We present state-of-the-art quantitative measurements of the\nscale-by-scale {\\it transition to synchronization} and a detailed discussion of\nthe probability distribution function of the reconstruction error, by comparing\nthe nudged field and the {\\it truth} point-by-point. Furthermore, we show that\nfor more complex flow configurations, like the case of anisotropic rotating\nturbulence, the presence of cyclonic and anticyclonic structures leads to\nunexpectedly better performances of the algorithm. We discuss potential further\napplications of nudging to a series of applied flow configurations, including\nthe problem of field-reconstruction in thermal Rayleigh-B\\'enard convection and\nin magnetohydrodynamics (MHD), and to the determination of optimal\nparametrisation for small-scale turbulent modeling. Our study fixes the\nstandard requirements for future applications of nudging to complex turbulent\nflows.", "category": "physics_flu-dyn" }, { "text": "Oceanic internal solitary wave interactions via the KP equation in a\n three-layer fluid with shear flow: The various patterns of internal solitary wave interactions are complex\nphenomena in the ocean, susceptible to the influence of shear flow and density\ndistributions. Satellite imagery serves as an effective tool for investigating\nthese interactions, but usually does not provide information on the structure\nof internal waves and their associated dynamics. Considering a three-layer\nconfiguration that approximates ocean stratification, we analytically\ninvestigate two-dimensional internal solitary waves (ISW) in a three-layer\nfluid with shear flow and continuous density distribution using the\n(2+1)-dimensional Kadomtsev-Petviashvili (KP) model. Firstly, the KP equation\nis derived from the basic governing equations which include mass and momentum\nconservations, along with free surface boundary conditions. The coefficients of\nthe KP equation are determined by the vertical distribution of fluid density,\nshear flow, and layer depth. Secondly, it is found that the interactions of ISW\ncan be carefully classified into five types: ordinary interactions including\nO-type, asymmetric interactions including P-type, TP-type and TO-type, and\nMiles resonance. The genuine existence of these interaction types is observed\nfrom satellite images in the Andaman Sea, the Malacca Strait, and the coast of\nWashington state. Finally, the ``bright\" and ``dark\" internal solitary\ninteractions are discovered in the three-layer fluid, which together constitute\nthe fluctuating forms of oceanic ISW. It is revealed that shear flow is the\nprimary factor to determine whether these types of interactions are ``bright\"\nor ``dark\". Besides, a detailed analysis is conducted to show how the ratio of\ndensities influences the properties of these interactions, such as amplitude,\nangle, and wave width.", "category": "physics_flu-dyn" }, { "text": "Drag reduction by polymer additives from turbulent spectra: We extend the analysis of the friction factor for turbulent pipe flow\nreported by G. Gioia, P. Chakraborty and N. Goldenfeld (GCG) (G. Gioia and P.\nChakraborty, Phys. Rev. Lett. 96, 044502 (2006), N. Goldenfeld, Phys. Rev.\nLett. 96, 044503 (2006)) to the case where drag is reduced by polymer\nadditives.", "category": "physics_flu-dyn" }, { "text": "Reinterpreting Shock Wave Structure Predictions using the Navier-Stokes\n Equations: Classical Navier-Stokes equations fail to predict shock wave profiles\naccurately. In this paper, the Navier-Stokes system is fully transformed using\na velocity variable transformation. The transformed equations termed the\nre-casted Navier-Stokes equations display physics not initially included. We\nthen analyse the stationary shock structure problem in a monatomic gas by\nsolving both the classical and the re-casted Navier-Stokes equations\nnumerically using a finite difference global solution (FDGS) scheme. The\nnumerical results are presented for different upstream Mach numbers ranging\nfrom supersonic to hypersonic flows. We found that the re-casted Navier-Stokes\nequations show better agreements with the experimentally measured density and\nreciprocal shock thickness profiles.", "category": "physics_flu-dyn" }, { "text": "Accelerating the Galerkin Reduced-Order Model with the Tensor\n Decomposition for Turbulent Flows: Galerkin-based reduced-order models (G-ROMs) have provided efficient and\naccurate approximations of laminar flows. In order to capture the complex\ndynamics of the turbulent flows, standard G-ROMs require a relatively large\nnumber of reduced basis functions (on the order of hundreds and even\nthousands). Although the resulting G-ROM is still relatively low-dimensional\ncompared to the full-order model (FOM), its computational cost becomes\nprohibitive due to the 3rd-order convection tensor contraction. The tensor\nrequires storage of $N^3$ entries with a corresponding work of $2N^3$\noperations per timestep, which makes such ROMs impossible to use in realistic\napplications, such as control of turbulent flows. In this paper, we focus on\nthe scenario where the G-ROM requires large $N$ values and propose a novel\napproach that utilizes the CANDECOMC/PARAFAC decomposition (CPD), a tensor\ndecomposition technique, to accelerate the G-ROM by approximating the 3rd-order\nconvection tensor by a sum of $R$ rank-1 tensors. In addition, we show that the\ntensor is partially skew-symmetric and derive two conditions for the CP\ndecomposition for preserving the skew-symmetry. Moreover, we investigate the\nG-ROM with the singular value decomposition (SVD). The G-ROM with CP\ndecomposition is investigated in several flow configurations from 2D periodic\nflow to 3D turbulent flows. Our numerical investigation shows CPD-ROM achieves\nat least a factor of 10 speedup. Additionally, the skew-symmetry preserving\nCPD-ROM is more stable and allows the usage of smaller rank $R$. Moreover, from\nthe singular value behavior, the advection tensor formed using the $H^1_0$-POD\nbasis has a low-rank structure, and is preserved even in higher Reynolds\nnumbers. Furthermore, for a given level of accuracy, the CP decomposition is\nmore efficient in size and cost than the SVD.", "category": "physics_flu-dyn" }, { "text": "Self-similar, spatially localized structures in turbulent pipe flow from\n a data-driven wavelet decomposition: This study aims to extract and characterize structures in fully developed\npipe flow at a friction Reynolds number of $\\text{Re}_\\tau = 12\\,400$. To do\nso, we employ data-driven wavelet decomposition (DDWD) [D.~Floryan and\nM.~D.~Graham, PNAS 118, e2021299118 (2021)], a method that combines features of\nproper orthogonal decomposition and wavelet analysis in order to extract\nenergetic and spatially localized structures from data. We apply DDWD to\nstreamwise velocity signals measured separately via a thermal anemometer at 40\nwall-normal positions. The resulting localized velocity structures, which we\ninterpret as being reflective of underlying eddies, are self-similar across\nstreamwise extents of 40 wall units to one pipe radius, and across wall-normal\npositions from $y^+=350$ to $y/R=1$. Notably, the structures are similar in\nshape to Meyer wavelets. Projections of the data onto the DDWD wavelet\nsubspaces are found to be self-similar as well, but in Fourier space; the\nbounds of self-similarity are the same as before, except streamwise\nself-similarity starts at a larger length scale of $450$ wall units. The\nevidence of self-similarity provided in this study lends further support to\nTownsend's attached eddy hypothesis, although we note that the self-similar\nstructures are detected beyond the log layer and extend to large length scales.", "category": "physics_flu-dyn" }, { "text": "Parabolic velocity profile causes drift of inertial prolate spheroids --\n but gravity is stronger: Motion of elongated particles in shear is studied. In applications where\nparticles are much heavier than the carrying fluid, e.g. aerosols, the\ninfluence of particle inertia dominates the particle dynamics. Assuming that\nthe particle only experiences a local linear velocity profile, its rotational\nand translational motion are independent. However, we show that quadratic terms\nof the local velocity profile combined with particle inertia cause a lateral\ndrift of prolate spheroidal particles. We find that this drift is maximal when\nparticle inertial forces are of the same order of magnitude as viscous forces,\nand that both extremely light and extremely heavy particles have negligible\ndrift. In the non-inertial case, the particle rotates according to the local\nlinear velocity profile, with each instantaneous orientation corresponding to a\nvelocity that gives zero force on the particle. This results in a translational\nmotion in the flow direction with periodic velocity fluctuations. With added\nparticle inertia, the particle is slow to react to the surrounding fluid motion\nand the particle will switch between slower and faster rotation compared to the\nzero-force solution. The final motion that gives zero integrated force over a\nrotational period, is a motion with a lateral drift. We show that this drift is\npurely an effect of the non-sphericity of the particle and its translational\ninertia, while rotational inertia is negligible. Finally, although this\ninertial drift will contribute to the lateral motion of heavy elongated\nparticles in channel flow, sedimentation due to gravity will dominate in any\npractical application on earth.", "category": "physics_flu-dyn" }, { "text": "Developed and quasi-developed macro-scale flow in micro- and\n mini-channels with arrays of offset strip fins: We investigate to what degree the steady laminar flow in typical micro- and\nmini-channels with offset strip fin arrays can be described as developed on a\nmacro-scale level, in the presence of channel entrance and side-wall effects.\nHereto, the extent of the developed and quasi-developed flow regions in such\nchannels is determined through large-scale numerical flow simulations. It is\nobserved that the onset point of developed flow increases linearly with the\nReynolds number and channel width, but remains small relative to the total\nchannel length. Further, we find that the local macro-scale pressure gradient\nand closure force for the (double) volume-averaged Navier-Stokes equations are\nadequately modeled by a developed friction factor correlation, as typical\ndiscrepancies are below 15% in both the developed and developing flow region.\nWe show that these findings can be attributed to the eigenvalues and mode\namplitudes which characterize the quasi-developed flow in the entrance region\nof the channel. Finally, we discuss the influence of the channel side walls on\nthe flow periodicity, the mass flow rate, as well as the macro-scale velocity\nprofile, which we capture by a displacement factor and slip length coefficient.\nOur findings are supported by extensive numerical data for fin height-to-length\nratios up to 1, fin pitch-to-length ratios up to 0.5, and channel aspect ratios\nbetween 1/5 and 1/17, covering Reynolds numbers from 28 to 1224.", "category": "physics_flu-dyn" }, { "text": "Physics informed machine learning with Smoothed Particle Hydrodynamics:\n Hierarchy of reduced Lagrangian models of turbulence: Building efficient, accurate and generalizable reduced order models of\ndeveloped turbulence remains a major challenge. This manuscript approaches this\nproblem by developing a hierarchy of parameterized reduced Lagrangian models\nfor turbulent flows, and investigates the effects of enforcing physical\nstructure through Smoothed Particle Hydrodynamics (SPH) versus relying on\nneural networks (NN)s as universal function approximators. Starting from Neural\nNetwork (NN) parameterizations of a Lagrangian acceleration operator, this\nhierarchy of models gradually incorporates a weakly compressible and\nparameterized SPH framework, which enforces physical symmetries, such as\nGalilean, rotational and translational invariances. Within this hierarchy, two\nnew parameterized smoothing kernels are developed in order to increase the\nflexibility of the learn-able SPH simulators. For each model we experiment with\ndifferent loss functions which are minimized using gradient based optimization,\nwhere efficient computations of gradients are obtained by using Automatic\nDifferentiation (AD) and Sensitivity Analysis (SA). Each model within the\nhierarchy is trained on two data sets associated with weekly compressible\nHomogeneous Isotropic Turbulence (HIT): (1) a validation set using weakly\ncompressible SPH; and (2) a high fidelity set from Direct Numerical Simulations\n(DNS). Numerical evidence shows that encoding more SPH structure improves\ngeneralizability to different turbulent Mach numbers and time shifts, and that\nincluding the novel parameterized smoothing kernels improves the accuracy of\nSPH at the resolved scales.", "category": "physics_flu-dyn" }, { "text": "High- and low-entropy layers in solids behind shock and ramp compression\n waves: Non-uniform temperature fields are analyzed, which arise in the problems of\nformation of the steady shock wave at impact and ramp loading of metals, exit\nof the steady shock wave to the free surface, and the shock wave passing\nthrough the interface between two different materials. Theoretical analysis and\ncomputations show that high-entropy (with the temperature increase) and\nlow-entropy (with the temperature decrease) layers arise near the interfaces in\nthe above problems of shock and ramp loading. The impact produces the\nhigh-entropy layer; while the ramp loading can result in the both high- and\nlow-entropy layers. At the shock wave passing through the interface, the\nhigh-entropy layer is formed in the lower-impedance material and the\nlow-entropy -- in the higher-impedance one. The formation of high-entropy layer\nat impact is supported by molecular-dynamics simulations in addition to\ncontinuum modeling. The high- and low-entropy layers should be taken into\naccount in simulations of shock-wave processes in thin targets or in other\ncases where surface effects are important.", "category": "physics_flu-dyn" }, { "text": "Efficient swimming of an assembly of rigid spheres at low Reynolds\n number: The swimming of an assembly of rigid spheres immersed in a viscous fluid of\ninfinite extent is studied in low Reynolds number hydrodynamics. The\ninstantaneous swimming velocity and rate of dissipation are expressed in terms\nof the time-dependent displacements of sphere centers about their collective\nmotion. For small amplitude swimming with periodically oscillating\ndisplacements, optimization of the mean swimming speed at given mean power\nleads to an eigenvalue problem involving a velocity matrix and a power matrix.\nThe corresponding optimal stroke permits generalization to large amplitude\nmotion in a model of spheres with harmonic interactions and corresponding\nactuating forces. The method allows straightforward calculation of the swimming\nperformance of structures modeled as assemblies of interacting rigid spheres. A\nmodel of three collinear spheres with motion along the common axis is studied\nas an example.", "category": "physics_flu-dyn" }, { "text": "Viscous lubrication force between spherical bubbles with time-dependent\n radii: Motivated by the dynamics of microbubbles in dissolved gas flotation\nprocesses, we consider theoretically the approach between two shear-free\nspherical bubbles with time-dependent radii. We make use of the lubrication\nassumption to obtain the thin film flow between the bubbles. Our analysis\nunderscores that for the shear-free condition and spherical shape assumption to\nhold, both the viscosity ratio and the capillary number must be significantly\nsmaller than the thickness of the film. We demonstrate that the lubrication\nforce exhibits weak singular behavior, scaling logarithmically with the ratio\nof bubble radius to film thickness. To assess the accuracy of our findings, we\ncompare the obtained solution to results from Stokes flow theory. The\ncomparison demonstrates that our current results are reliable, provided that we\ncombine the lubrication forces with subdominant corrections, which require\nproper matching and computation to the solution far from the film. In practice,\nwe compute these subdominant corrections in the case of two equal bubbles or a\nbubble close to a plane-free surface either by a curve fit of numerical results\nfrom bi-spherical coordinate solutions or by using results from the literature.\nWe illustrate the relevance of the solution to determine the drainage time of a\nsmall bubble rising to a free surface and the drainage rate of expanding\nbubbles under force-free conditions. Finally, in the discussion, we relax the\nassumption of negligible shear and show that even a small but non-negligible\nshear induced by fluid motion within the bubble introduces a singular term in\nthe lubrication force.", "category": "physics_flu-dyn" }, { "text": "What connects ignition and deflagration? -- On explosive transition of\n deflagration: The general relation between ignition and deflagration analyses was studied\ntheoretically and computationally. Simple analysis showed that the temporal\nevolutions of normalized fuel mass fraction and the normalized temperature in a\n0D homogeneous ignition are equivalent to the spatial evolutions of those in a\n1D laminar premixed flame at Lewis number of unity if a spatio-temporal\ntransformation of the flame is applied. Furthermore, computations showed that\nthe degree of reduction in normalized fuel mass fraction in the preheat zone\nwas found to be in the following order: when the Lewis number is less than\nunity, when the Lewis number is unity (= ignition), and when the Lewis number\nis greater than unity. This suggested that ignition in the preheat zone near\nthe explosive transition can only occur for 1D laminar premixed flames with\nLewis number greater than unity. In summary, the characteristic times of\nignition and flame propagation can be of the same order of magnitude in\npremixed gases compressed by piston in engines or shock waves, and thus the\nLewis number determines the existence of premixed flame structure under such\nconditions.", "category": "physics_flu-dyn" }, { "text": "On the topology of the atmosphere advected by a periodic array of\n axisymmetric thin-cored vortex rings: The fluid motion produced by a periodic array of identical, axisymmetric,\nthin-cored vortex rings is investigated. It is well known that such an array\nmoves uniformly without change of shape or form in the direction of the central\naxis of symmetry, and is therefore an equilibrium solution of Euler's\nequations. In a frame of reference moving with the system of vortex rings, the\nmotion of passive fluid particles is investigated as a function of the two\nnon-dimensional parameters that define this system: $\\varepsilon = a/R$, the\nratio of minor radius to major radius of the torus-shaped vortex rings, and\n$\\lambda = L/R$, the separation of the vortex rings normalized by their radii.\nTwo bifurcations in the streamline topology are found that depend significantly\non $\\varepsilon$ and $\\lambda$; these bifurcations delineate three distinct\nshapes of the 'atmosphere' of fluid particles that move together with the\nvortex ring for all time. Analogous to the case of an isolated vortex ring, the\natmospheres can be 'thin-bodied' or 'thick-bodied'. Additionally, we find the\noccurrence of a 'connected' system, in which the atmospheres of neighboring\nrings touch at an invariant ring of fluid particles that is stationary in a\nframe of reference moving with the rings.", "category": "physics_flu-dyn" }, { "text": "Lubricated wrinkles: imposed constraints affect the dynamics of wrinkle\n coarsening: We study the dynamic coarsening of wrinkles in an elastic sheet that is\ncompressed while lying on a thin layer of viscous liquid. When the ends of the\nsheet are instantaneously brought together by a small distance, viscous\nresistance initially prevents the sheet from adopting a globally buckled shape.\nInstead, the sheet accommodates the compression by wrinkling. Previous scaling\narguments suggested that a balance between the sheet's bending stiffness and\nviscous effects lead to a wrinkle wavelength $\\lambda$ that increases with time\n$t$ according to $\\lambda\\propto t^{1/6}$. We show that taking proper account\nof the compression constraint leads to a logarithmic correction of this result,\n$\\lambda\\propto (t/\\log t)^{1/6}$. This correction is significant over\nexperimentally observable time spans, and leads us to reassess previously\npublished experimental data.", "category": "physics_flu-dyn" }, { "text": "Breaking anchored droplets in a microfluidic Hele-Shaw cell: We study microfluidic self digitization in Hele-Shaw cells using pancake\ndroplets anchored to surface tension traps. We show that above a critical flow\nrate, large anchored droplets break up to form two daughter droplets, one of\nwhich remains in the anchor. Below the critical flow velocity for breakup the\nshape of the anchored drop is given by an elastica equation that depends on the\ncapillary number of the outer fluid. As the velocity crosses the critical\nvalue, the equation stops admitting a solution that satisfies the boundary\nconditions; the drop breaks up in spite of the neck still having finite width.\nA similar breaking event also takes place between the holes of an array of\nanchors, which we use to produce a 2D array of stationary drops in situ.", "category": "physics_flu-dyn" }, { "text": "A method for simulating interfacial mass transfer on arbitrary meshes: This paper presents a method for modelling interfacial mass transfer in\nInterface Capturing simulations of two-phase flow with phase change. The model\nenables mechanistic prediction of the local rate of phase change at the\nvapour-liquid interface on arbitrary computational meshes and is applicable to\nrealistic cases involving two-phase mixtures with large density ratios. The\nsimulation methodology is based on the Volume Of Fluid (VOF) representation of\nthe flow, whereby an interfacial region in which mass transfer occurs is\nimplicitly identified by a phase indicator, in this case the volume fraction of\nliquid, which varies from the value pertaining to the \"bulk\" liquid to the\nvalue of the bulk vapour. The novel methodology proposed here has been\nimplemented using the Finite Volume framework and solution methods typical of\n\"industrial\" CFD practice. The proposed methodology for capturing mass transfer\nis applicable to arbitrary meshes without the need to introduce elaborate but\nartificial smearing of the mass transfer term as is often done in other\ntechniques. The method has been validated via comparison with analytical\nsolutions for planar interface evaporation and bubble growth test cases, and\nagainst experimental observations of steam bubble growth.", "category": "physics_flu-dyn" }, { "text": "WaterLily.jl: A differentiable fluid simulator in Julia with fast\n heterogeneous execution: Integrating computational fluid dynamics (CFD) software into optimization and\nmachine-learning frameworks is hampered by the rigidity of classic\ncomputational languages and the slow performance of more flexible high-level\nlanguages. WaterLily.jl is an open-source incompressible viscous flow solver\nwritten in the Julia language. The small code base is multi-dimensional,\nmulti-platform and backend-agnostic (serial CPU, multi-threaded, & GPU\nexecution). The simulator is differentiable and uses automatic-differentiation\ninternally to immerse solid geometries and optimize the pressure solver. The\ncomputational time per time step scales linearly with the number of degrees of\nfreedom on CPUs, and we see up to a 182x speed-up using CUDA kernels. This\nleads to comparable performance with Fortran solvers on many research-scale\nproblems opening up exciting possible future applications on the cutting edge\nof machine-learning research.", "category": "physics_flu-dyn" }, { "text": "Thermodynamically Consistent, Frame Indifferent Diffuse Interface Models\n for Incompressible Two-Phase Flows with Different Densities: A new diffuse interface model for a two-phase flow of two incompressible\nfluids with different densities is introduced using methods from rational\ncontinuum mechanics. The model fulfills local and global dissipation\ninequalities and is frame indifferent. Moreover, it is generalized to\nsituations with a soluble species. Using the method of matched asymptotic\nexpansions we derive various sharp interface models in the limit when the\ninterfacial thickness tends to zero. Depending on the scaling of the mobility\nin the diffusion equation we either derive classical sharp interface models or\nmodels where bulk or surface diffusion is possible in the limit. In the latter\ncase a new term resulting from surface diffusion appears in the momentum\nbalance at the interface. Finally, we show that all sharp interface models\nfulfill natural energy inequalities.", "category": "physics_flu-dyn" }, { "text": "A numerical model to predict unsteady cavitating flow behaviour in\n inducer blade cascades: The cavitation behaviour of a four-blade rocket engine turbopump inducer is\nsimulated. A 2D numerical model of unsteady cavitation was applied to a blade\ncascade drawn fromthe inducer geometry. The physical model is based on a\nhomogeneous approach of cavitation, coupled with a barotropic state law for the\nliquid/vapour mixture. The numericalresolution uses a pressure-correction\nmethod derived from the SIMPLE algorithm and a finite volume discretization.\nUnsteadybehaviour of sheet cavities attached to the blade suction side depends\non the flow rate and cavitation number. Two differentunstable configurations of\nrotating cavitation, respectively sub-synchronous and super-synchronous, are\nidentified. The mechanisms that are responsible for these unstable behaviours\nare discussed, and the stress fluctuations induced on the blade by the rotating\ncavitation are estimated.", "category": "physics_flu-dyn" }, { "text": "Backpropagation and gradient descent for an optimized dynamic mode\n decomposition: We present a robust and flexible optimization approach for dynamic mode\ndecomposition analysis of data with complex dynamics and low signal-to-noise\nratios. The approach borrows techniques and insights from the field of deep\nlearning. Specifically, we employ automatic differentiation and stochastic\ngradient descent to compute eigenvalues, modes, and mode amplitudes\nsimultaneously. The method allows embedding regularization or physical\nconstraints into the operator definition. The optimization approach is applied\nto three examples of increasing complexity, the most challenging of which is an\nexperimental dataset of transonic shock buffet on a swept at realistic flight\nconditions.", "category": "physics_flu-dyn" }, { "text": "Field theoretical formulation of the asymptotic relaxation states of\n two-dimensional ideal fluids: The ideal incompressible fluid in two dimensions (Euler fluid) evolves at\nrelaxation from turbulent states to highly coherent states of flow. For the\ncase of double spatial periodicity and zero total vorticity it is known that\nthe streamfunction verifies the \\emph{sinh}-Poisson equation. These exceptional\nstates can only be identified in a description based on the extremum of an\naction functional. Starting from the discrete model of interacting point-like\nvortices it was possible to write a Lagrangian in terms of a matter function\nand a gauge potential. They provide a dual representation of the same physical\nobject, the vorticity. This classical field theory identifies the stationary,\ncoherent, states of the $2D$ Euler fluid as derived from the self-duality. We\nfirst provide a more detailed analysis of this model, including a comparison\nwith the approach based on the statistical physics of point-like vortices. The\nsecond main objective is the study of the dynamics in close proximity of the\nstationary self-dual state, \\emph{i.e.} before the system has reached the\nabsolute extremum of the action functional. Finally, limitations and possible\nextensions of this field theoretical model for the $2D$ fluids model are\ndiscussed and some possible applications are mentioned.", "category": "physics_flu-dyn" }, { "text": "Transitions in turbulent rotating Rayleigh-B\u00e9nard convection: Numerical simulations of rotating Rayleigh-B\\'enard convection are presented\nfor both no slip and free slip boundaries. The goal is to find a criterion\ndistinguishing convective flows dominated by the Coriolis force from those\nnearly unaffected by rotation. If one uses heat transport as an indicator of\nwhich regime the flow is in, one finds that the transition between the flow\nregimes always occurs at the same value of a certain combination of Reynolds,\nPrandtl and Ekman numbers for both boundary conditions. If on the other hand\none uses the helicity of the velocity field to identify flows nearly\nindependent of rotation, one finds the transition at a different location in\nparameter space.", "category": "physics_flu-dyn" }, { "text": "Critical Prandtl number for Heat Transfer Enhancement in Rotating\n Convection: Rotation, which stabilizes flow, can enhance the heat transfer in\nRayleigh-B\\'enard convection (RBC) through Ekman pumping. In this Letter, we\npresent the results of our direct numerical simulations of rotating RBC,\nproviding a comprehensive analysis of this heat transfer enhancement relative\nto non-rotating RBC in the parameter space of Rayleigh number ($Ra$), Prandtl\nnumber ($Pr$), and Taylor number ($Ta$). We show that for a given $Ra$, there\nexists a critical Prandtl number ($Pr_{cr}$) below which no significant heat\ntransfer enhancement occurs at any rotation rate, and an optimal Prandtl number\n($Pr_{opt}$) at which maximum heat transfer enhancement occurs at an optimal\nrotation rate ($Ta_{opt}$). Notably, $Pr_{cr}$, $Pr_{opt}$, $Ta_{opt}$, and the\nmaximum heat transfer enhancement all increase with increasing $Ra$. We also\ndemonstrate a significant heat transfer enhancement up to $Ra=2\\times 10^{10}$\nand predict that the enhancement would become even more pronounced at higher\n$Ra$, provided $Pr$ is also increased commensurately.", "category": "physics_flu-dyn" }, { "text": "Drag reduction in boiling Taylor-Couette turbulence: We create a highly controlled lab environment-accessible to both global and\nlocal monitoring-to analyse turbulent boiling flows and in particular their\nshear stress in a statistically stationary state. Namely, by precisely\nmonitoring the drag of strongly turbulent Taylor-Couette flow (the flow in\nbetween two co-axially rotating cylinders, Reynolds number $\\textrm{Re}\\approx\n10^6$) during its transition from non-boiling to boiling, we show that the\nintuitive expectation, namely that a few volume percent of vapor bubbles would\ncorrespondingly change the global drag by a few percent, is wrong. Rather, we\nfind that for these conditions a dramatic global drag reduction of up to 45%\noccurs. We connect this global result to our local observations, showing that\nfor major drag reduction the vapor bubble deformability is crucial,\ncorresponding to Weber numbers larger than one. We compare our findings with\nthose for turbulent flows with gas bubbles, which obey very different physics\nthan vapor bubbles. Nonetheless, we find remarkable similarities and explain\nthese.", "category": "physics_flu-dyn" }, { "text": "Techniques for Generating Centimetric Drops in Microgravity and\n Application to Cavitation Studies: This paper describes the techniques and physical parameters used to produce\nstable centimetric water drops in microgravity, and to study single cavitation\nbubbles inside such drops (Parabolic Flight Campaigns, European Space Agency\nESA). While the main scientific results have been presented in a previous\npaper, we shall herein provide the necessary technical background, with\npotential applications to other experiments. First, we present an original\nmethod to produce and capture large stable drops in microgravity. This\ntechnique succeeded in generating quasi-spherical water drops with volumes up\nto 8 ml, despite the residual g-jitter. We find that the equilibrium of the\ndrops is essentially dictated by the ratio between the drop volume and the\ncontact surface used to capture the drop, and formulate a simple stability\ncriterion. In a second part, we present a setup for creating and studying\nsingle cavitation bubbles inside those drops. In addition, we analyze the\ninfluence of the bubble size and position on the drop behaviour after collapse,\ni.e. jets and surface perturbations.", "category": "physics_flu-dyn" }, { "text": "Lagrangian Manifestation of Anomalies in Active Turbulence: We show that Lagrangian measurements in active turbulence bear imprints of\nturbulent and anomalous streaky hydrodynamics leading to a self-selection of\npersistent trajectories - Levy walks - over diffusive ones. This emergent\ndynamical heterogeneity results in a super-diffusive first passage distribution\nwhich could lead to biologically advantageous motility. We then go beyond\nsingle-particle statistics to show that for the pair-dispersion problem as\nwell, active flows are at odds with inertial turbulence. Our study, we believe,\nwill readily inform experiments in establishing the extent of universality of\nanomalous behaviour across a variety of active flows.", "category": "physics_flu-dyn" }, { "text": "Surface wave dynamics in orbital shaken cylindrical containers: Be it to aerate a glass of wine before tasting, to accelerate a chemical\nreaction or to cultivate cells in suspension, the \"swirling\" (or orbital\nshaking) of a container ensures good mixing and gas exchange in an efficient\nand simple way. Despite being used in a large range of applications this\nintuitive motion is far from being understood and presents a richness of\npatterns and behaviors which has not yet been reported. The present research\ncharts the evolution of the waves with the operating parameters identifying a\nlarge variety of patterns, ranging from single and multiple crested waves to\nbreaking waves. Free surface and velocity fields measurements are compared to a\npotential sloshing model, highlighting the existence of various flow regimes.\nOur research assesses the importance of the modal response of the shaken\nliquids, laying the foundations for a rigorous mixing optimization of the\norbital agitation in its applications. Copyright (2014) American Institute of\nPhysics. This article may be downloaded for personal use only. Any other use\nrequires prior permission of the author and the American Institute of Physics.\nThe following article appeared in Physics of Fluids 26, 052104 (2014) and may\nbe found at http://dx.doi.org/10.1063/1.4874612", "category": "physics_flu-dyn" }, { "text": "Note on Onsager's conjecture: Onsager conjectured that solutions of the incompressible Euler equations\npossessing a certain degree of roughness do not conserve the kinetic energy.\nSince, within the physical frame of Onsager's conjecture, the kinetic energy is\nthe only occurring energy, and thus identical with the total energy, the\nimplication would be that the conservation of energy is not absolute, but\nsubject to the properties of mathematical solutions. Further, Onsager\nintroduced the concept of anomalous dissipation of kinetic energy without\nviscosity. Both these aspects are critically discussed and their shortcomings\nunveiled.", "category": "physics_flu-dyn" }, { "text": "Interpolatory-based data-driven pulsed fluidic actuator control design\n and experimental validation: Pulsed fluidic actuators play a central role in the fluid flow experimental\ncontrol strategy to achieve better performances of aeronautic devices. In this\npaper, we demonstrate, through an experimental test bench, how the\ninterpolatory-based Loewner Data-Driven Control (L-DDC) framework is an\nappropriate tool for accurately controlling the outflow velocity of this family\nof actuators. L-DDC combines the concept of ideal controller with the Loewner\nframework in a single data-driven rationale, appropriate to experimental users.\nThe contributions of the paper are, first, to emphasise the simplicity and\nversatility of such a data-driven rationale in a constrained experimental\nsetup, and second, to solve some practical fluid engineers concerns by\ndetailing the complete workflow and key ingredients for successfully\nimplementing a pulsed fluidic actuator controller from the data acquisition to\nthe control implementation and validation stages.", "category": "physics_flu-dyn" }, { "text": "Linear and nonlinear optimal growth mechanisms for generating turbulent\n bands: Linear and nonlinear energy optimizations in a tilted domain are used to\nunveil the main mechanisms allowing the creation of a turbulent band in a\nchannel flow. Linear optimization predicts an optimal growth for streamwse and\nspanwise wavenumbers $k_x = 1.2$, $k_z = - 1.75$, corresponding to the peak\nvalues of the premultiplied energy spectra of direct numerical simulations. At\ntarget time, the linear optimal perturbation is composed by oblique streaks,\nwhich, for a sufficiently large initial energy, induce turbulence in the whole\ndomain, due to the lack of spatial localization. When localization is achieved\nby adding nonlinear effects to the optimization, or by artificially confining\nthe linear optimal to a localized region in the spanwise direction, a\nlarge-scale flow is created, which leads to the generation of a localised\nturbulent band. These results suggest that inducing transition towards\nturbulent bands in a tilted domain, two main elements are needed: a linear\nenergy growth mechanism such as the lift-up for generating large-amplitude flow\nstructures which produce inflection points; large-scale vortices ensuring\nspatial localisation. Remarkably, these two elements alone are able to generate\nan isolated turbulent band also in a large, non-tilted domain.", "category": "physics_flu-dyn" }, { "text": "Structure of A Heterogeneous Two-Phase Rotating Detonation Wave with\n Ethanol-Hydrogen-Air Mixture: In this short Letter, the structure of a rotating detonation wave (RDW)\nfueled by biofuel is revealed and expounded. The simulation is carried out\nunder an Eulerian-Lagrangian framework in which the main characteristics of the\ntwo-phase RDW are analyzed in detail. Results suggest a self-sustained rotating\ndetonation fueled by liquid ethanol and air can be achieved with hydrogen\naddition for combustion enhancement, and a laminated dual-front structure of\nthe RDW due to the effect of droplet evaporation is captured and clarified.", "category": "physics_flu-dyn" }, { "text": "How the growth of ice depends on the fluid dynamics underneath: Convective flows coupled with solidification or melting in water bodies play\na major role in shaping geophysical landscapes. Particularly in relation to the\nglobal climate warming scenario, it is essential to be able to accurately\nquantify how water-body environments dynamically interplay with ice formation\nor melting process. Previous studies have revealed the complex nature of the\nicing process, but have often ignored one of the most remarkable particularity\nof water, its density anomaly, and the induced stratification layers\ninteracting and coupling in a complex way in presence of turbulence and phase\nchange. By combining experiments, numerical simulations, and theoretical\nmodeling, we investigate solidification of freshwater, properly considering\nphase transition, water density anomaly, and real physical properties of ice\nand water phases, which we show to be essential for correctly predicting the\ndifferent qualitative and quantitative behaviors. We identify, with increasing\nthermal driving, four distinct flow-dynamics regimes, where different levels of\ncoupling among ice front, stably and unstably stratified water layers occur.\nDespite the complex interaction between the ice front and fluid motions,\nremarkably, the average ice thickness and growth rate can be well captured with\nthe theoretical model. It is revealed that the thermal driving has major\neffects on the temporal evolution of the global icing process, which can vary\nfrom a few days to a few hours in the current parameter regime. Our model can\nbe applied to general situations where the icing dynamics occurs under\ndifferent thermal and geometrical conditions (e.g. cooling conditions or water\nlayer depth).", "category": "physics_flu-dyn" }, { "text": "Turbulence appearance and non-appearance in thin fluid layers: Flows in fluid layers are ubiquitous in industry, geophysics and\nastrophysics. Large-scale flows in thin layers can be considered\ntwo-dimensional (2d) with bottom friction added. Here we find that the\nproperties of such flows depend dramatically on the way they are driven. We\nargue that wall-driven (Couette) flow cannot sustain turbulence at however\nsmall viscosity and friction. Direct numerical simulations (DNS) up to the\nReynolds number $Re=10^6$ confirm that all perturbations die in a plane Couette\nflow. On the contrary, for sufficiently small viscosity and friction, we show\nthat finite perturbations destroy the pressure-driven laminar (Poiseuille)\nflow. What appears instead is a traveling wave in the form of a jet slithering\nbetween wall vortices. For $10^4 1$ and $\\phi_{xy} < 1$ were\ndesigned and fabricated for experiments in a benchtop water channel experiment.\nParticle Image Velocimetry (PIV) measurements were used to compute mean\nturbulence statistics and to educe coherent structure via snapshot Proper\nOrthogonal Decomposition (POD). Friction velocity estimates based on the\nReynolds shear stress profiles do not show evidence of discernible friction\nreduction (or increase) over the streamwise-preferential substrate with\n$\\phi_{xy}>1$ relative to a smooth wall flow at identical bulk Reynolds number.\nA significant increase in friction is observed over the substrate with\n$\\phi_{xy} < 1$. This increase in friction is linked to the emergence of\nspanwise rollers resembling Kelvin-Helmholtz vortices. Coherent structures\nextracted via POD analysis show qualitative agreement with model predictions.", "category": "physics_flu-dyn" }, { "text": "Profile of a Two-Dimensional Vortex Condensate Beyond the Universal\n Limit: It is well known that an inverse turbulent cascade in a finite ($2 \\pi \\times\n2 \\pi$) two-dimensional periodic domain leads to the emergence of a\nsystem-sized coherent vortex dipole. We report a numerical hyperviscous study\nof the spatial vorticity profile inside one of the vortices. The exciting force\nwas shortly correlated in time, random in space, and had a correlation length\n$l_f = 2\\pi/k_f$ with $k_f$ ranging from $100$ to $12.5$. Previously, it was\nfound that in the asymptotic limit of small-scale forcing, the vorticity\nexhibits the power-law behavior $\\Omega(r) = (3 \\epsilon/\\alpha)^{1/2} r^{-1}$,\nwhere $r$ is the distance to the vortex center, $\\alpha$ is the bottom friction\ncoefficient, and $\\epsilon$ is the inverse energy flux. Now we show that for a\nspatially homogeneous forcing with finite $k_f$ the vorticity profile becomes\nsteeper, with the difference increasing with the pumping scale but decreasing\nwith the Reynolds number at the forcing scale. Qualitatively, this behaviour is\nrelated to a decrease in the effective pumping of the coherent vortex with\ndistance from its center. To support this statement, we perform an additional\nsimulation with spatially localized forcing, in which the effective pumping of\nthe coherent vortex, on the contrary, increases with $r$ and show for the first\ntime that in this case the vorticity profile can be flatter than the asymptotic\nlimit.", "category": "physics_flu-dyn" }, { "text": "Sub-Kolmogorov-Scale Fluctuations in Fluid Turbulence: We relate the intermittent fluctuations of velocity gradients in turbulence\nto a whole range of local dissipation scales generalizing the picture of a\nsingle mean dissipation length. The statistical distribution of these local\ndissipation scales as a function of Reynolds number is determined in numerical\nsimulations of forced homogeneous isotropic turbulence with a spectral\nresolution never applied before which exceeds the standard one by at least a\nfactor of eight. The core of the scale distribution agrees well with a\ntheoretical prediction. Increasing Reynolds number causes the generation of\never finer local dissipation scales. This is in line with a less steep decay of\nthe large-wavenumber energy spectra in the dissipation range. The energy\nspectrum for the highest accessible Taylor microscale Reynolds number\nR_lambda=107 does not show a bottleneck.", "category": "physics_flu-dyn" }, { "text": "Tuning turbine rotor design for very large wind farms: A new theoretical method is presented for future multi-scale aerodynamic\noptimisation of very large wind farms. The new method combines a recent\ntwo-scale coupled momentum analysis of ideal wind turbine arrays with the\nclassical blade-element-momentum (BEM) theory for turbine rotor design, making\nit possible to explore some potentially important relationships between the\ndesign of rotors and their performance in a very large wind farm. The details\nof the original two-scale momentum model are described first, followed by the\nnew coupling procedure with the classical BEM theory and some example\nsolutions. The example solutions, obtained using a simplified but still\nrealistic NREL S809 airfoil performance curve, illustrate how the\naerodynamically optimal rotor design may change depending on the farm density.\nIt is also shown that the peak power of the rotors designed optimally for a\ngiven farm (i.e. 'tuned' rotors) could be noticeably higher than that of the\nrotors designed for a different farm (i.e. 'untuned' rotors) even if the blade\npitch angle is allowed to be adjusted optimally during the operation. The\nresults presented are for ideal very large wind farms and a possible future\nextension of the present work for real large wind farms is also discussed\nbriefly.", "category": "physics_flu-dyn" }, { "text": "Experimental observations on interaction between a root and droplets in\n relation to aeroponic agriculture: Aeroponics or Soil-less agriculture is a relatively new and recent type of\npractice, where plants are grown without soil while nutrient-rich water is\nprovided via an atomized spray system to the suspended roots. Spray nozzles are\neasy-to-use in supplying water (and fertilizers) to (mainly) the roots and root\nhairs of the desired crop (or plant) for production. We characterize a spray\nnozzle delivering water vertically above against the gravity by measuring,\nexperimentally, its (a) spray drift, (b) spray height, (c) maximum spray angle,\n(d) spray width, and (e) droplets sizes. Experiments were carried out at\ndifferent inlet pressures and a majority of the above mentioned parameters were\nobtained by processing the images captured using optical (or high speed)\ncamera, sometimes along a plane lighted by a high-power laser source. We also\nstudied the spray (or jet) behaviour at different vertical heights and\ndifferent horizontal planes using a unique polythene sponge method. We studied\nthe mass flow rate, the absorption rate, and droplet size dynamics (as a\nfunction of time and pressure) using this method. The water drop/droplet\ninteraction was also studied in the case of simpler porous and impervious\nsurfaces as well. We believe that this study can be extrapolated to other\nnozzles (especially sprays) to obtain similar characteristic parameters. This\nstudy, hence, is critical in selecting the desired spray system for a given\ncanopy and is also expected to be of some use in controlled agricultural\npractices such as in greenhouses and apartment rooms.", "category": "physics_flu-dyn" }, { "text": "A Kinematic Evolution Equation for the Dynamic Contact Angle and some\n Consequences: We investigate the moving contact line problem for two-phase incompressible\nflows with a kinematic approach. The key idea is to derive an evolution\nequation for the contact angle in terms of the transporting velocity field. It\nturns out that the resulting equation has a simple structure and expresses the\ntime derivative of the contact angle in terms of the velocity gradient at the\nsolid wall. Together with the additionally imposed boundary conditions for the\nvelocity, it yields a more specific form of the contact angle evolution. Thus,\nthe kinematic evolution equation is a tool to analyze the evolution of the\ncontact angle. Since the transporting velocity field is required only on the\nmoving interface, the kinematic evolution equation also applies when the\ninterface moves with its own velocity independent of the fluid velocity. We\napply the developed tool to a class of moving contact line models which employ\nthe Navier slip boundary condition. We derive an explicit form of the contact\nangle evolution for sufficiently regular solutions, showing that such solutions\nare unphysical. Within the simplest model, this rigorously shows that the\ncontact angle can only relax to equilibrium if some kind of singularity is\npresent at the contact line. Moreover, we analyze more general models including\nsurface tension gradients at the contact line, slip at the fluid-fluid\ninterface and mass transfer across the fluid-fluid interface.", "category": "physics_flu-dyn" }, { "text": "Dynamic triad interactions and evolving turbulence spectra: We investigate the effect of a four-dimensional Fourier transform on the\nformulation of Navier-Stokes equation in Fourier space and the way the energy\nis transferred between Fourier components (modal interactions, commonly\nreferred to as triad interactions in the classical 3-dimensional analysis). We\nspecifically consider the effect of a spatially and temporally finite,\ndigitally sampled velocity record on the modal interactions and find that\nFourier components may interact within a broadened frequency window as compared\nto the usual integrals over infinite ranges. We also see how these finite\nvelocity records have a significant effect on the efficiency of the different\nmodal interactions and thereby on the shape and development of velocity power\nspectra. The observation that mismatches in the wavevector triadic interactions\nmay be compensated by a corresponding mismatch in the frequencies supports the\nempirically deduced delayed interactions reported in [Josserand \\textit{et\nal.}, \\textit{J. Stat. Phys.} (2017)]. Collectively, these results explain the\noccurrence and time development of the so-called Richardson cascade and also\nwhy deviations from the classical Richardson cascade may occur. Finally, we\nquote results from companion papers that deal with measurements and computer\nsimulations of the time development of velocity power spectra in a turbulent\njet flow into which a single Fourier mode (narrow band oscillation) is\ninjected.", "category": "physics_flu-dyn" }, { "text": "Chaos, randomization, and turbulence in particle-laden flows: The randomization effect of the two-way (particle-flow) interaction has been\nstudied and quantified using the notion of distributed chaos and the results of\nnumerical simulations and laboratory measurements. It is shown, in particular,\nthat an increase of such parameters as the particle volume fraction, particle\nmass loading, and Stokes number results generally in stronger randomization of\nthe particle-laden flows. An important role of spontaneous breaking of the\nlocal reflectional symmetry in the randomization of the particle-laden flows\nhas been also analyzed using relevant dynamical invariants.", "category": "physics_flu-dyn" }, { "text": "Near-wall depletion and layering affect contact line friction of\n multicomponent liquids: The main causes of energy dissipation in micro- and nano-scale wetting are\nviscosity and liquid-solid friction localized in the three-phase contact line\nregion. Theoretical models predict the contactline friction coefficient to\ncorrelate with the shear viscosity of the wetting fluid. Experiments conducted\nto investigate such correlation have not singled out a unique scaling law\nbetween the two coefficients. We perform Molecular Dynamics simulations of\nliquid water-glycerol droplets wetting silica-like surfaces, aimed to demystify\nthe effect of viscosity on contact line friction. The viscosity of the fluid is\ntuned by changing the relative mass fraction of glycerol in the mixture and it\nis estimated both via equilibrium and non-equilibrium Molecular Dynamics\nsimulations. Contact line friction is measured directly by inspecting the\nvelocity of the moving contact line and the microscopic contact angle. It is\nfound that the scaling between contact line friction and viscosity is\nsub-linear, contrary to the prediction of Molecular Kinetic Theory. The\ndisagreement is explained by accounting for the depletion of glycerol in the\nnear-wall region. A correction is proposed, based on multicomponent Molecular\nKinetic Theory and the definition of a re-scaled interfacial friction\ncoefficient.", "category": "physics_flu-dyn" }, { "text": "Analysis of vortex breakdown in an enclosed cylinder based on the energy\n gradient theory: Numerical simulation is carried out to study the phenomenon of vortex\nbreakdown in an enclosed cylinder. The energy gradient theory is used to\nexplain the vortex breakdown in the cylinder with consideration of centrifugal\nforce, Coriolis force, angular momentum and azimuthal vorticity. The research\nresults show that the large value of energy gradient function K is mainly\nlocated at the centerline and the region between the circulation vortices on\nboth sides of the cylinder and the vortex breakdown bubbles at the centerline.\nIt is found that the position of the local peak value of the energy gradient\nfunction K at the centerline corresponds to the location of vortex breakdown\nfirst occurrence. The position of the local peak value of K function in\nhorizontal direction corresponds to the velocity inflection points except for\nthe centerline. The vortex breakdown is mainly determined by the high K value\nat the centerline for low aspect ratio. The influence of the region of high K\nvalue between the circulation vortices on both sides of the cylinder and the\nvortex breakdown bubbles at the centerline becomes larger with the increase of\nthe aspect ratio. The occurrence and development of the vortex breakdown bubble\nmay be affected by the region of high K value between the circulation vortices\non both sides of the cylinder and the vortex breakdown bubbles at the\ncenterline for high aspect ratio.", "category": "physics_flu-dyn" }, { "text": "Capillary-gravity wave transport over spatially random drift: We derive transport equations for the propagation of water wave action in the\npresence of a static, spatially random surface drift. Using the Wigner\ndistribution $\\W(\\x,\\k,t)$ to represent the envelope of the wave amplitude at\nposition $\\x$ contained in waves with wavevector $\\k$, we describe surface wave\ntransport over static flows consisting of two length scales; one varying\nsmoothly on the wavelength scale, the other varying on a scale comparable to\nthe wavelength. The spatially rapidly varying but weak surface flows augment\nthe characteristic equations with scattering terms that are explicit functions\nof the correlations of the random surface currents. These scattering terms\ndepend parametrically on the magnitudes and directions of the smoothly varying\ndrift and are shown to give rise to a Doppler coupled scattering mechanism. The\nDoppler interaction in the presence of slowly varying drift modifies the\nscattering processes and provides a mechanism for coupling long wavelengths\nwith short wavelengths. Conservation of wave action (CWA), typically derived\nfor slowly varying drift, is extended to systems with rapidly varying flow. At\nyet larger propagation distances, we derive from the transport equations, an\nequation for wave energy diffusion. The associated diffusion constant is also\nexpressed in terms of the surface flow correlations. Our results provide a\nformal set of equations to analyse transport of surface wave action, intensity,\nenergy, and wave scattering as a function of the slowly varying drifts and the\ncorrelation functions of the random, highly oscillatory surface flows.", "category": "physics_flu-dyn" }, { "text": "Analysis of Resonance in Jet Screech with Large-Eddy Simulations: Screech resonance is studied with experimentally validated large-eddy\nsimulation data for a 4:1 rectangular under-expanded jet at three nozzle\npressure ratios. The analysis uses spectral proper orthogonal decomposition\n(SPOD) and spatial cross correlation to characterize the oppositely-traveling\nwaves in the jet at the screech fundamental frequency. The results support\nrecent theoretical framing of screech as absolute instability, and further\nreveal the spatial separation of individual processes for screech generation.\nFrom the leading-order SPOD mode, direct evidence of the guided jet mode being\nthe screech closure mechanism, not the external acoustic feedback, is observed.\nA match in the spatial wavenumber suggests the guided jet mode is generated via\ninteractions between the Kelvin-Helmholtz wave and the shock cells. The energy\nof the oppositely-moving waves shows spatially global and non-periodic behavior\nof the coherent structures in the streamwise direction. The ratio of wave\nenergy identifies regions where distinct processes in screech generation take\nplace by comparing the rate of energy propagation in the downstream direction\nto that of the upstream direction. The distinct regions correspond to initial\nshear layer receptivity, sound emission, guided jet mode excitation and decay\nof coherence. The leading-order SPOD mode also enables the approximation of\nLighthill's stress tensor and allows for accurate calculation of the far-field\nscreech tone amplitude with the acoustic analogy formulation. The current\nfindings provide insights on building a physics-based reduced order model for\nscreech amplitude prediction in the future.", "category": "physics_flu-dyn" }, { "text": "Particle and thermo-hydraulic maldistribution of nanofluids in parallel\n microchannel systems: Fluidic maldistribution in microscale multichannel devices requires deep\nunderstanding to achieve optimized flow and heat transfer characteristics. A\nthorough computational study has been performed to understand the concentration\nand thermohydraulic maldistribution of nanofluids in parallel microchannel\nsystems using an Eulerian Lagrangian twin phase model. The study reveals that\nnanofluids cannot be treated as homogeneous single phase fluids in such complex\nflow domains and effective property models fail drastically to predict the\nperformance parameters. To comprehend the distribution of the particulate\nphase, a novel concentration maldistribution factor has been proposed. It has\nbeen observed that distribution of particles need not essentially follow the\nflow pattern, leading to higher thermal performance than expected from\nhomogeneous models. Particle maldistribution has been conclusively shown to be\ndue to various migration and diffusive phenomena like Stokesian drag, Brownian\nmotion, thermophoretic drift, etc. The implications of particle distribution on\nthe cooling performance have been illustrated and smart fluid effects (reduced\nmagnitude of maximum temperature) have been observed and a mathematical model\nto predict the enhanced cooling performance in such flow geometries has been\nproposed. The article presents lucidly the effectiveness of discrete phase\napproach in modelling nanofluid thermohydraulics and sheds insight on behavior\nof nanofluids in complex flow domains.", "category": "physics_flu-dyn" }, { "text": "Two-dimensional pulse dynamics and the formation of bound states on\n electrified falling films: The flow of an electrified liquid film down an inclined plane wall is\ninvestigated with the focus on coherent structures in the form of travelling\nwaves on the film surface, in particular, single-hump solitary pulses and their\ninteractions. The flow structures are analysed first using a long-wave model,\nwhich is valid in the presence of weak inertia, and second using the Stokes\nequations. For obtuse angles, gravity is destablising and solitary pulses exist\neven in the absence of an electric field. For acute angles, spatially\nnon-uniform solutions exist only beyond a critical value of the electric field\nstrength; moreover, solitary-pulse solutions are present only at sufficiently\nhigh supercritical elec- tric field strengths. The electric field increases the\namplitude of the pulses, can generate recirculation zones in the humps, and\nalters the far-field decay of the pulse tails from ex- ponential to algebraic\nwith a significant impact on pulse interactions. A weak-interaction theory\npredicts an infinite sequence of bound-state solutions for non-electrified\nflow, and a finite set for electrified flow. The existence of single-hump pulse\nsolutions and two-pulse bound states is confirmed for the Stokes equations via\nboundary-element computations. In addition, the electric field is shown to\ntrigger a switch from absolute instability to convective instability, thereby\nregularising the dynamics, and this is confirmed by time- dependent simulations\nof the long-wave model.", "category": "physics_flu-dyn" }, { "text": "On Satisfying the Kutta Condition in Unsteady Thin Aerofoil Theory: Unsteady thin-aerofoil theory is a low-order method for solving\npotential-flow aerodynamics on a camber-line undergoing arbitrary motion. In\nthis method, a Kutta condition must be applied at the trailing edge to uniquely\nspecify the net circulation about the aerofoil. This article provides a\ncritical discussion on applying the Kutta condition in unsteady flows, and\nintroduces an improved method of doing so in unsteady thin-aerofoil theory.\nSpecifically, the shed wake at any discrete time step is represented by a\ncontinuous distribution of vorticity derived from the exact Wagner solution\nrather than by a point vortex or regularized vortex blob. Results in the\narticle illustrate the effects of this improvement for cases of step change in\nangle of attack (Wagner problem), harmonic heaving motion (Theodorsen problem),\nand a pitch-ramp-hold manoeuvre. Exact analytical solutions and CFD simulations\nof the incompressible Euler equations are used for verification. The new\napproach is seen to satisfy the Kutta condition at all reduced frequencies,\nwith velocities being finite and pressure difference going to zero at the\ntrailing edge. It improves unsteady thin-aerofoil theory in terms of\ntheoretical rigour, computational cost and numerical accuracy.", "category": "physics_flu-dyn" }, { "text": "Intrinsic viscosity of a suspension of weakly Brownian ellipsoids in\n shear: We analyze the angular dynamics of triaxial ellipsoids in a shear flow\nsubject to weak thermal noise. By numerically integrating an overdamped angular\nLangevin equation, we find the steady angular probability distribution for a\nrange of triaxial particle shapes. From this distribution we compute the\nintrinsic viscosity of a dilute suspension of triaxial particles. We determine\nhow the viscosity depends on particle shape in the limit of weak thermal noise.\nWhile the deterministic angular dynamics depends very sensitively on particle\nshape, we find that the shape dependence of the intrinsic viscosity is weaker,\nin general, and that suspensions of rod-like particles are the most sensitive\nto breaking of axisymmetry. The intrinsic viscosity of a dilute suspension of\ntriaxial particles is smaller than that of a suspension of axisymmetric\nparticles with the same volume, and the same ratio of major to minor axis\nlengths.", "category": "physics_flu-dyn" }, { "text": "The effect of $Re_\u03bb$ and Rouse numbers on the settling of inertial\n particles in homogeneous isotropic turbulence: We present an experimental study on the settling velocity of dense\nsub-Kolmogorov particles in active-grid-generated turbulence in a wind tunnel.\nUsing phase Doppler interferometry, we observe that the modifications of the\nsettling velocity of inertial particles, under homogeneous isotropic turbulence\nand dilute conditions $\\phi_v\\leq O(10)^{-5}$, is controlled by the\nTaylor-based Reynolds number $Re_\\lambda$ of the carrier flow. On the contrary,\nwe did not find a strong influence of the ratio between the fluid and gravity\naccelerations (i.e., $\\gamma\\sim(\\eta/\\tau_\\eta^2)/g$) on the particle settling\nbehavior. Remarkably, our results suggest that the hindering of the settling\nvelocity (i.e. the measured particle settling velocity is smaller than its\nrespective one in still fluid conditions) experienced by the particles\nincreases with the value of $Re_\\lambda$, reversing settling enhancement found\nunder intermediate $Re_\\lambda$ conditions. This observation applies to all\nparticle sizes investigated, and it is consistent with previous experimental\ndata in the literature. At the highest $Re_\\lambda$ studied, $Re_\\lambda>600$,\nthe particle enhancement regime ceases to exist. Our data also show that for\nmoderate Rouse numbers, the difference between the measured particle settling\nvelocity and its velocity in still fluid conditions scales linearly with Rouse,\nwhen this difference is normalized by the carrier phase rms fluctuations, i.e.,\n$(V_p-V_T)/u\\sim -Ro$.", "category": "physics_flu-dyn" }, { "text": "A theoretical study of the air-sea drag-saturation in very strong winds: The goal of this note is to provide a theoretical explanation for the\nsaturation of the drag coefficient in strong wind conditions. The hydrodynamic\nmodel under consideration takes into account the important effects of airborne\ndroplets of water in a thin layer above the water surface that effectively\nbehave as a different fluid between the water and the air. Above this layer the\nmodel is coupled with a log-wind profile for the strong winds blowing above the\nsea. The main underlying mechanism governing the behavior of the drag\ncoefficient is the Kelvin Helmholtz instability for capillary waves on the\nwater surface and the continuity of shear stress along the intermediate\ninterface.", "category": "physics_flu-dyn" }, { "text": "Physics-Aware Neural Network Flame Closure for Combustion Instability\n Modeling in a Single-Injector Engine: Neural networks (NN) are implemented as sub-grid flame models in a large-eddy\nsimulation of a single-injector liquid-propellant rocket engine with the aim to\nreplace a look-up table approach. The NN training process presents an\nextraordinary challenge. The multi-dimensional combustion instability problem\ninvolves multi-scale lengths and characteristic times in an unsteady flow\nproblem with nonlinear acoustics, addressing both transient and\ndynamic-equilibrium behaviors, superimposed on a turbulent reacting flow with\nvery narrow, moving flame regions. Accurate interpolation between the points of\nthe training data becomes vital. A major novel aspect of the proposed NNs is\nthat they are trained to reproduce relevant portions of the information stored\nin a flamelet table by using only limited data from a few CFD simulations of a\nsingle-injector liquid-propellant rocket engine under different dynamical\nconfigurations. This is made possible by enriching the training set with\ncontrived data resulting from the physical characteristics of the combustion\nmodel and also by including the flame temperature as an extra input to the NNs\nthat are trained to model other flame variables of interest. These\nphysics-aware NN-based closure models are first tested offline by comparing\nthem directly with the flamelet table and then are successfully implemented\ninto CFD simulations in place of the flamelet table and verified on various\ndynamical configurations. The results from those tests compare favorably with\ncounterpart table-based CFD simulations. Computational advantages of the\napproach are discussed.", "category": "physics_flu-dyn" }, { "text": "On cascade of kinetic energy in compressible hydrodynamic turbulence: Properties of the turbulent cascade of kinetic energy are studied using\ndirect numerical simulations of three-dimensional hydrodynamic decaying\nturbulence with a moderate Reynolds number and the initial Mach number $M=1$.\nCompressible and incompressible versions of the Karman-Howarth-Monin (KHM) and\nlow-pass filtering/coarse-graining approaches are compared. In the simulation\nthe total energy is well conserved; the scale dependent KHM and coarse-grained\nenergy equations are also well conserved; the two approaches show similar\nresults, the system does not have an inertial range for the cascade of kinetic\nenergy, the region where this cascade dominates also have a non-negligible\ncontribution of the kinetic-energy decay, dissipation, and pressure-dilatation\neffects. While the two approaches give semi-quantitatively similar results for\nthe kinetic energy cascade, dissipation and pressure-dilatation rates, they\ndiffer in the increment separation and filtering scales; these scales are not\nsimply related. The two approaches may be used to find the inertial range and\nto determine the cascade/dissipation rate of the kinetic energy.", "category": "physics_flu-dyn" }, { "text": "Vortices around Dragonfly Wings: Dragonfly beats its wings independently, resulting in its superior\nmaneuverability. Depending on the magnitude of phase difference between the\nfore- and hind-wings of dragonfly, the vortical structures and their\ninteraction with wings become significantly changed, and so does the\naerodynamic performance. In this study, we consider hovering flights of\nmodelled dragonfly with three different phase differences (phi=-90, 90, 180\ndegrees). The three-dimensional wing shape is based on that of Aeschna juncea\n(Norberg, 1972), and the Reynolds number is 1,000 based on the maximum\ntranslational velocity and mean chord length. The numerical method is based on\nan immersed boundary method (Kim et al., 2001). In counter-stroke (phi=180\ndegree), the wing-tip vortices from both wings are connected in the wake,\ngenerating an entangled wing-tip vortex (e-WTV). A strong downward motion\ninduced by this vortex decreases the lift force in the following downstroke\n(Kweon and Choi, 2008). When the fore-wing leads the hind-wing (phi=90 degree),\nthe hind-wing is submerged in the vortices generated by the fore-wing and\nsuffers from their induced downwash flow throughout the downstroke, resulting\nin a significant reduction of lift force. On the other hand, when the hind-wing\nleads the fore-wing (phi=-90 degree), the e-WTV is found only near the start of\nhind-wing upstroke. In the following downstroke of hind-wing, most of the e-WTV\ndisappears and the hind-wing is little affected by this vortex, which produces\nrelatively large lift force.", "category": "physics_flu-dyn" }, { "text": "Fundamental solutions of MHD Stokes flow: A simple analytical solution is obtained for the MHD stokeslet in a\nhomogeneous magnetic field. This solution represents the flow past a small\nparticle and can also be interpreted as the flow sufficiently far away from a\nbody of finite size. Fundamental solutions are found in terms of velocity,\npressure and scalar potential distributions for the flows due to either a\nconcentrated force or a current source. The former consists of two basic\nsolutions for the force parallel and transverse to the magnetic field,\nrespectively. All fundamental solutions have the characteristic length scale of\nthe Hartmann boundary layer and two parabolic wakes developing along the\nmagnetic field.", "category": "physics_flu-dyn" }, { "text": "Flows and dynamos in a model of stellar radiative zones: Stellar radiative zones are typically assumed to be motionless in standard\nmodels of stellar structure but there is sound theoretical and observational\nevidence that this cannot be the case. We investigate by direct numerical\nsimulations a three-dimensional and time-dependent model of stellar radiation\nzones consisting of an electrically-conductive and stably-stratified anelastic\nfluid confined to a rotating spherical shell and driven by a baroclinic torque.\nAs the baroclinic driving is gradually increased a sequence of transitions from\nan axisymmetric and equatorially-symmetric time-independent flow to flows with\na strong poloidal component and lesser symmetry are found. It is shown that all\nflow regimes characterised with significant non-axisymmetric components are\ncapable of generating self-sustained magnetic field. As the value of the\nPrandtl number is decreased and the value of the Ekman number is decreased\nflows become strongly time-dependent with progressively complex spatial\nstructure and dynamos can be generated at lower values of the magnetic Prandtl\nnumber.", "category": "physics_flu-dyn" }, { "text": "Transitional channel flow: A minimal stochastic model: In line with Pomeau's conjecture about the relevance of directed percolation\n(DP) to turbulence onset/decay in wall-bounded flows, we propose a minimal\nstochastic model dedicated to the interpretation of the spatially intermittent\nregimes observed in channel flow before its return to laminar flow. Numerical\nsimulations show that a regime with bands obliquely drifting in two stream-wise\nsymmetrical directions bifurcates into an asymmetrical regime, before\nultimately decaying to laminar flow. The model is expressed in terms of a\nprobabilistic cellular automaton evolving von Neumann neighbourhoods with\nprobabilities educed from a close examination of simulation results. It\nimplements band propagation and the two main local processes: longitudinal\nsplitting involving bands with the same orientation, and transversal splitting\ngiving birth to a daughter band with orientation opposite to that of its\nmother. The ultimate decay stage observed to display one-dimensional DP\nproperties in a two-dimensional geometry is interpreted as resulting from the\nirrelevance of lateral spreading in the single-orientation regime. The model\nalso reproduces the bifurcation restoring the symmetry upon variation of the\nprobability attached to transversal splitting, which opens the way to a study\nof the critical properties of that bifurcation, in analogy with thermodynamic\nphase transitions.", "category": "physics_flu-dyn" }, { "text": "Geometry of elastic hydrofracturing by injection of an over pressured\n non-Newtonian Fluid: The nucleation and propagation of hydrofractures by injection of over\npressured fluids in an elastic and isotropic medium are studied experimentally.\nNon-Newtonian fluids are injected inside a gelatine whose mechanical properties\nare assumed isotropic at the experimental strain rates. Linear elastic theory\npredicts that plastic deformation associated to breakage of gelatin bonds is\nlimited to a small zone ahead of the tip of the propagating fracture and that\npropagation will be maintained while the fluid pressure exceeds the normal\nstress to the fracture walls (Ch\\'avez-\\'Alvarez,2008) (i.e., the minimum\ncompressive stress), resulting in a single mode I fracture geometry. However,\nwe observed the propagation of fractures type II and III as well as nucleation\nof secondary fractures, with oblique to perpendicular trajectories with respect\nto the initial fracture. In the Video (http://hdl.handle.net/1813/14122)\nexperimental evidence shows that the fracture shape depends on the viscoelastic\nproperties of gelatine coupled with the strain rate achieved by fracture\npropagation.", "category": "physics_flu-dyn" }, { "text": "Turbulent kinematic dynamos in ellipsoids driven by mechanical forcing: Dynamo action in planetary cores has been extensively studied in the context\nof convectively-driven flows. We show in this letter that mechanical forcings,\nnamely tides, libration and precession, are also able to kinematically sustain\na magnetic field against ohmic diffusion. Previous attempts published in the\nliterature focused on the laminar response or considered idealized spherical\nconfigurations. In contrast, we focus here on the developed turbulent regime\nand we self-consistently solve the magnetohydrodynamic (MHD) equations in an\nellipsoidal container. Our results open new avenues of research in dynamo\ntheory where both convection and mechanical forcing can play a role,\nindependently or simultaneously.", "category": "physics_flu-dyn" }, { "text": "Unsteady Stokes flow near boundaries: the point-particle approximation\n and the method of reflections: Problems of particle dynamics involving unsteady Stokes flows in confined\ngeometries are typically harder to solve than their steady counterparts.\nApproximation techniques are often the only resort. Felderhof (see e.g. 2005,\n2009b) has developed a point-particle approximation framework to solve such\nproblems, especially in the context of Brownian motion. Despite excellent\nagreement with past experiments, this framework has an inconsistency which we\naddress in this work. Upon implementing our modifications, the framework passes\nconsistency checks that it previously failed. Further, it is not obvious that\nsuch an approximation should work for short time-scale motion. We investigate\nits validity by deriving it from a general formalism based on integral\nequations through a series of systematic approximations. We also compare\nresults from the point-particle framework against a calculation performed using\nthe method of reflections, for the specific case of a sphere near a full-slip\nplane boundary. We find from our analysis that the reasons for the success of\nthe point-particle approximation are subtle and have to do with the nature of\nthe unsteady Oseen tensor. Finally, we provide numerical predictions for\nBrownian motion near a full-slip and a no-slip plane wall based on the\npoint-particle approximation as used by Felderhof, our modified point-particle\napproximation, and the method of reflections. We show that our modifications to\nFelderhof's framework would become significant for systems of metallic\nnanoparticles in liquids.", "category": "physics_flu-dyn" }, { "text": "Time-averaged transport in oscillatory squeeze flow of a viscoelastic\n fluid: Periodically-driven flows are known to generate non-zero, time-averaged\nfluxes of heat or solute species, due to the interactions of out-of-phase\nvelocity and temperature/concentration fields, respectively. Herein, we\ninvestigate such transport (a form of the well-known Taylor--Aris dispersion)\nin the gap between two parallel plates, one of which oscillates vertically,\ngenerating a time-periodic squeeze flow of either a newtonian or Maxwellian\nfluid. Using the method of multiple time-scale homogenization, the mass/heat\nbalance equation describing transport in this flow is reduced to a\none-dimensional advection--diffusion--reaction equation. This result indicates\nthree effective mechanisms in the mass/heat transfer in the system: an\neffective diffusion that spreads mass/heat along the concentration/temperature\ngradient, an effective advective flux, and an effective reaction that releases\nor absorbs mass/heat - in the time-averaged frame. Our results demonstrate that\nthere exist resonant modes under which the velocity peaks when the\ndimensionless plate oscillation frequency (embodied by the Womersley number,\nthe ratio of the transient inertia to viscous forces) approaches specific\nvalues. As a result, transport in this flow is significantly influenced by the\ndimensionless frequency. On the one hand, the effective, time-averaged\ndispersion coefficient is always larger than the molecular diffusivity, and is\nsharply enhanced near resonance. The interaction between fluid elasticity and\nthe oscillatory forcing enhances the efficiency of transport in the system. On\nthe other hand, the identified effective advection and reaction mechanisms may\ntransport mass/heat from regions of high concentration/temperature to those of\nlow concentration/temperature, or vice versa, depending on the value of\ndimensionless frequency.", "category": "physics_flu-dyn" }, { "text": "Robust estimate of dynamo thresholds in the von K\u00e1rm\u00e1n sodium\n experiment using the Extreme Value Theory: We apply a new threshold detection method based on the extreme value theory\nto the von K\\'arm\\'an sodium (VKS) experiment data. The VKS experiment is a\nsuccessful attempt to get a dynamo magnetic field in a laboratory liquid-metal\nexperiment. We first show that the dynamo threshold is associated to a change\nof the probability density function of the extreme values of the magnetic\nfield. This method does not require the measurement of response functions from\napplied external perturbations, and thus provides a simple threshold estimate.\nWe apply our method to different configurations in the VKS experiment showing\nthat it yields a robust indication of the dynamo threshold as well as evidence\nof hysteretic behaviors. Moreover, for the experimental configurations in which\na dynamo transition is not observed, the method provides a way to extrapolate\nan interval of possible threshold values.", "category": "physics_flu-dyn" }, { "text": "The Significance of Simple Invariant Solutions in Turbulent Flows: Recent remarkable progress in computing power and numerical analysis is\nenabling us to fill a gap in the dynamical systems approach to turbulence. One\nof the significant advances in this respect has been the numerical discovery of\nsimple invariant sets, such as nonlinear equilibria and periodic solutions, in\nwell-resolved Navier--Stokes flows. This review describes some fundamental and\npractical aspects of dynamical systems theory for the investigation of\nturbulence, focusing on recently found invariant solutions and their\nsignificance for the dynamical and statistical characterization of\nlow-Reynolds-number turbulent flows. It is shown that the near-wall\nregeneration cycle of coherent structures can be reproduced by such solutions.\nThe typical similarity laws of turbulence, i.e. the Prandtl wall law and the\nKolmogorov law for the viscous range, as well as the pattern and intensity of\nturbulence-driven secondary flow in a square duct can also be represented by\nthese simple invariant solutions.", "category": "physics_flu-dyn" }, { "text": "Transition of Planar Couette Flow at infinite Reynolds numbers: An outline of the state space of planar Couette flow at high Reynolds numbers\n($Re < 10^5$) is investigated via a variety of efficient numerical techniques.\nIt is verified from nonlinear analysis that the lower branch of {\\it Hairpin\nVortex State} (HVS) asymptotically approaches the primary (laminar) state with\nincreasing $Re$. It is also predicted that the lower branch of HVS at high $Re$\nbelongs to the stability boundary that initiates transition to turbulence, and\nthat one of the unstable manifolds of the lower branch of HVS lies on the\nboundary. These facts suggest HVS may provide a criterion to estimate a minimum\nperturbation arising transition to turbulent states at the infinite $Re$ limit.", "category": "physics_flu-dyn" }, { "text": "Additive Layers: An Alternate Classification of Flow Regimes: It is argued that ejections of wall fluid in the bursting process disturb the\nflow beyond the wall layer and result in the emergence of two new layers in the\nflow field: the law of the wake and log-law layers. The wall layer represents\nthe extent of penetration of viscous momentum into the main flow and remains\nconstant once it has reached a critical value at the end of the laminar regime.\nThe identification of the three flow regimes: laminar, transition and fully\nturbulent is conveniently achieved by monitoring the emergence of these three\nlayers", "category": "physics_flu-dyn" }, { "text": "Some theoretical results on the second-order conservative phase field\n equation: In this paper, a theoretical research on the second-order conservative phase\nfield (SOCPF) equation is presented. The theoretical results include the\nfollowing three aspects. First, three new derivation methods for the SOCPF\nequation are given. The SOCPF equation can be viewed as the gradient flow, the\nspecial diffusion equation and the diffuse interface form of a sharp interface\nformulation for the piecewise constant function, respectively. These derivation\nmethods help us to understand the SOCPF equation at different perspectives.\nSecond, the conservation's properties of the solution of SOCPF equation are\nstudied. Compared with the Cahn-Hilliard equation and the Allen-Cahn equation,\nit is found that the solution of SOCPF equation satisfies more conservation\nlaws. Third, the wetting boundary condition for the SOCPF equation is\ninvestigated. We find that the no-flux boundary condition is equivalent to the\nwetting boundary condition for two-component phase field model. Moreover,\napplying the no-flux boundary conditions for $N$-component phase field model,\nwe give a set of wetting boundary conditions for $N$ phase field parameters.", "category": "physics_flu-dyn" }, { "text": "Acoustic streaming and its suppression in inhomogeneous fluids: We present a theoretical and experimental study of boundary-driven acoustic\nstreaming in an inhomogeneous fluid with variations in density and\ncompressibility. In a homogeneous fluid this streaming results from dissipation\nin the boundary layers (Rayleigh streaming). We show that in an inhomogeneous\nfluid, an additional non-dissipative force density acts on the fluid to\nstabilize particular inhomogeneity configurations, which markedly alters and\neven suppresses the streaming flows. Our theoretical and numerical analysis of\nthe phenomenon is supported by ultrasound experiments performed with\ninhomogeneous aqueous iodixanol solutions in a glass-silicon microchip.", "category": "physics_flu-dyn" }, { "text": "Analysis of anisotropic subgrid-scale stress for coarse large-eddy\n simulation: This study discusses the necessity of anisotropic subgrid-scale (SGS) stress\nin large-eddy simulations (LESs) of turbulent shear flows using a coarse grid\nresolution. We decompose the SGS stress into two parts to observe the role of\nSGS stress in turbulent shear flows in addition to the energy transfer between\ngrid-scale (GS or resolved scale) and SGS. One is the isotropic eddy-viscosity\nterm, which contributes to energy transfer, and the other is the residual\nanisotropic term, which is separated from the energy transfer. We investigate\nthe budget equation for GS Reynolds stress in turbulent channel flows\naccompanied by the SGS stress decomposition. In addition, we examine the medium\nand coarse filter length cases; the conventional eddy-viscosity models can\nfairly predict the mean velocity profile for the medium filter case and fails\nfor the coarse filter case. The budget for GS turbulent kinetic energy shows\nthat the anisotropic SGS stress has a negligible contribution to energy\ntransfer. In contrast, the anisotropic stress has a large and non-dissipative\ncontribution to the streamwise and spanwise components of GS Reynolds stress\nwhen the filter size is large. Even for the medium-size filter case, the\nanisotropic stress contributes positively to the budget for the spanwise GS\nReynolds stress. Spectral analysis of the budget reveals that the positive\ncontribution is prominent at a scale consistent with the spacing of streaks in\nthe near-wall region. Therefore, we infer that anisotropic stress contributes\nto the generation mechanism of coherent structures. Predicting the positive\ncontribution of the anisotropic stress to the budget is key to further\nimproving SGS models.", "category": "physics_flu-dyn" }, { "text": "Data-driven Modeling of Two-Dimensional Detonation Wave Fronts: Historical experimental testing of high-altitude nuclear explosions (HANEs)\nare known to cause severe and detrimental effects to radio frequency signals\nand communications infrastructure. In order to study and predict the impact of\nHANEs, tractable computational approaches are required to model the complex\nphysical processes involved in the detonation wave physics. Modern\nreduced-order models (ROMs) can enable long-time and many-parameter simulations\nwith minimal computational cost. However, translational and scale invariances\ninherent to this type of wave propagation problem are known to limit\ntraditional ROM approaches. Specifically, dimensionality reduction methods are\ntypically ineffective in producing low-rank models when invariances are present\nin the data. In this work, an unsupervised machine learning method is used to\ndiscover coordinate systems that make such invariances amenable to traditional\ndimensionality reduction methods. The method, which has previously been\ndemonstrated on one-dimensional translations, is extended to higher dimensions\nand additional invariances. A surrogate HANE system, i.e. a HANE-ROM, with one\ndetonation wave is captured well at extremely low-rank. Two detonation-waves\nare also considered with various amounts of interaction between the waves, with\nimprovements to low-rank models for multiple wave quantities with limited\ninteraction.", "category": "physics_flu-dyn" }, { "text": "Impact of Engine Nacelle Flow on Buffet: The transonic flow around the OAT15A airfoil is computed at buffet\nconditions, i.e., freestream Mach number $Ma_\\infty = 0.73$, chord-based\nfreestream Reynolds number $Re_c = 2\\cdot10^6$, and angle of attack $\\alpha =\n3.5^\\circ$ using wall-modeled LES. Two configurations are considered, one which\nincludes a generic ultra-high bypass ratio (UHBR) engine nacelle geometry and\none without an engine, which is denoted the baseline case. The introduction of\nthe UHBR-engine nacelle leads to a significant deviation of the flow onto the\nairfoil from the baseline case and has an essential effect on the occurring\nshock dynamics. The flow field of the nacelle configuration is characterized by\na shock wave on the upper part of the nacelle sharing dynamic features with the\nshock on the airfoil. This impact of the nacelle shock on the airfoil shock\nmeans a reduced strength of the airfoil shock resulting in a less developed\nbuffet. The perturbation of the general flow field is evaluated as to\nestablished buffet models and the dynamic features of the shock waves are\nanalyzed by the sparsity-promoting dynamic mode decomposition. This analysis\nshows the existence of a shared dynamic mode of the nacelle and the airfoil\nshock which suggests a coupling mechanism between them.", "category": "physics_flu-dyn" }, { "text": "Relative dispersion of particle pairs in turbulent channel flow: Lagrangian tracking of particle pairs is of fundamental interest in a large\nnumber of environmental applications dealing with contaminant dispersion and\npassive scalar mixing. The aim of the present study is to extend the\nobservations available in the literature on relative dispersion of fluid\nparticle pairs to wall-bounded turbulent flows, by means of particle pair\ntracking in direct numerical simulations (DNS) of a turbulent channel flow. The\nmean-square change of separation between particle pairs follows a clear\nballistic regime at short times for all wall distances. The Eulerian structure\nfunctions governing this short-time separation are characterised in the\nchannel, and allow to define a characteristic time scale for the ballistic\nregime, as well as a suitable normalisation of the mean-square separation\nleading to an overall collapse for different wall distances. Through fluid\nparticle pair tracking backwards and forwards in time, the temporal asymmetry\nof relative dispersion is illustrated. At short times, this asymmetry is linked\nto the irreversibility of turbulence, as in previous studies on homogeneous\nisotropic flows. The influence of the initial separation (distance and\norientation) as well as the influence of mean shear are addressed. By\ndecomposing the mean-square separation into the dispersion by the fluctuating\nvelocity field and by the average velocity, it is shown that the influence of\nmean shear becomes important at early stages of dispersion close to the wall\nbut also near the channel centre. The relative dispersion tensor $\\Delta_{ij}$\nis also presented and particularly the sign and time evolution of the\ncross-term $\\Delta_{xy}$ are discussed. Finally, a ballistic cascade model\npreviously proposed for homogeneous isotropic turbulence is adapted here to\nturbulent channel flows. Preliminary results are given and compared to the DNS.", "category": "physics_flu-dyn" }, { "text": "Investigation of a lattice Boltzmann model with a variable speed of\n sound: A lattice Boltzmann model is considered in which the speed of sound can be\nvaried independently of the other parameters. The range over which the speed of\nsound can be varied is investigated and good agreement is found between\nsimulations and theory. The onset of nonlinear effects due to variations in the\nspeed of sound is also investigated and good agreement is again found with\ntheory. It is also shown that the fluid viscosity is not altered by changing\nthe speed of sound.", "category": "physics_flu-dyn" }, { "text": "DNS of Turbulent Flows Laden with Droplets or Bubbles: This review focuses on Direct numerical simulations (DNS) of turbulent flows\nladen with droplets or bubbles. DNS of these flows are more challenging than\nthose of flows laden with solid particles due to the surface deformation in the\nformer. The classification of the discussed numerical methods is based on\nwhether the initial diameter of the bubble/droplet is smaller or larger than\nthe Kolmogorov length scale, and whether the instantaneous surface deformation\nis fully resolved or obtained via a phenomenological model. Numerical methods\nthat account for the breakup of single droplet/bubble as well as multiple\ndroplet/bubble in canonical turbulent flows are also discussed.", "category": "physics_flu-dyn" }, { "text": "A playground for compressible natural convection with a nearly uniform\n density: In the quest to understand the basic universal features of compressible\nconvection, one would like to disentangle genuine consequences of compression\nfrom spatial variations of transport properties. In the present work, we\nconsider a very peculiar equation of state, whereby entropy is solely dependent\non density, so that a nearly isentropic fluid domain is nearly isochoric.\nWithin this class of equations of state, there is a thermal adiabatic gradient\nand a key property of compressible convection is still present, namely its\ncapacity to viscously dissipate a large fraction of the thermal energy\ninvolved, of the order of the well-named dissipation number. In a series of\nanelastic approximations, under the assumption of an infinite Prandtl number,\nthe number of governing parameters can be brought down to two, the Rayleigh\nnumber and the dissipation number. This framework is proposed as a playground\nfor compressible convection, an opportunity to extend the vast corpus of\ntheoretical analyses on the Oberbeck-Boussinesq equations regarding stability,\nbifurcations or the determination of upper bounds for the turbulent heat\ntransfer. Here, in a two-dimensional geometry, we concentrate on the structure\nof upward and downward plumes depending on the dissipation number, on the heat\nflux dependence on the dissipation number and on the ratio of dissipation to\nconvective heat flux. For dissipation numbers of order unity, in the limit of\nlarge Rayleigh numbers, dissipation becomes related to the entropy heat flux at\neach depth, so that the vertical dissipation profile can be predicted, and\nconsequently so does the total ratio of dissipation to convective heat flux.", "category": "physics_flu-dyn" }, { "text": "A fractional PDE model for turbulent velocity fields near solid walls: This paper presents a class of turbulence models written in terms of\nfractional partial differential equations (FPDEs) with stochastic loads. Every\nsolution of these FPDE models is an incompressible velocity field and the\ndistribution of solutions is Gaussian. Interaction of the turbulence with solid\nwalls is incorporated through the enforcement of various boundary conditions.\nThe various boundary conditions deliver extensive flexibility in the near-wall\nstatistics that can be modelled. Reproduction of both fully-developed\nshear-free and uniform shear boundary layer turbulence are highlighted as two\nsimple physical applications; the first of which is also directly validated\nwith experimental data. The rendering of inhomogeneous synthetic turbulence\ninlet boundary conditions is an additional application, motivated by\ncontemporary numerical wind tunnel simulations. Calibration of model parameters\nand efficient numerical methods are also conferred upon.", "category": "physics_flu-dyn" }, { "text": "Modelling Settling-Driven Gravitational Instabilities at the Base of\n Volcanic Clouds Using the Lattice Boltzmann Method: Field observations and laboratory experiments have shown that ash\nsedimentation can be significantly affected by collective settling mechanisms\nthat promote premature ash deposition, with important implications for\nassociated impacts. Among these mechanisms, settling-driven gravitational\ninstabilities result from the formation of a gravitationally-unstable particle\nboundary layer (PBL) that grows between volcanic ash clouds and the underlying\natmosphere. The PBL destabilises once it reaches a critical thickness,\ntriggering the formation of rapid, downward-moving ash fingers that remain\npoorly characterised. We simulate this process by coupling a Lattice Boltzmann\nmodel, which solves the Navier-Stokes equations for the fluid phase, with a\nWeighted Essentially Non Oscillatory (WENO) finite difference scheme which\nsolves the advection-diffusion-settling equation describing particle transport.\nSince the physical problem is advection dominated, the use of the WENO scheme\nreduces numerical diffusivity and ensures accurate tracking of the temporal\nevolution of the interface between the layers. We have validated the new model\nby showing that the simulated early-time growth rate of the instability is in\nvery good agreement with that predicted by linear stability analysis, whilst\nthe modelled late-stage behaviour also successfully reproduces quantitative\nresults from published laboratory experiments.", "category": "physics_flu-dyn" }, { "text": "Detailed analysis of the lattice Boltzmann method on unstructured grids: The lattice Boltzmann method has become a standard for efficiently solving\nproblems in fluid dynamics. While unstructured grids allow for a more efficient\ngeometrical representation of complex boundaries, the lattice Boltzmann methods\nis often implemented using regular grids. Here we analyze two implementations\nof the lattice Boltzmann method on unstructured grids, the standard forward\nEuler method and the operator splitting method. We derive the evolution of the\nmacroscopic variables by means of the Chapman-Enskog expansion, and we prove\nthat it yields the Navier-Stokes equation and is first order accurate in terms\nof the temporal discretization and second order in terms of the spatial\ndiscretization. Relations between the kinetic viscosity and the integration\ntime step are derived for both the Euler method and the operator splitting\nmethod. Finally we suggest an improved version of the bounce-back boundary\ncondition. We test our implementations in both standard benchmark geometries\nand in the pore network of a real sample of a porous rock.", "category": "physics_flu-dyn" }, { "text": "Even-odd alternative dispersions and beyond. Part II. Noninertial and\n inertial particles, and, astrophysical chirality analogy: Particle transports in carriers with even-odd alternating dispersions\n(introduced in Part I) are investigated. For the third-order dispersion as in\nKorteweg-de-Vries (KdV), such alternating dispersion has the effects of not\nonly regularizing the velocity from forming shock singularity (thus the\nattenuation of particle clustering strength) but also symmetrizing the\noscillations (thus the corresponding skewness of the particle densities), among\nothers, as demonstrated numerically. The analogy of such dispersion effects and\nconsequences (on particle transports in particular) with those of helicity in\nBurgers turbulence, addressed in the context of astrophysics and cosmology, is\nmade for illumination and promoting models. Both dispersion and helicity\nregularize the respective systems, and both are shown to be transferred by the\ndrag to the flows of the respective inertial particles carried by the latter\nand to similarly affect the particle clustering. A reward from studying\nparticle transports is the understanding of the (asymptotic) $k^0$-scaling\n(equipartition among the wavenumbers, $k$s), before large-$k$ exponential\ndecay, of the power spectrum of KdV solitons [resulting in the more general\nstatement (valid beyond the KdV soliton and Burgers shock) that \"a\n(one-dimensional) soliton is the derivative of a classical shock, just like the\nDirac delta is the derivative of a step function\"], motivated by the\nexplanation of the the same scaling law of the particle densities as the\napparent approximation of the Dirac deltas; while, the \"shocliton\" from the\neven-odd alternating dispersion in aKdV appears to be, indeed, $shock \\oplus\nsoliton$, accordingly the decomposition of the averaged odd-mode spectrum, from\nsinusoidal initial field, into a $k^{-2}$ part for the shock and a\n$k^0$-scaling part for the solitonic pulses, only the latter being contained in\nthe averaged even-mode spectrum.", "category": "physics_flu-dyn" }, { "text": "Microswimmer trapping in surface waves with shear: Many species of phytoplankton migrate vertically near the surface of the\nocean, either in search of light or nutrients. These motile organisms are\naffected by ocean waves at the surface. We derive a set of wave-averaged\nequations to describe the motion of spheroidal microswimmers. We include\nseveral possible effects, such as gyrotaxis, settling, and wind-driven shear.\nIn addition to the well-known Stokes drift, the microswimmer orbits depend on\ntheir orientation in a way that can lead to trapping at a particular depth;\nthis in turn can affect transport of organisms, and may help explain observed\nphytoplankton layers in the ocean.", "category": "physics_flu-dyn" }, { "text": "Inertial wave super-attractor in a truncated elliptical cone: We consider inertial waves propagating in a fluid contained in a\nnon-axisymmetric three-dimensional rotating cavity. We focus on the particular\ncase of a fluid enclosed inside a truncated cone or frustum, which is the\nvolume that lies between two horizontal parallel planes cutting an upright\ncone. While this geometry has been studied in the past, we generalise it by\nbreaking its axisymmetry and consider the case of a truncated elliptical cone\nfor which the horizontal sections are elliptical instead of circular. The\nproblem is first tackled using ray tracing where local wave packets are\ngeometrically propagated and reflected within the closed volume without\nattenuation. We complement these results with a local asymptotic analysis and\nnumerical simulations of the original linear viscous problem. We show that the\nattractors, well-known in two dimensional or axisymmetric domains, can be\ntrapped in a particular plane in three-dimension provided that the axisymmetry\nof the domain is broken. Contrary to previous examples of attractors in\nthree-dimensional domains, all rays converge towards the same limit cycle\nregardless of initial conditions, and it is localised in the bulk of the fluid.", "category": "physics_flu-dyn" }, { "text": "On the internal gravity waves in the stratified ocean with shear flows: In this paper, we consider a fundamental problem of describing the dynamics\nof internal gravity waves in the stratified ocean with shear flows. We develop\nan asymptotic representation of the wave fields in terms of the Green's\nfunctions. We explore the far field of the internal gravity waves generated by\ndisturbing sources, and propose asymptotic algorithms for calculating its\ndynamics.", "category": "physics_flu-dyn" }, { "text": "Microscopic structure of electrowetting-driven transitions on\n superhydrophobic surfaces: We investigate directly at the microscale the morphology of the\nelectrowetting induced transition between the Cassie-Baxter and Wenzel states\nfor a water droplet on a superhydrophobic surface. Our experiments demonstrate\nthat the transition originates in a very narrow annular region near the\nmacroscopic contact line, which is first invaded by water and causes a thin\nfilm of air to be entrapped below. At high applied voltages, a growing fraction\nof microscopic air-pockets collapse, resulting in a partialWenzel state.\nModulations in the intensity of the light reflected from individual\nmicro-menisci clarify that the local contact angles near the filling transition\nare close to the usual advancing values for contact lines on smooth surfaces.", "category": "physics_flu-dyn" }, { "text": "An implicit wetting and drying approach for non-hydrostatic baroclinic\n flows in high aspect ratio domains: A new approach to modelling free surface flows is developed that enables, for\nthe first time, 3D consistent non-hydrostatic baroclinic physics that wets and\ndries in the large aspect ratio spatial domains that characterise geophysical\nsystems. This is key in the integration of physical models to permit seamless\nsimulation in a single consistent arbitrarily unstructured multiscale and\nmulti-physics dynamical model. A high order continuum representation is\nachieved through a general Galerkin finite element formulation that guarantees\nlocal and global mass conservation, and consistent tracer advection. A flexible\nspatial discretisation permits conforming domain bounds and a variable spatial\nresolution, whilst atypical use of fully implicit time integration ensures\ncomputational efficiency. Notably this brings the natural inclusion of\nnon-hydrostatic baroclinic physics and a consideration of vertical inertia to\nflood modelling in the full 3D domain. This has application in improving\nmodelling of inundation processes in geophysical domains, where dynamics\nproceeds over a large range of horizontal extents relative to vertical\nresolution, such as in the evolution of a tsunami, or in urban environments\ncontaining complex geometric structures at a range of scales.", "category": "physics_flu-dyn" }, { "text": "Doing more with less: the flagellar end piece enhances the propulsive\n effectiveness of human spermatozoa: Spermatozoa self-propel by propagating bending waves along a predominantly\nactive elastic flagellum. The organized structure of the \"9 + 2\" axoneme is\nlost in the most-distal few microns of the flagellum, and therefore this region\nis unlikely to have the ability to generate active bending; as such it has been\nlargely neglected in biophysical studies. Through elastohydrodynamic modeling\nof human-like sperm we show that an inactive distal region confers significant\nadvantages, both in propulsive thrust and swimming efficiency, when compared\nwith a fully active flagellum of the same total length. The beneficial effect\nof the inactive end piece on these statistics can be as small as a few percent\nbut can be above 430%. The optimal inactive length, between 2-18% of the total\nlength, depends on both wavenumber and viscous-elastic ratio, and therefore is\nlikely to vary in different species. Potential implications in evolutionary\nbiology and clinical assessment are discussed.", "category": "physics_flu-dyn" }, { "text": "Spatial and temporal evolution of an experimental debris flow,\n exhibiting coupled fluid and particulate phases: The internal behaviour of debris flows provides fundamental insight into the\nmechanics responsible for their motion. We provide velocity data within a\nsmall-scale experimental debris flow, consisting of the instantaneous release\nof a water-granular mixture along a rectangular flume, inclined at 31 degrees.\nThe results show a transition from a collisional, turbulent front to a\nviscous-type, steady flow body, exhibiting strong fluid-particulate coupling.\nThis is the first time that both the spatial and temporal evolution of the\ninternal mechanics of a small-scale debris flow have been considered. Our\nresults serve as invaluable data for testing two-phase fluid-particulate\nnumerical models.", "category": "physics_flu-dyn" }, { "text": "Transitions in a magnetized quasi-laminar spherical Couette Flow: First results of a new spherical Couette experiment are presented. The liquid\nmetal flow in a spherical shell is exposed to a homogeneous axial magnetic\nfield. For a Reynolds number Re=1000, we study the effect of increasing\nHartmann number Ha. The resulting flow structures are inspected by ultrasound\nDoppler velocimetry. With a weak applied magnetic field, we observe an\nequatorially anti-symmetric jet instability with azimuthal wave number m=3. As\nthe magnetic field strength increases, this instability vanishes. When the\nfield is increased further, an equatorially symmetric return flow instability\narises. Our observations are shown to be in good agreement with linear\nstability analysis and non-linear flow simulations.", "category": "physics_flu-dyn" }, { "text": "Small deformation theory for a magnetic droplet in a rotating field: A three dimensional small deformation theory is developed to examine the\nmotion of a magnetic droplet in a uniform rotating magnetic field. The\nequations describing the droplet's shape evolution are derived using two\ndifferent approaches - a phenomenological equation for the tensor describing\nthe anisotropy of the droplet, and the hydrodynamic solution using perturbation\ntheory. We get a system of ordinary differential equations for the parameters\ndescribing the droplet's shape, which we further analyze for the particular\ncase when the droplet's elongation is in the plane of the rotating field. The\nqualitative behavior of this system is governed by a single dimensionless\nquantity $\\tau\\omega$ - the product of the characteristic relaxation time of\nsmall perturbations and the angular frequency of the rotating magnetic field.\nValues of $\\tau\\omega$ determine whether the droplet's equilibrium will be\ncloser to an oblate or a prolate shape, as well as whether it's shape will\nundergo oscillations as it settles to this equilibrium.We show that for small\ndeformations, the droplet pseudo-rotates in the rotating magnetic field - its\nlong axis follows the field, which is reminiscent of a rotation, nevertheless\nthe torque exerted on the surrounding fluid is zero. We compare the analytic\nresults with a boundary element simulation to determine their accuracy and the\nlimits of the small deformation theory.", "category": "physics_flu-dyn" }, { "text": "A finite element formulation of the outlet gradient boundary condition\n for convective-diffusive transport problems: A simple finite element formulation of the outlet gradient boundary condition\nis presented in the general context of convective-diffusive transport\nprocesses. Basically, the method is based on an upstream evaluation of the\ndependent variable gradient along open boundaries. Boundary normal unit vectors\nand gradient operators are evaluated using covariant bases and metric tensors,\nwhich allow handling finite elements of mixed dimensions. Even though the\npresented method has implications for many fields where diffusion processes are\ninvolved, discussion and illustrative examples address more particularly the\nframework of contaminant transport in porous media, in which the outlet\ngradient concentration is classically, but wrongly assumed to be zero.", "category": "physics_flu-dyn" }, { "text": "Diluted-dispersed mass transfer within an AWE: The goal of this document is describe the multiphase transfer processes\ndescribing the bubble dynamics of a water electrolyzer. The motivation is to\ndescribe the dilute-dispersed mass transfer within and Alkaline Water\nElectrolyzer. Special emphasis is put on the mathematical formulation. The\npresentation starts by posing the governing equations and their dimensionless\ncounterpart. By filtering the equations, the two-fluid model is presented along\nwith the need to sub-scale and wall models. To the later aim, boundary layer\nequations are introduced. By reviewing self-similiarity transformations, the\nanalysis of Blasius, Ostrach and Sparrow is reviewed for Prandtl's boundary\nlayer equations; along with that of Leveque.", "category": "physics_flu-dyn" }, { "text": "Contribution to the theory of waves in multi-dimensional linear\n dispersive media: The asymptotic solutions for linear waves generated by oscillating source of\nelliptic shape in the motionless media is constructed with the recently\ndeveloped Reference Solution Approach (RSA). Pronounced anisotropy of the\nsolutions is found for elongated sources both for amplitudes and phases of the\nresulting wave pattern. The classic Kelvin angles of the ship wave patterns\ndetermine specific directions of this anisotropy, thus, demonstrating the role\nof wave dispersion. The analytical results within the RSA are shown to agree\nremarkably well with exact solutions of the linear wave problem.", "category": "physics_flu-dyn" }, { "text": "Dynamic wetting effects in finite mobility ratio Hele-Shaw flow: In this paper we study the effects of dynamic wetting on the immiscible\ndisplacement of a high viscosity fluid subject to the radial injection of a\nless viscous fluid in a Hele-Shaw cell. The displaced fluid can leave behind a\ntrailing film that coats the cell walls, dynamically affecting the pressure\ndrop at the fluid interface. By considering the non-linear pressure drop in a\nboundary element formulation, we construct a Picard scheme to iteratively\npredict the interfacial velocity and subsequent displacement in finite mobility\nratio flow regimes. Dynamic wetting delays the onset of finger bifurcation in\nthe late stages of interfacial growth, and at high local capillary numbers can\nalter the fundamental mode of bifurcation, producing vastly different finger\nmorphologies. In low mobility ratio regimes, we see that finger interaction is\nreduced and characteristic finger breaking mechanisms are delayed but never\nfully inhibited. In high mobility ratio regimes, finger shielding is reduced\nwhen dynamic wetting is present. Finger bifurcation is delayed which allows the\nprimary fingers to advance further into the domain before secondary fingers are\ngenerated, reducing the level of competition.", "category": "physics_flu-dyn" }, { "text": "Learned Coarse Models for Efficient Turbulence Simulation: Turbulence simulation with classical numerical solvers requires\nhigh-resolution grids to accurately resolve dynamics. Here we train learned\nsimulators at low spatial and temporal resolutions to capture turbulent\ndynamics generated at high resolution. We show that our proposed model can\nsimulate turbulent dynamics more accurately than classical numerical solvers at\nthe comparably low resolutions across various scientifically relevant metrics.\nOur model is trained end-to-end from data and is capable of learning a range of\nchallenging chaotic and turbulent dynamics at low resolution, including\ntrajectories generated by the state-of-the-art Athena++ engine. We show that\nour simpler, general-purpose architecture outperforms various more specialized,\nturbulence-specific architectures from the learned turbulence simulation\nliterature. In general, we see that learned simulators yield unstable\ntrajectories; however, we show that tuning training noise and temporal\ndownsampling solves this problem. We also find that while generalization beyond\nthe training distribution is a challenge for learned models, training noise,\nadded loss constraints, and dataset augmentation can help. Broadly, we conclude\nthat our learned simulator outperforms traditional solvers run on coarser\ngrids, and emphasize that simple design choices can offer stability and robust\ngeneralization.", "category": "physics_flu-dyn" }, { "text": "Characterization of a Canonical Helicopter Hub Wake: The current study investigates the long-age wake behind rotating helicopter\nhub models composed of geometrically simple, canonical bluff body shapes. The\nmodels consisted of a 4-arm rotor mounted on a shaft above a 2-arm (scissor)\nrotor with all the rotor arms having a rectangular cross section. The relative\nphase between the 2- and 4-arm rotors was either $0\\deg$ (in-phase) or $45\\deg$\n(out-of-phase). The rotors were oriented at zero angle-of-attack and rotated at\n30 Hz. Their wakes were measured with particle-image-velocimetry within a water\ntunnel at a hub diameter based Reynolds number of $820,000$ and an advance\nratio of $0.2$. Mean profiles, fluctuating profiles and spectral analysis using\ntime-series analysis as well as dynamic mode decomposition were used to\ncharacterize the wake and identify coherent structures associated with specific\nfrequency content. The canonical geometry produced coherent structures that\nwere consistent with previous results using more complex geometries. It was\nshown that the dominant structures (2 and 4 times per hub revolution) decay\nslowly and were not sensitive to the relative phase between the rotors.\nConversely, the next strongest structure (6 times per hub revolution) was\nsensitive to the relative phase with almost no coherence observed for the\nin-phase model. This is strong evidence that the 6 per revolution content is a\nnonlinear interaction between the 2 and 4 revolution structures. This study\ndemonstrates that the far wake region is dominated by the main rotor arms wake,\nthe scissor rotor wake and interactions between these two features.", "category": "physics_flu-dyn" }, { "text": "Large-scale motions and self-similar structures in compressible\n turbulent channel flows: In this work, we study the scale characteristics of the log- and outer-region\nmotions and structures in subsonic and supersonic turbulence. To this end, a\nseries of direct numerical simulations of the compressible turbulent channel\nflow at medium Reynolds numbers are performed. Based on this database, the\nstreamwise and spanwise length scales of the outer-region motions are\ninvestigated by the two-point correlations and the one-dimensional spectra. The\nenergy distribution among the multi-scale structures in the outer region is\nfound to be dominated by the semilocal friction-Reynolds-number effects rather\nthan the Mach-number effects. This conclusion not only holds for the velocity\nfluctuations but also the fluctuations of the thermodynamic variables. Besides,\nthe streamwise and spanwise length scales of the outer motions do not alter\nsignificantly when the flow passes the sound barrier as reported by a previous\nexperimental study (Bross et al., J. Fluid Mech., vol. 911, 2021, A2). On the\nother hand, the self-similar structures populating the logarithmic region are\ninvestigated by adopting a linear coherence spectrum. The\nstreamwise/wall-normal aspect ratio of the self-similar wall-attached\nstructures of the streamwise velocity and temperature fluctuations is\napproximately 15.5, and the counterpart of density and pressure fluctuations is\n1.8. The present study confirms the existence of self-similar structures in\ncompressible wall turbulence and assesses their geometrical characteristics.", "category": "physics_flu-dyn" }, { "text": "Near-wall turbulence alteration with the Transpiration-Resistance Model: A set of boundary conditions called the Transpiration-Resistance Model (TRM)\nare investigated in altering near-wall turbulence. The TRM has been previously\nproposed by \\citet{Lacis2020} as a means of representing the net effect of\nsurface micro-textures on their overlying bulk flows. It encompasses\nconventional Navier-slip boundary conditions relating the streamwise and\nspanwise velocities to their respective shears through the slip lengths\n$\\ell_x$ and $\\ell_z$. In addition, it features a transpiration condition\naccounting for the changes induced in the wall-normal velocity by expressing it\nin terms of variations of the wall-parallel velocity shears through the\ntranspiration lengths $m_x$ and $m_z$. Greater levels of drag increase occur\nwhen more transpiration takes place at the boundary plane, with turbulent\ntranspiration being predominately coupled to the spanwise shear component for\ncanonical near-wall turbulence. The TRM can reproduce the effect of a\nhomogeneous and structured roughness up to ${k^+}\\,{\\approx18}$. In this\ntransitionally rough flow regime, the transpiration lengths of the TRM must be\nempirically determined. The \\emph{transpiration factor} is defined as the\nproduct between the slip and transpiration lengths, i.e. $(m\\ell)_{x,z}$. This\nfactor contains the compound effect of the wall-parallel velocity occurring at\nthe boundary plane and increased permeability, both of which lead to the\ntransport of momentum in the wall-normal direction. A linear relation between\nthe transpiration factor and the roughness function is observed for regularly\ntextured surfaces in the transitionally rough regime of turbulence. The\nrelations obtained between the transpiration factor and the roughness function\nshow that such effective flow quantities can be suitable measures for\ncharacterizing rough surfaces in this flow regime.", "category": "physics_flu-dyn" }, { "text": "Droplet size distribution in homogeneous isotropic turbulence: We study the physics of droplet breakup in a statistically stationary\nhomogeneous and isotropic turbulent flow by means of high resolution numerical\ninvestigations based on the multicomponent lattice Boltzmann method. We\nverified the validity of the criterion proposed by Hinze (1955) for droplet\nbreakup and we measured the full probability distribution function (pdf) of\ndroplets radii at different Reynolds numbers and for different volume fraction.\nBy means of a Lagrangian tracking we could follow individual droplets along\ntheir trajectories, define a local Weber number based on the velocity gradients\nand study its cross-correlation with droplet deformation.", "category": "physics_flu-dyn" }, { "text": "Exact mathematical formulas for wall-heat flux in compressible turbulent\n channel flows: In this paper, several exact expressions for the mean heat flux at the wall\n($q_w$) for the compressible turbulent channel flows are derived by using the\ninternal energy equation or the total energy equation. Two different routes,\nincluding the FIK method and the RD method, can be applied. The direct\nnumerical simulations data of compressible channel flows at different Reynolds\nand Mach numbers verify the correctness of the derived formulas. Discussions\nrelated to the different energy equations, and different routes are carried\nout, and we may arrive at the conclusion that most of the formulas derived in\nthe present work are just mathematically ones which generally are lack of clear\nphysical interpretation. They can be used to estimate $q_w$, but might not be\nsuitable for explore the underlying physics.", "category": "physics_flu-dyn" }, { "text": "Charge Transport Scalings in Turbulent Electroconvection: We describe a local-power law scaling theory for the mean dimensionless\nelectric current $Nu$ in turbulent electroconvection. The experimental system\nconsists of a weakly conducting, submicron thick liquid crystal film supported\nin the annulus between concentric circular electrodes. It is driven into\nelectroconvection by an applied voltage between its inner and outer edges. At\nsufficiently large voltage differences, the flow is unsteady and electric\ncharge is turbulently transported between the electrodes. Our theoretical\ndevelopment, which closely parallels the Grossmann-Lohse model for turbulent\nthermal convection, predicts the local-power law $Nu \\sim F(\\Gamma) {\\cal\nR}^{\\gamma} {\\cal P}^{\\delta}$. ${\\cal R}$ and ${\\cal P}$ are dimensionless\nnumbers that are similar to the Rayleigh and Prandtl numbers of thermal\nconvection, respectively. The dimensionless function $F(\\Gamma)$, which is\nspecified by the model, describes the dependence of $Nu$ on the aspect ratio\n$\\Gamma$. We find that measurements of $Nu$ are consistent with the theoretical\nmodel.", "category": "physics_flu-dyn" }, { "text": "Method for Flow Measurement in Microfluidic Channels Based on Electrical\n Impedance Spectroscopy: We have developed and characterized two novel micro flow sensors based on\nmeasuring the electrical impedance of the interface between the flowing liquid\nand metallic electrodes embedded on the channel walls. These flow sensors are\nvery simple to fabricate and use, are extremely compact and can easily be\nintegrated into most microfluidic systems. One of these devices is a micropore\nwith two tantalum/platinum electrodes on its edges; the other is a micro\nchannel with two tantalum /platinum electrodes placed perpendicular to the\nchannel on its walls. In both sensors the flow rate is measured via the\nelectrical impedance between the two metallic electrodes, which is the\nimpedance of two metal-liquid junctions in series. The dependency of the\nmetal-liquid junction impedance on the flow rate of the liquid has been\nstudied. The effects of different parameters on the sensor's outputs and its\nnoise behavior are investigated. Design guidelines are extracted and applied to\nachieve highly sensitive micro flow sensors with low noise.", "category": "physics_flu-dyn" }, { "text": "Statistics of phase fluctuations of an acoustic wave propagating through\n a turbulent flow: We investigate the statistics of phase fluctuations of an acoustic wave\npropagating through a turbulent flow in line of sight (LOS) configuration.\nExperiments are performed on a closed von Karman swirling flow whose boundaries\nare maintained at a constant temperature. In particular, we analyze the root\nmean square (RMS) and the power spectrum density (PSD) of phase fluctuations. A\nmodel is developed and analytical predictions obtained for these quantities\nusing geometrical acoustics are shown to be in agreement with experimental\nobservations.", "category": "physics_flu-dyn" }, { "text": "Emergence of intense jets and Jupiter Great Red Spot as maximum entropy\n structures: We explain the emergence and robustness of intense jets in highly turbulent\nplanetary atmospheres, like on Jupiter, by a general approach of statistical\nmechanics of potential vorticity patches. The idea is that potential vorticity\nmixing leads to the formation of a steady organized coarse grained flow,\ncorresponding to the statistical equilibrium state. Our starting point is the\nquasi-geostrophic 1-1/2 layer model, and we consider the relevant limit of a\nsmall Rossby radius of deformation. Then narrow jets are obtained, scaling like\nthe Rossby radius of deformation. These jets can be either zonal, or closed\ninto a ring bounding a vortex. Taking into account the effect of the beta\neffect and a sublayer deep shear flow, we predict an organization of the\nturbulent atmospheric layer into an oval-shaped vortex amidst a background\nshear. Such an isolated vortex is centered over an extremum of the equivalent\ntopography (determined by the deep shear flow and beta-effect). This prediction\nis in agreement with analysis of wind data in major Jovian vortices (Great Red\nSpot and Oval BC).", "category": "physics_flu-dyn" }, { "text": "Accurate Multi-physics Numerical Analysis of Particle Preconcentration\n Based on Ion Concentration Polarization: This paper studies mechanism of preconcentration of charged particles in a\nstraight micro-channel embedded with permselective membranes, by numerically\nsolving coupled transport equations of ions, charged particles and solvent\nfluid without any simplifying assumptions. It is demonstrated that trapping and\npreconcentration of charged particles are determined by the interplay between\ndrag force from the electroosmotic fluid flow and the electrophoretic force\napplied trough the electric field. Several insightful characteristics are\nrevealed, including the diverse dynamics of co-ions and counter ions,\nreplacement of co-ions by focused particles, lowered ion concentrations in\nparticle enriched zone, and enhanced electroosmotic pumping effect etc.\nConditions for particles that may be concentrated are identified in terms of\ncharges, sizes and electrophoretic mobilities of particles and co-ions.\nDependences of enrichment factor on cross-membrane voltage, initial particle\nconcentration and buffer ion concentrations are analyzed and the underlying\nreasons are elaborated. Finally, post priori a condition for validity of\ndecoupled simulation model is given based on charges carried by focused charge\nparticles and that by buffer co-ions. These results provide important guidance\nin the design and optimization of nanofluidic preconcentration and other\nrelated devices.", "category": "physics_flu-dyn" }, { "text": "Optimal Strokes of Low Reynolds Number Linked-Sphere Swimmers: Optimal gait design is important for micro-organisms and micro-robots that\npropel themselves in a fluid environment in the absence of external force or\ntorque. The simplest models of shape changes are those that comprise a series\nof linked-spheres that can change their separation and their sizes. We examine\nthe dynamics of three existing linked-sphere types of modeling swimmers in low\nReynolds number Newtonian fluids using asymptotic analysis, and obtain their\noptimal swimming strokes by solving the Euler-Lagrange equation using the\nshooting method. The numerical results reveal that (1) with the minimal 2\ndegrees of freedom in shape deformations, the model swimmer adopting the mixed\nshape deformation modes strategy is more efficient than those with a\nsingle-mode of shape deformation modes, and (2) the swimming efficiency mostly\ndecreases as the number of spheres increases, indicating that more degrees of\nfreedom in shape deformations might not be a good strategy in optimal gait\ndesign in low Reynolds number locomotion.", "category": "physics_flu-dyn" }, { "text": "Deceleration Driven Wetting Transition During \"Gentle\" Drop Deposition: We present high speed video of Cassie-Baxter to Wenzel drop transition during\ngentle deposition of droplets where the modest amount of energy is channeled\nvia rapid deceleration into a high water hammer pressure.", "category": "physics_flu-dyn" }, { "text": "Stability and sensitivity analysis of bird flapping flight: This paper investigates stability analysis of flapping flight. Due to\ntime-varying aerodynamic forces, such systems do not display fixed points of\nequilibrium. The problem is therefore approached via a limit cycle analysis\nbased on Floquet theory. Stability is assessed from the eigenvalues of the\nJacobian matrix associated to the limit cycle, also known as the Floquet\nmultipliers. We developed this framework to analyze the flapping flight\nequations of motion of a bird in the longitudinal plane. Such a system is known\nto be not only non-linear and time-dependent, but also driven by\nstate-dependent forcing aerodynamic forces. A model accounting for wing\nmorphing under prescribed kinematics is developed for generating realistic\nstate-dependent aerodynamic forces. The morphing wing geometry results from the\nenvelope of continuously articulated rigid bodies, modeling bones and feather\nrachises, and capturing biologically relevant degrees of freedom. A sensitivity\nanalysis is carried out which allows studying several flight configurations in\ntrimmed state. Our numerical results show that in such a system one instability\nmode is ubiquitous, thus suggesting the importance of sensory feedback to\nachieve steady-state flapping flight in birds. The effect of wingbeat\namplitude, governed by the shoulder joint, is found to be crucial in tuning the\ngait towards level flight, but marginally affects stability. In contrast, the\nrelative position between the wing and the center of mass is found to\nsignificantly affect the values of Floquet multipliers, suggesting that the\ndistribution of pitching moment plays a very important role in flapping flight\nstability.", "category": "physics_flu-dyn" }, { "text": "Coagulation of inertial particles in supersonic turbulence: Coagulation driven by supersonic turbulence is primarily an astrophysical\nproblem because coagulation processes on Earth are normally associated with\nincompressible fluid flows at low Mach numbers, while dust aggregation in the\ninterstellar medium (ISM) for instance is an example of the opposite regime. We\nstudy coagulation of inertial particles in compressible turbulence using\nhigh-resolution direct and shock-capturing numerical simulations with a wide\nrange of Mach numbers from nearly incompressible to moderately supersonic. The\nparticle dynamics is simulated by representative particles and the effects on\nthe size distribution and coagulation rate due to increasing Mach number is\nexplored. We show that the time evolution of particle size distribution mainly\ndepends on the compressibility (Mach number). We find that the average\ncoagulation kernel $\\langle C_{ij}\\rangle$ scales linearly with the average\nMach number $\\mathcal{M}_{\\rm rms}$ multiplied by the combined size of the\ncolliding particles, that is, $\\langle C_{ij}\\rangle \\sim \\langle (a_i +\na_j)^3\\rangle\\, \\mathcal{M}_{\\rm rms}\\tau_\\eta^{-1}$, which is qualitatively\nconsistent with expectations from analytical estimates. A quantitative\ncorrection $\\langle C_{ij}\\rangle \\sim \\langle(a_i + a_j)^3\\rangle(v_{\\rm\np,rms}/c_{\\rm s})\\tau_\\eta^{-1}$ is proposed and can serve as a benchmark for\nfuture studies. We argue that the coagulation rate $\\langle R_c\\rangle$ is also\nenhanced by compressibility-induced compaction of particles.", "category": "physics_flu-dyn" }, { "text": "Thin film flow and heat transfer over an unsteady stretching sheet with\n thermal radiation, internal heating in presence of external magnetic field: In this paper we present a mathematical analysis of thin film flow and heat\ntransfer to a laminar liquid film from a horizontal stretching sheet. The flow\nof thin liquid film and subsequent heat transfer from the stretching surface is\ninvestigated with the aid of similarity transformations. Similarity\ntransformations are used to convert unsteady boundary layer equations to a\nsystem of non-linear ordinary differential equations. The resulting non-linear\ndifferential equations are solved numerically using Runge-kutta-Fehlberg and\nNewton-Raphson schemes. A relationship between film thickness $\\beta$ and the\nunsteadiness parameter $S$ is found, the effect of unsteadiness parameter $S$,\nand the Prandtl number $Pr$, Magnetic field parameter $Mn$, Radiation parameter\n$Nr$ and viscous dissipation parameter $Ec$ and heat source parameter $\\gamma$\non the temperature distributions are presented and discussed in detail. Present\nanalysis shows that the combined effect of magnetic field, thermal radiation,\nheat source and viscous dissipation. The results which form special case of the\npresent study are in excellent agreement with the results reported in the\nliterature.", "category": "physics_flu-dyn" }, { "text": "Spatially resolved measurement of the electrostatic charge of turbulent\n powder flows: This article reports on measurements of the electrostatic charge of particles\nin a turbulent duct flow. In contrast to previous charge measurements, which do\nnot apply to turbulent flows or give only the sum of all particles' charges,\nthe new method resolves the charge of a turbulent powder flow spatially. The\nexperiment consists of a Particle Tracking Velocimetry (PTV) system and\nelectrode plates that generate an electric field. By comparing particle\nvelocities and accelerations with and without the electric field, the\ntime-averaged local particle charge profile is derived. Spatially resolving the\ncharge profiles unveiled bipolar particle flow. The average of the charge\nprofiles agreed well with a conventional Faraday pail measurement,\ndemonstrating the accuracy of our measurements. However, the peak value of the\ncharge profiles was 76 times higher than the average of the particles' charge.", "category": "physics_flu-dyn" }, { "text": "State-of-the-art SPH solver DualSPHysics: from fluid dynamics to\n multiphysics problems: DualSPHysics is a weakly compressible smoothed particle hydrodynamics (SPH)\nNavier-Stokes solver initially conceived to deal with coastal engineering\nproblems, especially those related to wave impact with coastal structures.\nSince the first release back in 2011, DualSPHysics has shown to be robust and\naccurate for simulating extreme wave events along with a continuous improvement\nin efficiency thanks to the exploitation of hardware such as graphics\nprocessing units (GPUs) for scientific computing or the coupling with wave\npropagating models such as SWASH and OceanWave3D. Numerous additional\nfunctionalities have also been included in the DualSPHysics package over the\nlast few years which allow the simulation of fluid-driven objects. The use of\nthe discrete element method (DEM) has allowed the solver to simulate the\ninteraction among different bodies (sliding rocks, for example), which provides\na unique tool to analyse debris flows. In addition, the recent coupling with\nother solvers like Project Chrono or MoorDyn has been a milestone in the\ndevelopment of the solver. Project Chrono allows the simulation of articulated\nstructures with joints, hinges, sliders and springs and MoorDyn allows\nsimulating moored structures. Both functionalities make DualSPHysics one of the\nmeshless model world leaders in the simulation of offshore energy harvesting\ndevices. Lately, the present state of maturity of the solver goes beyond single\nphase simulations, allowing multi-phase simulations with gas-liquid and a\ncombination of Newtonian and non-Newtonian models expanding further the\ncapabilities and range of applications for the DualSPHysics solver. These\nadvances and functionalities make DualSPHysics a state-of-the-art meshless\nsolver with emphasis on free-surface flow modelling.", "category": "physics_flu-dyn" }, { "text": "Dynamic modelling of near-surface turbulence in large eddy simulation of\n wind farms: In large eddy simulation of atmospheric boundary layer flows over wind farms,\nwall-layer models are generally imposed for the surface fluxes without\nconsidering the spatial variability of the surface roughness. In this study, we\nconsider the near-surface model in conjunction with square of the velocity\ngradient tensor to model the adaptive dissipation of turbulence production. The\nsurface roughness is incorporated through Monin-Obhukhov similarity theory for\nthe computational cells immediately adjacent to the Earth's surface. The\nunderlying proposed near-surface model captures the significant amount of\nReynolds stresses in the near-surface and is able to maintain the log-law\nprofile in wind farms. The present study indicates that the suggested\n`near-surface model' is relatively robust in comparison to the classical\n`near-wall model'.", "category": "physics_flu-dyn" }, { "text": "Low-wavenumber forcing and turbulent energy dissipation: A popular method of forcing the fluid in Direct Numerical Simulations of\nturbulence is to take the body force proportional to the projection of the\nvelocity of the fluid onto its lowest Fourier modes, while keeping the injected\nexternal power constant. In this paper we perform a simple but rigorous\nanalysis to establish bounds on the relationship between the energy dissipation\nrate and the resulting Reynolds number for this type of forcing. While this\nanalysis cannot give detailed information of the energy spectrum, it does\nprovide some indication of the balance of energy between the lower, directly\nforced, modes, and those excited by the cascade.", "category": "physics_flu-dyn" }, { "text": "Numerical study of Bingham flow in macrosopic two dimensional\n heterogenous porous media: The flow of non-Newtonian fluids is ubiquitous in many applications in the\ngeological and industrial context. We focus here on yield stress fluids (YSF),\ni.e. a material that requires minimal stress to flow. We study numerically the\nflow of yield stress fluids in 2D porous media on a macroscopic scale in the\npresence of local heterogeneities. As with the microscopic problem,\nheterogeneities are of crucial importance because some regions will flow more\neasily than others. As a result, the flow is characterized by preferential flow\npaths with fractal features. These fractal properties are characterized by\ndifferent scale exponents that will be determined and analyzed. One of the\nsalient features of these results is that these exponents seem to be\nindependent of the amplitude of heterogeneities for a log-normal distribution.\nIn addition, these exponents appear to differ from those at the microscopic\nlevel, illustrating the fact that, although similar, the two scales are\ngoverned by different sets of equations.", "category": "physics_flu-dyn" }, { "text": "The effect of polydispersity in a turbulent channel flow laden with\n finite-size particles: We study turbulent channel flows of monodisperse and polydisperse suspensions\nof finite-size spheres by means of Direct Numerical Simulations using an\nimmersed boundary method to account for the dispersed phase. Suspensions with 3\ndifferent Gaussian distributions of particle radii are considered (i.e. 3\ndifferent standard deviations). The distributions are centered on the reference\nparticle radius of the monodisperse suspension. In the most extreme case, the\nradius of the largest particles is 4 times that of the smaller particles. We\nconsider two different solid volume fractions, 2% and 10%. We find that for all\npolydisperse cases, both fluid and particles statistics are not substantially\naltered with respect to those of the monodisperse case. Mean streamwise fluid\nand particle velocity profiles are almost perfectly overlapping. Slightly\nlarger differences are found for particle velocity fluctuations. These increase\nclose to the wall and decrease towards the centerline as the standard deviation\nof the distribution is increased. Hence, the behavior of the suspension is\nmostly governed by excluded volume effects regardless of particle size\ndistribution (at least for the radii here studied). Due to turbulent mixing,\nparticles are uniformly distributed across the channel. However, smaller\nparticles can penetrate more into the viscous and buffer layer and velocity\nfluctuations are therein altered. Non trivial results are presented for\nparticle-pair statistics.", "category": "physics_flu-dyn" }, { "text": "Injection rate of cylinder lubrication oil in large two-stroke marine\n diesel engines using a common rail lubrication system: This paper investigates a common rail cylinder lubrication system for large\ntwo-stroke marine diesel engines using electronically controlled injectors. The\nsystem is studied using the Bosch rate of injection measurement technique The\ncommon rail injector has a buildup of mass flow of approximately 1 ms as the\ninjector opens until the nozzle is choked from cavitation. Using a highly\nviscous fluid, the Bosch rate of injection method is able to predict the\ninjected amount with an error of 5% or lower for nearly the entire tested\ndelivery range of 2 mg to 21 mg. Lubrication of cylinder liners and piston\nrings is a crucial parameter in operating a two-stoke marine diesel engine\nefficiently. Both over and under lubrication is harmful for the engine, so the\nability to accurately dose the cylinder oil is very important. A mass flow\nbuild up time of 1 ms promises high accuracy of dosage even down to 2.5 mg per\ninjection. This paves the way for injecting the oil where and when it is\nneeded, which in turn will improve engine performance and lower harmful\nemissions.", "category": "physics_flu-dyn" }, { "text": "Enhanced Water Nucleation and Growth Based on Microdroplet Mobility on\n Lubricant-Infused Surfaces: Lubricant-infused surfaces (LISs) can promote stable dropwise condensation\nand improve heat transfer rates due to a low nucleation free-energy barrier and\nhigh droplet mobility. Topographical differences in the oil surface cause water\nmicrodroplets to rigorously self-propel long distances, continuously\nredistributing the oil film and potentially refreshing the surface for\nre-nucleation. Using high-speed microscopy, we reveal that during water\ncondensation on LISs, the smallest visible droplets (diameter ~ 1um,\nqualitatively representing nucleation) predominantly emerge in oil-poor regions\ndue to a smaller thermal activation barrier. Considering the significant heat\ntransfer performance of microdroplets (< 10um) and transient characteristic of\nmicrodroplet movement, we compare the apparent nucleation rate density and\nwater collection rate for LISs with oils of different viscosity and a solid\nhydrophobic surface at a wide range of subcooling temperatures. Generally, the\nlowest lubricant viscosity leads to the highest nucleation rate density. We\ncharacterize the length and frequency of microdroplet movement and attribute\nthe nucleation enhancement primarily to higher droplet mobility and surface\nrefreshing frequency. Interestingly and unexpectedly, hydrophobic surfaces\noutperform high-viscosity LISs at high subcooling temperatures, but are\ngenerally inferior to any of the tested LISs at low temperature differences. To\nexplain the observed non-linearity between LISs and the solid hydrophobic\nsurface, we introduce two dominant regimes that influence the condensation\nefficiency: mobility-limited and coalescence-limited. Our findings advance the\nunderstanding of dynamic water-lubricant interactions and provide new design\nrationales for choosing surfaces for enhanced dropwise condensation and water\ncollection efficiencies.", "category": "physics_flu-dyn" }, { "text": "Molecular dynamics simulation of flow around a circular nano-cylinder: In this study, the wake flow around a circular nano-cylinder is numerically\ninvestigated with molecular dynamics simulation to reveal the micro/nano size\neffect on the wake flow. The cavitation occurring when Reynolds number (Re) >\n101 can effectively influence the wake flow. The Strouhal number (St) of the\nwake flow increases with the Re at low Re, but steadily decreases with the Re\nafter the cavitation appears. The dominant frequency of the lift force\nfluctuation can be higher than that of the velocity fluctuation, and be drowned\nin the chaotic fluctuating background of the Brownian forces when Re {\\geq}\n127. Also because of the strong influence of the Brownian forces, no dominant\nfrequency of the drag force fluctuation can be observed. The Jz number, which\nis defined as the ratio between the mean free path {\\lambda} of the fluid\nmolecules and the equilibrium distance of potential energy {\\sigma}, is newly\nintroduced in order to consider the internal size effect of fluid. The St of\nthe wake flow increases with the Jz until it falls to zero sharply when Jz\n{\\approx} 1.7. It denotes the discontinuity of the fluid can eventually\neliminate the vortex generation and shedding. Meanwhile, the St decreases with\nthe Kn because of the intensification of the cavitation.", "category": "physics_flu-dyn" }, { "text": "Stirring up trouble: Multi-scale mixing measures for steady scalar\n sources: The mixing efficiency of a flow advecting a passive scalar sustained by\nsteady sources and sinks is naturally defined in terms of the suppression of\nbulk scalar variance in the presence of stirring, relative to the variance in\nthe absence of stirring. These variances can be weighted at various spatial\nscales, leading to a family of multi-scale mixing measures and efficiencies. We\nderive a priori estimates on these efficiencies from the advection--diffusion\npartial differential equation, focusing on a broad class of statistically\nhomogeneous and isotropic incompressible flows. The analysis produces bounds on\nthe mixing efficiencies in terms of the Peclet number, a measure the strength\nof the stirring relative to molecular diffusion. We show by example that the\nestimates are sharp for particular source, sink and flow combinations. In\ngeneral the high-Peclet number behavior of the bounds (scaling exponents as\nwell as prefactors) depends on the structure and smoothness properties of, and\nlength scales in, the scalar source and sink distribution. The fundamental\nmodel of the stirring of a monochromatic source/sink combination by the random\nsine flow is investigated in detail via direct numerical simulation and\nanalysis. The large-scale mixing efficiency follows the upper bound scaling\n(within a logarithm) at high Peclet number but the intermediate and small-scale\nefficiencies are qualitatively less than optimal. The Peclet number scaling\nexponents of the efficiencies observed in the simulations are deduced\ntheoretically from the asymptotic solution of an internal layer problem arising\nin a quasi-static model.", "category": "physics_flu-dyn" }, { "text": "The effect of spanwise heterogeneous surfaces on mixed convection in\n turbulent channels: Turbulent mixed convection in channel flows with heterogeneous surfaces is\nstudied using direct numerical simulations. The relative importance between\nbuoyancy and shear effects, characterized by the bulk Richardson number $Ri_b$,\nis varied in order to cover the flow regimes of forced, mixed and natural\nconvection, which are associated with different large-scale flow organization.\nThe heterogeneous surface consists of streamwise-aligned ridges, which are\nknown to induce secondary motion in case of forced convection. The large-scale\nstreamwise rolls emerging under smooth-wall mixed convection conditions are\nsignificantly affected by the heterogeneous surfaces and their appearance is\nconsiderably reduced for dense ridge spacings. It is found that the formation\nof these rolls requires larger buoyancy forces than over smooth walls due to\nthe additional drag induced by the ridges. Therefore, the transition from\nforced convection structures to rolls is delayed towards larger $Ri_b$ for\nspanwise heterogeneous surfaces. The influence of the heterogeneous surface on\nthe flow organization of mixed convection is particularly pronounced in the\nroll-to-cell transition range, where ridges favor the transition to convective\ncells at significantly lower $Ri_b$. In addition, the convective cells are\nobserved to align perpendicular to the ridges with decreasing ridge spacing. We\nattribute this reorganization to the fact that flow parallel to the ridges\nexperience less drag than flow across the ridges, which is energetically more\nbeneficial. Furthermore, we find that streamwise rolls exhibit very slow\ndynamics for $Ri_b=1$ and $Ri_b=3.2$ when the ridge spacing is in the order of\nthe rolls' width. For these cases the up- and downdrafts of the rolls move\nslowly across the entire channel instead of being fixed in space as observed\nfor the smooth-wall cases.", "category": "physics_flu-dyn" }, { "text": "Absolute instability in shock-containing jets: We present an analysis of the linear stability characteristics of\nshock-containing jets. The flow is linearised around a spatially periodic mean,\nwhich acts as a surrogate for a mean flow with a shock-cell structure, leading\nto a set of partial differential equations with periodic coefficients in space.\nDisturbances are written using the Floquet ansatz and Fourier modes in the\nstreamwise direction, leading to an eigenvalue problem for the Floquet\nexponent. The characteristics of the solution are directly compared to the\nlocally parallel case, and some of the features are similar. The inclusion of\nperiodicity induces minor changes in the growth rate and phase velocity of the\nrelevant modes for small shock amplitudes. On the other hand, the\neigenfunctions are now subject to modulation related to the periodicity of the\nflow. Analysis of the spatio-temporal growth rates led to the identification of\na saddle point between the Kelvin-Helmholtz mode and the guided jet mode,\ncharacterising an absolute instability mechanism. Frequencies and mode shapes\nrelated to the saddle points for two conditions (associated with axisymmetric\nand helical modes) are compared with screech frequencies and the most energetic\ncoherent structures of screeching jets, resulting in a good agreement for both.\nThe analysis shows that a periodic shock-cell structure has an impulse response\nthat grows upstream, leading to oscillator behaviour. The results suggest that\nscreech can occur in the absence of a nozzle, and that the upstream reflection\ncondition is not essential for screech frequency selection. Connections to\nprevious models are also discussed.", "category": "physics_flu-dyn" }, { "text": "Nonexistence of two-dimensional sessile drops in the diffuse-interface\n model: The diffuse-interface model (DIM) is a widely used tool for modeling fluid\nphenomena involving interfaces -- such as, for example, sessile drops (liquid\ndrops on a solid substrate, surrounded by saturated vapor) and liquid ridges\n(two-dimensional sessile drops). In this work, it is proved that, surprisingly,\nthe DIM does not admit solutions describing static liquid ridges. If, however,\nthe vapor-to-liquid density ratio is small -- as, for example, for water at\nroom temperature -- the ridges can still be observed as quasi-static states, as\ntheir evolution is too slow to be distinguishable from evaporation.\nInterestingly, the nonexistence theorem cannot be extended to axisymmetric\nsessile drops and ridges near a vertical wall, which are not ruled out.", "category": "physics_flu-dyn" }, { "text": "CFD analysis of electroviscous effects in electrolyte liquid flow\n through heterogeneously charged non-uniform microfluidic device: In this work, the pressure-driven flow of symmetric electrolyte liquid\nthrough a heterogeneously charged contraction-expansion (4:1:4) microfluidic\ndevice has been investigated numerically. Total potential ($U$), ion\nconcentrations ($n_\\pm$), velocity (${V}$), and pressure ($P$) fields are\nobtained after solving the mathematical model consisting of the Poisson's,\nNernst-Planck (NP), Navier-Stokes (NS), and continuity equations numerically\nusing the finite element method (FEM). Results are presented for wide ranges of\ndimensionless parameters such as inverse Debye length ($2\\le K\\le 20$), surface\ncharge density ($4\\le S_\\text{1}\\le 16$), and surface charge-heterogeneity\nratio ($0\\le S_\\text{rh}\\le 2$). Results show that the total potential ($\\Delta\nU$) and pressure ($\\Delta P$) drops change maximally by 3511.45% \\add{(0.2127\nto 7.6801)} (at $S_1=4$, $K=20$) and 41.4% \\add{(1.0941 to 1.5471)} (at\n$S_1=16$, $K=2$), respectively with overall enhancing charge-heterogeneity\n($0\\le S_\\text{rh}\\le 2$), over the ranges of $K$ and $S_1$. Electroviscous\ncorrection factor, $Y$ (i.e., ratio of apparent to physical viscosity)\nincreases maximally by 24.39\\% \\add{(1.1158 to 1.3879)} (at $K=4$,\n$S_\\text{rh}=1.75$), 37.52% \\add{(1.0597 to 1.4573)} (at $S_1=16$,\n$S_\\text{rh}=2$), and 41.4% \\add{(1.0306 to 1.4573)} (at $S_1=16$, $K=2$) with\nthe variation of $S_1$ from 4 to 16, $K$ from 20 to 2, and $S_\\text{rh}$ from 0\nto 2, respectively. Further, overall increment in $Y$ is noted as 45.73\\%\n\\add{(1 to 1.4573)} (at $K=2$, $S_1=16$, $S_\\text{rh}=2$), relative to non-EVF\n($S_1=0$ or $K=\\infty$). Thus, charge-heterogeneity enhances electroviscous\neffects in microfluidic devices, which enables the use of present numerical\nresults for designing reliable and essential micro-sized channels for practical\nmicrofluidic applications.", "category": "physics_flu-dyn" }, { "text": "Influence of a thin compressible insoluble liquid film on the eddy\n currents generated by interacting surface waves: Recently the generation of eddy currents by interacting surface waves was\nobserved experimentally. The phenomenon provides the possibility for\nmanipulation of particles which are immersed in the fluid. The analysis shows\nthat the amplitude of the established eddy currents produced by stationary\nsurface waves does not depend on the fluid viscosity in the free surface case.\nThe currents become parametrically larger being inversely proportional to the\nsquare root of the fluid viscosity in the case when the fluid surface is\ncovered by an almost incompressible thin liquid (i.e. shear elasticity is zero)\nfilm formed by an insoluble agent with negligible internal viscous losses as\ncompared to the dissipation in the fluid bulk. Here we extend the theory for a\nthin insoluble film with zero shear elasticity and small shear and dilational\nviscosities on the case of an arbitrary elastic compression modulus. We find\nboth contributions into the Lagrangian motion of passive tracers, which are the\nadvection by the Eulerian vertical vorticity and the Stokes drift. Whereas the\nStokes drift contribution preserves its value for the free surface case outside\na thin viscous sublayer, the Eulerian vertical vorticity strongly depends on\nthe fluid viscosity at high values of the film compression modulus. The Stokes\ndrift acquires a strong dependence on the fluid viscosity inside the viscous\nsublayer, however, the change is compensated by an opposite change in the\nEulerian vertical vorticity. As a result, the vertical dependence of the\nintensity of eddy currents is given by a sum of two decaying exponents with\nboth decrements being of the order of the wave number. The decrements are\nnumerically different, so the Eulerian contribution becomes dominant at some\ndepth for the surface film with any compression modulus.", "category": "physics_flu-dyn" }, { "text": "High-fidelity simulation of pebble beds: Toward an improved\n understanding of the wall channeling effect: Wall channeling is a phenomena of interest for Pebble Bed Reactors (PBRs)\nwhere flow is diverted into high-porosity regions near the wall. This diversion\nof flow can have a significant impact on maximum fuel temperatures and core\nbypass flow. Porous media models that are currently used to model PBRs for\ndesign scoping and transient simulation are lacking in their capabilities to\nmodel the wall channel effect. Recent efforts at Penn State have produced an\nimproved porous media pressure drop equation that is more capable of modeling\nthe velocity variations caused by the wall channel effect in a porous media\nmodel. Several pebble beds were divided into concentric rings of $0.05D_{peb}$,\nand average flow quantities and porosities were extracted for the ring. A\ncorrelation between the form loss coefficient and the local ring porosity was\nfound, allowing for the addition of a correction factor to the form loss term\nof the KTA equation. The developed correlation was purely empirical, and thus a\nmore thorough understanding of the underlying flow phenomena is desired. This\nstudy investigates geometric and flow features that can explain the observed\ncorrelation between the form coefficient and the local porosity that was used\nto generate the improved pressure drop equation. The solid surface area to\nvolume ratio $S_v$ along with the production of Turbulent Kinetic Energy (TKE)\nis analyzed. A relationship between $S_v$ and the local porosity and an inverse\nrelationship between the negative TKE production and the local porosity were\nfound, pointing to the idea that inertial effects caused by different pore\ngeometry in each ring contribute to the variation of the form constant with the\nlocal porosity.", "category": "physics_flu-dyn" }, { "text": "The Intrinsic Fragility of the Liquid-Vapor Interface: A Stress Network\n Perspective: The evolution of the liquid-vapour interface of a Lennard-Jones fluid is\nexamined with molecular dynamics simulations using the intrinsic sampling\nmethod. Results suggest, in agreement with capillary wave theory, clear damping\nof the density profiles as the temperature is increased. We identify a linear\nvariation of the space-filling nature (fractal dimension) of the\nstress-clusters at the intrinsic surface with increasing surface tension, or\nequivalently, with decreasing temperature. A percolation analysis of these\nstress networks indicates that the stress field is more disjointed at higher\ntemperatures. This leads to more fragile interfaces that result in a reduction\nin surface tension at higher temperature.", "category": "physics_flu-dyn" }, { "text": "Non-integrable soliton gas: The Schamel equation framework: Soliton gas or soliton turbulence is a subject of intense studies due to its\ngreat importance to optics, hydrodynamics, electricity, chemistry, biology and\nplasma physics. Usually, this term is used for integrable models where solitons\ninteract elastically. However, soliton turbulence can also be a part of\nnon-integrable dynamics, where long-lasting solutions in the form of almost\nsolitons may exist. In the present paper, the complex dynamics of ensembles of\nsolitary waves is studied within the Schamel equation using direct numerical\nsimulations. Some important statistical characteristics (distribution\nfunctions, moments) are calculated numerically for unipolar and bipolar soliton\ngases. Comparison of results with integrable Korteweg-de Vries (KdV) and\nmodified KdV (mKdV) models are given qualitatively. Our results agree well with\nthe predictions of the KdV equation in the case of unipolar solitons. However,\nin the bipolar case, we observed a notable departure from the mKdV model,\nparticularly in the behavior of kurtosis. The observed increase in kurtosis\nsignifies the amplification of distribution function tails, which, in turn,\ncorresponds to the presence of high-amplitude waves.", "category": "physics_flu-dyn" }, { "text": "Learning nonlocal constitutive models with neural networks: Constitutive and closure models play important roles in computational\nmechanics and computational physics in general. Classical constitutive models\nfor solid and fluid materials are typically local, algebraic equations or flow\nrules describing the dependence of stress on the local strain and/or\nstrain-rate. Closure models such as those describing Reynolds stress in\nturbulent flows and laminar--turbulent transition can involve transport PDEs\n(partial differential equations). Such models play similar roles to\nconstitutive relation, but they are often more challenging to develop and\ncalibrate as they describe nonlocal mappings and often contain many submodels.\nInspired by the structure of the exact solutions to linear transport PDEs, we\npropose a neural network representing a region-to-point mapping to describe\nsuch nonlocal constitutive models. The range of nonlocal dependence and the\nconvolution structure are derived from the formal solution to transport\nequations. The neural network-based nonlocal constitutive model is trained with\ndata. Numerical experiments demonstrate the predictive capability of the\nproposed method. Moreover, the proposed network learned the embedded submodel\nwithout using data from that level, thanks to its interpretable mathematical\nstructure, which makes it a promising alternative to traditional nonlocal\nconstitutive models.", "category": "physics_flu-dyn" }, { "text": "Interactions between a propagating detonation wave and water spray cloud\n in hydrogen/air mixture: Inhibition of hydrogen explosion is crucial to realize its wide applications\nand fine water spray is an ideal mitigant due to numerous advantages. In this\nwork, interactions between a propagating hydrogen/air detonation wave and\ncircular water cloud are numerically studied. Eulerian-Lagrangian method\ninvolving two-way gas-droplets coupling is applied, with a two-dimensional\nconfiguration. Different droplet (diameter, concentration) and cloud (diameter)\nproperties are considered. Our results show that droplet size, concentration\nand cloud radius have significant effects on peak pressure trajectory of the\ndetonation wave. After interacting with cloud, the detonation wave exhibits\nthree propagation modes, including perturbed propagation, leeward\nre-detonation, and detonation extinction. Leeward re-detonation is analyzed\nfrom unsteady evolutions of gas and liquid droplet quantities. The refracted\ndetonation wave inside the cloud is decoupled and propagates more slowly than\nthe one outside the cloud. The detonation re-initiation is from a local hot\nspot, caused by shock focusing from upper and lower diffracted detonations.\nDisintegration of water droplets proceeds when the detonation wave crosses the\ncloud and multiphase interfacial instability is observed due to the difference\nin effective density of the two fluids. Furthermore, detonation extinction is\nobserved when we consider various water cloud size. It is featured by quickly\nfading peak pressure trajectories when the detonation passes the cloud, and no\nlocal autoignition occurs in the shock focusing area. Evolutions of\nthermochemical structures from the shocked area in an extinction process are\nalso studied. Moreover, parametric studies considering various droplet\nconcentrations and cloud radii are performed.", "category": "physics_flu-dyn" }, { "text": "A Quantum Inspired Approach to Exploit Turbulence Structures: Understanding turbulence is the key to our comprehension of many natural and\ntechnological flow processes. At the heart of this phenomenon lies its\nintricate multi-scale nature, describing the coupling between different-sized\neddies in space and time. Here we introduce a new paradigm for analyzing the\nstructure of turbulent flows by quantifying correlations between different\nlength scales using methods inspired from quantum many-body physics. We present\nresults for interscale correlations of two paradigmatic flow examples, and use\nthese insights along with tensor network theory to design a structure-resolving\nalgorithm for simulating turbulent flows. With this algorithm, we find that the\nincompressible Navier-Stokes equations can be accurately solved within a\ncomputational space reduced by over an order of magnitude compared to direct\nnumerical simulation. Our quantum-inspired approach provides a pathway towards\nconducting computational fluid dynamics on quantum computers.", "category": "physics_flu-dyn" }, { "text": "Swimming bacteria in Poiseuille flow: the quest for active\n Bretherton-Jeffery trajectories: Using a 3D Lagrangian tracking technique, we determine experimentally the\ntrajectories of non-tumbling E. coli mutants swimming in a Poiseuille flow. We\nidentify a typology of trajectories in agreement with a kinematic \"active\nBretherton-Jeffery\" model, featuring an axi-symmetric self-propelled ellipsoid.\nIn particular, we recover the \"swinging\" and \"shear tumbling\" kinematics\npredicted theoretically by Z\\\"ottl et al. Moreover using this model, we derive\nanalytically new features such as quasi-planar piece-wise trajectories,\nassociated with the high aspect ratio of the bacteria, as well as the existence\nof a drift angle around which bacteria perform closed cyclic trajectories.\nHowever, the agreement between the model predictions and the experimental\nresults remains local in time, due to the presence of Brownian rotational\nnoise.", "category": "physics_flu-dyn" }, { "text": "A Simple Analytical Model of Coupled Single Flow Channel over Porous\n Electrode in Vanadium Redox Flow Battery with Serpentine Flow Channel: A simple analytical model of a layered system comprised of a single passage\nof a serpentine flow channel and a parallel underlying porous electrode (or\nporous layer) is proposed. This analytical model is derived from Navier-Stokes\nmotion in the flow channel and Darcy-Brinkman model in the porous layer. The\ncontinuities of flow velocity and normal stress are applied at the interface\nbetween the flow channel and the porous layer. The effects of the inlet\nvolumetric flow rate, thickness of the flow channel and thickness of a typical\ncarbon fiber paper porous layer on the volumetric flow rate within this porous\nlayer are studied. The maximum current density based on the electrolyte\nvolumetric flow rate is predicted, and found to be consistent with reported\nnumerical simulation. It is found that, for a mean inlet flow velocity of 33.3\ncm s-1, the analytical maximum current density is estimated to be 377 mA cm-2,\nwhich compares favorably with experimental result reported by others of ~400 mA\ncm-2.", "category": "physics_flu-dyn" }, { "text": "Order and Chaos: Collective Behavior of Crowded Drops in Microfluidic\n Systems: Sindy Tang is assistant professor at in the Department of Mechanical\nEngineering at Stanford University. In this contribution she describes how her\nteam uses droplet microfluidics to identify bacteria that could increase the\nefficiency of generation of bioplastics, and how this work motivated them to\ninvestigate the physics and design criteria necessary for further up-scaling of\nthe droplet technology.", "category": "physics_flu-dyn" }, { "text": "Numerical study on mechanism of C-J deflagration: The mechanism of detonation instability and deflagration-to-detonation\ntransition is studied by one-dimensional numerical simulation with overall\none-step chemical reaction kinetics in this paper. The detonation is ignited at\nthe left closed end of the one-dimensional detonation tube and propagates\ndownstream. The activation energy is increased to trigger the instability of\ndetonation. The numerical results show that the C-J detonation is stable at\nlower activation energy. The stable detonation does not have the von Neumann\nspike and the gas Mach number at detonation front is subsonic. The von Neumann\nspike appears and the gas Mach number becomes supersonic as the activation\nenergy is increased. The detonation instability appears with the von Neumann\nspike synchronously. At very higher activation energy, the detonation quenches\nabruptly and degenerates into a C-J deflagration. The detonation is\nextinguished abruptly by the rarefaction wave induced by the higher von Neumann\nspike. Then the rarefaction wave moves in front of the heat release region and\nweakens the leading shock wave gradually. The C-J deflagration is composed of a\nprecursor shock wave and a flame front, and the flame front is completely\ndecoupled from the shock wave. The gas static temperature behind the leading\nshock wave is too low to ignite the mixture. The rarefaction wave from the wall\nceases the mixture behind the leading shock, increases its static temperature\nand decrease its pressure. As a result, the combustion takes place at the\ninterface. The pressure rise caused by the combustion at the interface offsets\nthe influence of rarefaction wave, and this mechanism makes the C-J\ndeflagration propagate downstream with a relatively constant velocity for a\nlong time.", "category": "physics_flu-dyn" }, { "text": "MHD flow and heat transfer due to the axisymmetric stretching of a sheet\n with induced magnetic field: The full MHD equations, governing the flow due to the axisymmetric stretching\nof a sheet in the presence of a transverse magnetic field, can be cast in a\nself similar form. This allows evaluation of the induced magnetic field and its\neffect on the flow and heat transfer. The problem involves three parameters-\nthe magnetic Prandtl number, the magnetic interaction number, and the Prandtl\nnumber. Numerical solutions are obtained for the velocity field, the magnetic\nfield, and the temperature, at different values of the magnetic Prandtl number\nand the magnetic interaction number. The contributions of the viscous\ndissipation, Joule heating, and streamwise diffusion to the heat flux toward\nthe sheet are assessed.", "category": "physics_flu-dyn" }, { "text": "Uncertainty amplification due to density/refractive-index gradients in\n volumetric PTV and BOS experiments: We theoretically analyze the effect of density/refractive-index gradients on\nthe measurement precision of Volumetric Particle Tracking Velocimetry (V-PTV)\nand Background Oriented Schlieren (BOS) experiments by deriving the Cramer-Rao\nlower bound (CRLB) for the 2D centroid estimation process. A model is derived\nfor the diffraction limited image of a particle or dot viewed through a medium\ncontaining density gradients that includes the effect of various experimental\nparameters such as the particle depth, viewing direction and f-number. Using\nthe model we show that non-linearities in the density gradient field lead to\nblurring of the particle/dot image. This blurring amplifies the effect of image\nnoise on the centroid estimation process, leading to an increase in the CRLB\nand a decrease in the measurement precision. The ratio of position\nuncertainties of a dot in the reference and gradient images is a function of\nthe ratio of the dot diameters and dot intensities. We term this parameter the\nAmplification Ratio (AR), and we propose a methodology for estimating the\nposition uncertainties in tracking-based BOS measurements. The theoretical\npredictions of the particle/dot position estimation variance from the CRLB are\ncompared to ray tracing simulations with good agreement. The uncertainty\namplification is also demonstrated on experimental BOS images of flow induced\nby a spark discharge, where we show that regions of high amplification ratio\ncorrespond to regions of density gradients. This analysis elucidates the\ndependence of the position error on density and refractive-index gradient\ninduced distortion parameters, provides a methodology for accounting its effect\non uncertainty quantification and provides a framework for optimizing\nexperiment design.", "category": "physics_flu-dyn" }, { "text": "Multiphase CO2 Dispersions in Microfluidics: Formation, Phases, and Mass\n Transfer: The dissolution and microfluidic mass transfer of carbon dioxide in water at\nhigh-pressure conditions are crucial for a myriad of technological\napplications, including microreactors, extractions, and carbon capture,\nutilization, and sequestration (CCUS) processes. In this experimental work, we\nuse a high-pressure microfluidic method to elucidate the mass transfer process\nof CO2 in water at high pressure. An intriguing multiphase CO2 flow and\ndispersions are observed when operating at the pressure-temperature ($P$-$T$)\ncondition close to the CO2 gas-liquid phase boundary ($P=6.5$ MPa and\n$T=23.5\\pm 0.5~^{\\circ}$C). We propose a series of strategies to unravel this\ncomplex multi-phase dynamics by calculating each phase's volume and mass change\nin a gas-liquid coexistent CO2 dispersion, estimating the possible CO2\nconcentration change in water, and comparing with the CO2 solubility data.\nFinally, we quantify the CO2 mass transfer by directly calculating the CO2\ndissolution rate in water and estimating the volumetric mass transfer\ncoefficient ($k_{L}a$). The results show that the mass transfer may be\ninfluenced by the specific area ($a$), CO2 concentration gradient in the water\nslug, and the traveling speed of a dispersion.", "category": "physics_flu-dyn" }, { "text": "DeepClouds.ai: Deep learning enabled computationally cheap direct\n numerical simulations: Simulation of turbulent flows, especially at the edges of clouds in the\natmosphere, is an inherently challenging task. Hitherto, the best possible\ncomputational method to perform such experiments is the Direct Numerical\nSimulation (DNS). DNS involves solving non-linear partial differential\nequations for fluid flows, also known as Navier-Stokes equations, on\ndiscretized grid boxes in a three-dimensional space. It is a valuable paradigm\nthat has guided the numerical weather prediction models to compute rainfall\nformation. However, DNS cannot be performed for large domains of practical\nutility to the weather forecast community. Here, we introduce DeepClouds.ai, a\n3D-UNET that simulates the outputs of a rising cloud DNS experiment. The\nproblem of increasing the domain size in DNS is addressed by mapping an inner\n3D cube to the complete 3D cube from the output of the DNS discretized grid\nsimulation. Our approach effectively captures turbulent flow dynamics without\nhaving to solve the complex dynamical core. The baseline shows that the deep\nlearning-based simulation is comparable to the partial-differential\nequation-based model as measured by various score metrics. This framework can\nbe used to further the science of turbulence and cloud flows by enabling\nsimulations over large physical domains in the atmosphere. It would lead to\ncascading societal benefits by improved weather predictions via advanced\nparameterization schemes.", "category": "physics_flu-dyn" }, { "text": "Universal relaxation of turbulent binary fluids: Upon quenching the forcing, a turbulent system tends to attain the state of\nstable equilibrium through the process of turbulent relaxation. Such relaxation\nin binary fluids is of surmount interest for both fundamental science\nunderstanding and industrial applications. A systematic investigation of the\nsame has been carried out, for the first time, using direct numerical\nsimulations of Cahn-Hilliard-Navier-Stokes equations. With the help of a\nthorough scanning, the bulk of each fluid and its interface are found to relax\nin a different way. However, using the principle of vanishing nonlinear\ntransfer, we propose a convincing, universal pathway of obtaining the turbulent\nrelaxed states for both the bulk and the interface which attain a relaxed state\nwhen the turbulent cascades of the inviscid invariants are suppressed.\nInterestingly, the relaxation of the bulk turns up to be subtly different from\nthe turbulent relaxation of a single hydrodynamic fluid and the interface\nrelaxation is found to follow a Helmholtz-like pressure-balanced condition.", "category": "physics_flu-dyn" }, { "text": "Explicit predictability and dispersion scaling exponents in fully\n developed turbulence: We apply a simple method to provide explicit expressions for different\nscaling exponents in intermittent fully developed turbulence, that before were\nonly given through a Legendre transform. This includes predictability exponents\nfor infinitesimal and non infinitesimal perturbations, Lagrangian velocity\nexponents, and dispersion exponents. We obtain also new results concerning\ninverse statistics corresponding to exit-time moments.", "category": "physics_flu-dyn" }, { "text": "Mist Flow Visualization for Round Jets in Aerosol Jet Printing: With the microdroplets of water serving as light scattering particles, the\nmist flow patterns of round micro-jets can be visualized using the Aerosol\nJet(R) direct-write system. The visualization images show that the laminar mist\njet appears to extend to more than 20 times the diameter of nozzle orifice, D,\nfor jet Reynolds number Re < 600, especially with D = 0.3 mm and less. For\nsmaller jets (e.g., with D = 0.15 mm), laminar collimated mist flow might be\nretained to 40xD for Re < 600 and for Re ~ 1500 within 20xD from the nozzle.\nThe laminar part of mist flow associated with larger jets (e.g., with D = 1.0\nmm for Re < 600) tends to exhibit noticeable gradual widening due to viscous\ndiffusion. For free jets, their breakdown length--the distance from nozzle\nwhere transition from laminar to turbulent mist flow takes place as signaled by\nthe inception of a rapid widening of mist stream--is shown to decrease with\nincreasing Re. The presence of impingement wall tends to prevent turbulence\ndevelopment, even when the wall is placed further downstream of the free-jet\nbreakdown length for a given Re . The critical Re for an impinging jet to\ndevelop turbulence increases as the standoff S is reduced. The mist flow of\nimpinging jet of D = 1.0 mm seems to remain laminar even for Re > 4000 at S =\n12 mm.", "category": "physics_flu-dyn" }, { "text": "How a \"pinch of salt\" can tune chaotic mixing of colloidal suspensions: Efficient mixing of colloids, particles or molecules is a central issue in\nmany processes. It results from the complex interplay between flow deformations\nand molecular diffusion, which is generally assumed to control the\nhomogenization processes. In this work we demonstrate on the contrary that\ndespite fixed flow and self-diffusion conditions, the chaotic mixing of\ncolloidal suspensions can be either boosted or inhibited by the sole addition\nof trace amount of salt as a co-mixing species. Indeed, this shows that local\nsaline gradients can trigger a chemically-driven transport phenomenon,\ndiffusiophoresis, which controls the rate and direction of molecular transport\nfar more efficiently than usual Brownian diffusion. A simple model combining\nthe elementary ingredients of chaotic mixing with diffusiophoretic transport of\nthe colloids allows to rationalize our observations and highlights how\nsmall-scale out-of-equilibrium transport bridges to mixing at much larger\nscales in a very effective way. Considering chaotic mixing as a prototypal\nbuilding block for turbulent mixing, this suggests that these phenomena,\noccurring whenever the chemical environment is inhomogeneous, might bring\ninteresting perspective from micro-systems up to large-scale situations, with\nexamples ranging from ecosystems to industrial contexts.", "category": "physics_flu-dyn" }, { "text": "On dynamics of nonmagnetic accretion disks: Axisymmetric accretion disks in vicinity of a central compact body are\nstudied. In the case of non-viscous disk it is proven that all solutions for\nthe midplane circular velocity are unstable. Hence, the pure hydrodynamic\nturbulence in accretion disks is possible. It is disproved the well-known\narguments that an inviscid accretion disk must be sub-Keplerian. It is also\ndemonstrated that the regular asymptotic solutions, often used in astrophysics,\ncan lead to erroneous conclusions. It is proven that a laminar viscous disk can\nbe approximated with a great precision by the vortex motion. Assuming that a\nturbulent gas tends to flow with minimal losses, we have shown that a turbulent\ndisk tends to be Keplerian.", "category": "physics_flu-dyn" }, { "text": "Internal Gravity Waves in a Stratified Fluid with Smoothly Varying\n Bottom: The far field asymptotic of internal waves is constructed for the case when a\npoint source of mass moves in a layer of arbitrarily stratified fluid with\nslowly varying bottom. The solutions obtained describe the far field both near\nthe wave fronts of each individual mode and away from the wave fronts and are\nexpansions in Airy or Fresnel waves with the argument determined from the\nsolution of the corresponding eikonal equation. The amplitude of the wave field\nis determined from the energy conservation law along the ray tube. For model\ndistributions of the bottom shape and the stratification describing the typical\npattern of the ocean shelf exact analytic expressions are obtained for the\nrays, and the properties of the phase structure of the wave field are analyzed.", "category": "physics_flu-dyn" }, { "text": "An upper bound for passive scalar diffusion in shear flows: This study is concerned with the diffusion of a passive scalar $\\Theta(\\r,t)$\nadvected by general $n$-dimensional shear flows $\\u=u(y,z,...,t)\\hat{x}$ having\nfinite mean-square velocity gradients. The unidirectionality of the\nincompressible flows conserves the stream-wise scalar gradient,\n$\\partial_x\\Theta$, allowing only the cross-stream components to be amplified\nby shearing effects. This amplification is relatively weak because an important\ncontributing factor, $\\partial_x\\Theta$, is conserved, effectively rendering a\nslow diffusion process. It is found that the decay of the scalar variance\n$<\\Theta^2>$ satisfies $d<\\Theta^2>/dt\\ge -C\\kappa^{1/3}$, where $C>0$ is a\nconstant, depending on the fluid velocity gradients and initial distribution of\n$\\Theta$, and $\\kappa$ is the molecular diffusivity. This result generalizes to\naxisymmetric flows on the plane and on the sphere having finite mean-square\nangular velocity gradients.", "category": "physics_flu-dyn" }, { "text": "Toward a structural understanding of turbulent drag reduction: nonlinear\n coherent states in viscoelastic shear flows: Nontrivial steady flows have recently been found that capture the main\nstructures of the turbulent buffer layer. We study the effects of polymer\naddition on these \"exact coherent states\" (ECS) in plane Couette flow. Despite\nthe simplicity of the ECS flows, these effects closely mirror those observed\nexperimentally: Structures shift to larger length scales, wall-normal\nfluctuations are suppressed while streamwise ones are enhanced, and drag is\nreduced. The mechanism underlying these effects is elucidated. These results\nsuggest that the ECS are closely related to buffer layer turbulence.", "category": "physics_flu-dyn" }, { "text": "Flow Regime Transition in Countercurrent Packed Column Monitored by ECT: Vertical packed columns are widely used in absorption, stripping and\ndistillation processes. Flooding will occur in the vertical packed columns as a\nresult of excessive liquid accumulation, which reduces mass transfer efficiency\nand causes a large pressure drop. Pressure drop measurements are typically used\nas the hydrodynamic parameter for predicting flooding. They are, however, only\nindicative of the occurrence of transition of the flow regime across the packed\ncolumn. They offer limited spatial information to mass transfer packed column\noperators and designers. In this work, a new method using Electrical\nCapacitance Tomography (ECT) is implemented for the first time so that\nreal-time flow regime monitoring at different vertical positions is achieved in\na countercurrent packed bed column using ECT. Two normalisation methods are\nimplemented to monitor the transition from pre-loading to flooding in a column\nof 200 mm diameter, 1200 mm height filled with plastic structured packing.\nLiquid distribution in the column can be qualitatively visualised via\nreconstructed ECT images. A flooding index is implemented to quantitatively\nindicate the progression of local flooding. In experiments, the degree of local\nflooding is quantified at various gas flow rates and locations of ECT sensor.\nECT images were compared with pressure drop and visual observation. The\nexperimental results demonstrate that ECT is capable of monitoring liquid\ndistribution, identifying flow regime transitions and predicting local\nflooding.", "category": "physics_flu-dyn" }, { "text": "Numerical assessment of the penetroviscometer approach for large, rapid\n and transient shear deformations: The rheological characterisation of complex fluids is mostly performed under\nsimple shear flow in rotational rheometers. Their modern commercial versions\nare extremely sensitive instruments which are able to provide very accurate\nmeasurements of low values of different material functions, such as viscosity,\nviscoelastic moduli, etc., under certain ideal flow conditions. Nevertheless,\nthey fail in providing reliable data when characterising the response of\ncomplex fluids at short time scales, due to artefacts induced by either\ninstrument or fluid inertia. This is crucial in the analysis of the rheological\nproperties of new formulations of shear thickening fluids specifically\ndeveloped for protective applications, in which the performance is extremely\nlinked to the time dependent structural changes provoked by the sudden impact\nloads. Thus, the necessity of providing a reliable experimental tool able to\nimpose large, rapid, transient shear deformation for their rheological\ncharacterisation in conditions similar to those of the applications becomes\nevident. This numerical study aims at assessing the potential use of the\npenetroviscometer for the measurement of the transient shear viscosity of\ncomplex fluids at short time scales beyond the current limits of commercial\nrotational rheometers.", "category": "physics_flu-dyn" }, { "text": "Experimental study of fluid flows in a precessing cylindrical annulus: The flow inside a precessing fluid cavity has been given particular attention\nsince the end of the 19th century in geophysical and industrial contexts. The\npresent study aims at shedding light on the underlying mechanism by which the\nflow inside a precessing cylindrical annulus transitions from laminar to\nmultiple scale complex structures. We address this problem experimentally using\nultrasonic Doppler velocimetry to diagnose the fluid velocity in a rotating and\nprecessing cylindrical annulus. When precession is weak, the flow can be\ndescribed as a superposition of forced inertial modes. Above a critical value\nof the precession rate, the forced flow couples with two free inertial modes\nsatisfying triadic resonance conditions, leading to the classical growth and\ncollapse. Using a Bayesian approach, we extract the wavenumber, frequency,\ngrowth rate and amplitude of each mode involved in the instability. In some\ncases, we observe for the first time ever experimentally two pairs of free\nmodes coexisting with the forced flow. At larger precession rates, we do not\nobserve triadic resonance any more, instead we observe several harmonics whose\nfrequencies are integer multiples of the rotation frequency.", "category": "physics_flu-dyn" }, { "text": "Dynamics of Evaporating Respiratory Droplets in the Vicinity of Vortex\n Dipoles: A new mathematical analysis of exhaled respiratory droplet dynamics and\nsettling distances in the vicinity of vortical environments is presented.\nRecent experimental and theoretical studies suggest that vortical flow\nstructures may enhance the settling distances of exhaled respiratory droplets\nbeyond the two-meter distancing rule recommended by health authorities lately.\nWe propose a mathematical framework to study the underlying physical mechanism\nresponsible for the entrapment and subsequently delayed settling times of\nevaporating droplets and solid particles. A dipolar vortex is considered\nself-propelling through a cloud of micron-sized evaporating droplets. This\nconfiguration might be utilized to approximate an indoor environment in which\nsimilar unsteady vortical flow structures interact with exhaled respiratory\ndroplets. We demonstrate the vortex dipole effect on droplet and solid\nparticles settling distances, depending on the evaporation rate, the vorticity\nof the dipole, and the droplet's initial diameter and location relative to the\nvortex core. Our theoretical analysis reveals non-intuitive interactions\nbetween the vortex dipole, droplet relaxation time, gravity, and mass transfer.\nThe existence of optimal conditions for maximum displacement is suggested,\nwhere the droplet entrainment reaches up to an order of magnitude larger than\nthe vortex core length scale. We present a basic model that may be applied for\nevaluating the spread of exhaled respiratory droplets in vortical environments.\nOur theoretical study suggests that exhaled respiratory droplets initially at\nrest can translate to significant distances, hence implying that vortical flow\nmight enhance the transmission of airborne pathogens.", "category": "physics_flu-dyn" }, { "text": "Transition phenomena in unstably stratified turbulent flows: We study experimentally and theoretically transition phenomena caused by the\nexternal forcing from Rayleigh-Benard convection with the large-scale\ncirculation (LSC) to the limiting regime of unstably stratified turbulent flow\nwithout LSC whereby the temperature field behaves like a passive scalar. In the\nexperiments we use the Rayleigh-B\\'enard apparatus with an additional source of\nturbulence produced by two oscillating grids located nearby the side walls of\nthe chamber. When the frequency of the grid oscillations is larger than 2 Hz,\nthe large-scale circulation (LSC) in turbulent convection is destroyed, and the\ndestruction of the LSC is accompanied by a strong change of the mean\ntemperature distribution. However, in all regimes of the unstably stratified\nturbulent flow the ratio $\\big[(\\ell_x \\nabla_x T)^2 + (\\ell_y \\nabla_y T)^2 +\n(\\ell_z \\nabla_z T)^2\\big] / < \\theta^2 >$ varies slightly (even in the range\nof parameters whereby the behaviour of the temperature field is different from\nthat of the passive scalar). Here $\\ell_i$ are the integral scales of\nturbulence along x, y, z directions, T and \\theta are the mean and fluctuating\nparts of the fluid temperature. At all frequencies of the grid oscillations we\nhave detected the long-term nonlinear oscillations of the mean temperature. The\ntheoretical predictions based on the budget equations for turbulent kinetic\nenergy, turbulent temperature fluctuations and turbulent heat flux, are in\nagreement with the experimental results.", "category": "physics_flu-dyn" }, { "text": "Transition to turbulence in particle laden flows: Suspended particles can alter the properties of fluids and in particular also\naffect the transition from laminar to turbulent flow. In the present\nexperimental study, we investigate the impact of neutrally buoyant, spherical\ninertial particles on transition to turbulence in a pipe flow. At low particle\nconcentrations, like in single phase Newtonian fluids, turbulence only sets in\nwhen triggered by sufficiently large perturbations and, as characteristic for\nthis transition localized turbulent regions (puffs) co-exist with laminar flow.\nIn agreement with earlier studies this transition point initially moves to\nlower Reynolds number (Re) as the particle concentration increases. At higher\nconcentrations however the nature of the transition qualitatively changes:\nLaminar flow gives way to a globally fluctuating state following a continuous,\nnon-hysteretic transition. A further increase in Re results in a secondary\ninstability where localized puff-like structures arise on top of the uniformly\nfluctuating background flow. At even higher concentration only the uniformly\nfluctuating flow is found and signatures of Newtonian type turbulence are no\nlonger observed.", "category": "physics_flu-dyn" }, { "text": "Ventilation regime in a karstic system (Milandre Cave, Switzerland): Cave climatology and its impact on contemporary biogeochemical cycles are\nstill poorly documented. Ventilation in karst environment plays a fundamental\nrole in these two fields and its understanding could bring elements to study\nthem. However, only a few cavers have tried to understand and describe it, very\noften in a qualitative way or by theoretical approaches. The aim of this study\nis to test physical concepts with empirical data. For this purpose, a\nventilation model has been built and compared with field temperature and air\nvelocity measurements in the Milandre Cave Laboratory (Switzerland). The model\nexplains about 95% of the measured airflow thus confirming the major role of\ntemperature on the air dynamics. However, these first results also reveal that\nthe measured winter air flow is lower than predicted by the model and that the\nair flow reversal occurs at a lower temperature than anticipated. Combined with\na forced ventilation experiment these results underline the influence of the\natmospheric composition (particularly the water vapor and concentration in\nCO$_2$ and O$_2$), waterflow rates and network geometry on the air flow. This\nwork paves the way for a better quantification of heat and mass fluxes in\nrelation to underground ventilation.", "category": "physics_flu-dyn" }, { "text": "Ring bursting behavior en route to turbulence in quasi two-dimensional\n Taylor-Couette flows: We investigate the quasi two-dimensional Taylor-Couette system in the regime\nwhere the radius ratio is close to unity - a transitional regime between three\nand two dimensions. By systematically increasing the Reynolds number we observe\na number of standard transitions, such as one from the classical Taylor vortex\nflow (TVF) to wavy vortex flow (WVF), as well as the transition to fully\ndeveloped turbulence. Prior to the onset of turbulence we observe intermittent\nburst patterns of localized turbulent patches, confirming the experimentally\nobserved pattern of very short wavelength bursts (VSWBs). A striking finding is\nthat, for Reynolds number larger than the onset of VSWBs, a new type of\nintermittently bursting behaviors emerge: burst patterns of azimuthally closed\nrings of various orders. We call them ring-burst patterns, which surround the\ncylinder completely but remain localized and separated by non-turbulent mostly\nwavy structures in the axial direction. We use a number of quantitative\nmeasures, including the cross-flow energy, to characterize the ring-burst\npatterns and to distinguish them from the background flow. The ring-burst\npatterns are interesting because it does not occur in either three- or\ntwo-dimensional Taylor-Couette flow: it occurs only in the transition, quasi\ntwo-dimensional regime of the system, a regime that is less studied but\ncertainly deserves further attention so as to obtain deeper insights into\nturbulence.", "category": "physics_flu-dyn" }, { "text": "Direct numerical simulation of transition under free-stream turbulence\n and the influence of large integral length scales: Under action of free-stream turbulence (FST), elongated streamwise streaky\nstructures are generated inside the boundary layer, and their amplitude and\nwavelength are crucial for the transition onset. While turbulence intensity is\nstrongly correlated with the transitional Reynolds number, characteristic\nlength scales of the FST are often considered to have a slight impact on the\ntransition location. However, a recent experiment by Fransson & Shahinfar\n(2020} shows significant effects of FST scales. They found that, for higher\nfree-stream turbulence levels and larger integral length scales, an increase in\nthe length scale postpones transition, contrary to established literature.\nHere, we aim at understanding these results by performing a series of\nhigh-fidelity simulations. These results provide understanding why the FST\nintegral length scale affects the transition location differently. These\nintegral length scales are so large that the wide streaks introduced in the\nboundary layer have substantially lower growth in the laminar region upstream\nof the transition to turbulence, than streaks induced by smaller integral\nlength scales. The energy in the boundary layer subsequently propagate to\nsmaller spanwise scales as a result of the non-linear interaction. When the\nenergy has reached smaller spanwise scales larger amplitude streaks results in\nregions where the streak growth are larger. It takes longer for the energy from\nthe wider streaks to propagate to the spanwise scales associated with the\nbreakdown to turbulence, than for the those with smaller spanwise scales. Thus\nthere is a faster transition for FST with lower integral length scales in this\ncase.", "category": "physics_flu-dyn" }, { "text": "Evidence of thin-film precursors formation in hydrokinetic and atomistic\n simulations of nano-channel capillary filling: We present hydrokinetic Lattice Boltzmann and Molecular Dynamics simulations\nof capillary filling of high-wetting fluids in nano-channels, which provide\nclear evidence of the formation of thin precursor films, moving ahead of the\nmain capillary front. The dynamics of the precursor films is found to obey the\nLucas-Washburn law as the main capillary front, z2(t) proportional to t,\nalthough with a larger prefactor, which we find to take the same value for both\ngeometries under inspection. Both hydrokinetic and Molecular Dynamics\napproaches indicate a precursor film thickness of the order of one tenth of the\ncapillary diameter. The quantitative agreement between the hydrokinetic and\natomistic methods indicates that the formation and propagation of thin\nprecursors can be handled at a mesoscopic/hydrokinetic level, thereby opening\nthe possibility of using hydrokinetic methods to space-time scales and complex\ngeometries of direct experimental relevance.", "category": "physics_flu-dyn" }, { "text": "Large Eddy Simulation of urban boundary layer flows using a canopy\n stress method: Large-eddy simulation (LES) of a turbulent flow through an array of\nbuilding-like obstacles is an idealized model to study transport of\ncontaminants in the urban atmospheric boundary layer (UABL). A reasonably\naccurate LES prediction of turbulence in such an UABL must resolve a\nsignificant proportion of the small but energetic eddies in the roughness\nsublayer, which remains prohibitive even though computational power has been\nincreased significantly. In this article, we present a large-eddy simulation\nmethodology to study turbulence in UABLs, where the turbulence closure is based\non coupling the eddy viscosity method with the canopy stress method. Unlike the\nclassical Smagorinsky model that considers only the strain portion of the\nvelocity gradient tensor, we consider both the strain tensor and the rotation\ntensor to compute the eddy viscosity. This allows us to dynamically adapt the\nrate of energy dissipation to the scales of the energetic eddies in the\nroughness sublayer. Without employing a mesh conforming to the urban roughness\nelements, the effect of such solid bodies are represented in the LES model\nthrough a canopy stress method in which the loss of pressure and the sink of\nmomentum due to the interaction between eddies and roughness elements are\nparameterized using the instantaneous velocity field. Simulation results of the\nproposed canopy stress method is compared with that of a conventional\nComputational Fluid Dynamics (CFD) method employing a block-structured mesh\nconforming around the roughness elements. For urban flow simulations, the\nresults demonstrate that the proposed canopy stress model is accurate in\npredicting vertical profiles of mean and variance, as well as the temporal\nintermittency of coherent structures.", "category": "physics_flu-dyn" }, { "text": "Mechanism of flame acceleration and detonation transition from the\n interaction of a supersonic turbulent flame with an obstruction: The present paper seeks to determine the mechanism of flame acceleration and\ntransition to detonation when a turbulent flame preceded by a shock interacts\nwith a single obstruction in its path, taken as a cylindrical obstacle or a\nwall in the present study. The problem is addressed experimentally in a mixture\nof propane-oxygen at sub-atmospheric conditions. The turbulent flame was\ngenerated by passing a detonation wave through a perforated plate, yielding\nflames with turbulent burning velocities 10 to 20 larger than the laminar\nvalues and incident shock Mach numbers ranging between 2 and 2.5. Time resolved\nschlieren videos recorded at approximately 100 kHz and numerical reconstruction\nof the flow field permitted to determine the mechanism of flame acceleration\nand transition to detonation. It was found to be the enhancement of the\nturbulent burning rate of the flame through its interaction with the shock\nreflection on the obstacle. The amplification of the burning rate was found to\ndrive the flame burning velocity close to the speed of sound with respect to\nthe fresh gases, resulting in the amplification of a shock in front of the\nflame. The acceleration through this regime resulted in the strengthening of\nthis shock. Detonation was observed in regions of non-planarity of this\ninternal shock, inherited by the irregular shape of the turbulent flame itself.\nAuto-ignition at early times of this process was found to be negligibly slow\ncompared with the flow evolution time scale in the problem investigated,\nsuggesting that the relevant time scale is primarily associated with the\nincrease in turbulent burning rate by the interaction with reflected shocks.", "category": "physics_flu-dyn" }, { "text": "Squeeze cementing of micro-annuli: a visco-plastic invasion flow: Squeeze cementing is a process used to repair leaking oil and gas wells, in\nwhich a cement slurry is driven under pressure to fill an uneven leakage\nchannel. This results in a Hele-Shaw type flow problem involving a yield stress\nfluid. We solve the flow problem using an augmented Lagrangian approach and\nadvect forward the fluid concentrations until the flow stops. A planar invasion\nand a radial (perforation hole) invasion flow are studied. The characteristics\nof the flow penetration are linked to the channel thickness profile. The\ndistribution of streamlines, flowing and non-flowing zones, evolves during the\ninvasion flow. An interesting aspect of the results is the extreme variability\nin penetration metrics computed. These depend not only on the stochastic nature\nof the microannulus thickness, which has significant natural variation in both\nazimuthal and axial directions, but also on the ``luck'' of where the\nperforation hole is, relative to the larger/smaller microannulus gaps. This may\nexplain the unreliability of the process.", "category": "physics_flu-dyn" }, { "text": "Experimental observation of a strong mean flow induced by internal\n gravity waves: We report the experimental observation of a robust horizontal mean flow\ninduced by internal gravity waves. A wave beam is forced at the lateral\nboundary of a tank filled with a linearly stratified fluid initially at rest.\nAfter a transient regime, a strong jet appears in the wave beam, with\nhorizontal recirculations outside the wave beam. We present a simple physical\nmechanism predicting the growth rate of the mean flow and its initial spatial\nstructure. We find good agreement with experimental results.", "category": "physics_flu-dyn" }, { "text": "Reactive solute transport in physically and chemically heterogeneous\n porous media with multimodal reactive mineral facies: The Lagrangian approach: Physical and chemical heterogeneities have a large impact on reactive\ntransport in porous media. Examples of heterogeneous attributes affecting\nreactive mass transport are the hydraulic conductivity (K), and the equilibrium\nsorption distribution coefficient (Kd). This paper uses the Deng et al. (2013)\nconceptual model for multimodal reactive mineral facies and a Lagrangian-based\nstochastic theory in order to analyze the reactive solute dispersion in\nthree-dimensional anisotropic heterogeneous porous media with hierarchical\norganization of reactive minerals. An example based on real field data is used\nto illustrate the time evolution trends of reactive solute dispersion. The\nresults show that the correlation between the hydraulic conductivity and the\nequilibrium sorption distribution coefficient does have a significant effect on\nreactive solute dispersion. The anisotropy ratio does not have a significant\neffect on reactive solute dispersion. Furthermore, through a sensitivity\nanalysis we investigate the impact of changing the mean, variance, and integral\nscale of K and Kd on reactive solute dispersion.", "category": "physics_flu-dyn" }, { "text": "Incomplete to complete multiphysics forecasting -- a hybrid approach for\n learning unknown phenomena: Modeling complex dynamical systems with only partial knowledge of their\nphysical mechanisms is a crucial problem across all scientific and engineering\ndisciplines. Purely data-driven approaches, which only make use of an\nartificial neural network and data, often fail to accurately simulate the\nevolution of the system dynamics over a sufficiently long time and in a\nphysically consistent manner. Therefore, we propose a hybrid approach that uses\na neural network model in combination with an incomplete partial differential\nequations (PDE) solver that provides known, but incomplete physical\ninformation. In this study, we demonstrate that the results obtained from the\nincomplete PDEs can be efficiently corrected at every time step by the proposed\nhybrid neural network - PDE solver model, so that the effect of the unknown\nphysics present in the system is correctly accounted for. For validation\npurposes, the obtained simulations of the hybrid model are successfully\ncompared against results coming from the complete set of PDEs describing the\nfull physics of the considered system. We demonstrate the validity of the\nproposed approach on a reactive flow, an archetypal multi-physics system that\ncombines fluid mechanics and chemistry, the latter being the physics considered\nunknown. Experiments are made on planar and Bunsen-type flames at various\noperating conditions. The hybrid neural network - PDE approach correctly models\nthe flame evolution of the cases under study for significantly long time\nwindows, yields improved generalization, and allows for larger simulation time\nsteps.", "category": "physics_flu-dyn" }, { "text": "Elastic turbulence homogenizes fluid transport in stratified porous\n media: Many key environmental, industrial, and energy processes rely on controlling\nfluid transport within subsurface porous media. These media are typically\nstructurally heterogeneous, often with vertically-layered strata of distinct\npermeabilities -- leading to uneven partitioning of flow across strata, which\ncan be undesirable. Here, using direct in situ visualization, we demonstrate\nthat polymer additives can homogenize this flow by inducing a purely-elastic\nflow instability that generates random spatiotemporal fluctuations and excess\nflow resistance in individual strata. In particular, we find that this\ninstability arises at smaller imposed flow rates in higher-permeability strata,\ndiverting flow towards lower-permeability strata and helping to homogenize the\nflow. Guided by the experiments, we develop a parallel-resistor model that\nquantitatively predicts the flow rate at which this homogenization is optimized\nfor a given stratified medium. Thus, our work provides a new approach to\nhomogenizing fluid and passive scalar transport in heterogeneous porous media.", "category": "physics_flu-dyn" }, { "text": "A critical examination of the statistical symmetries admitted by the\n Lundgren-Monin-Novikov hierarchy of unconfined turbulence: We present a critical examination of the recent article by Waclawczyk et al.\n(2014) which proposes two new statistical symmetries in the classical theory\nfor turbulent hydrodynamic flows. We first show that both symmetries are\nunphysical in that they induce inconsistencies due to violating the principle\nof causality. In addition, they must get broken in order to be consistent with\nall physical constraints naturally arising in the statistical\nLundgren-Monin-Novikov (LMN) description of turbulence. As a result, we state\nthat besides the well-known classical symmetries of the LMN equations no new\nstatistical symmetries exist. Yet, aside from this particular issue, the\narticle by Waclawczyk et al. (2014) is flawed in more than one respect, ranging\nfrom an incomplete proof, to a self-contradicting statement up to an incorrect\nclaim. All these aspects will be listed, discussed and corrected, thus\nobtaining a completely opposite conclusion in our study than the article by\nWaclawczyk et al. (2014) is proposing.", "category": "physics_flu-dyn" }, { "text": "Buoyancy-driven attraction of active droplets: Active oil droplets in a liquid are believed to repel due to the Marangoni\neffect, while buoyancy effects caused by the density difference between the\ndroplets, diffusing product, and ambient fluid are usually overlooked. Recent\nexperiments have observed active droplet clustering phenomena due to\nbuoyancy-driven convection (Kruger et al. Eur. Phys. J. E, vol. 39, 2016,\npp.1-9). In this study, we numerically analyze the buoyancy effect in addition\nto Marangoni flow, characterized by Peclet number $Pe$. The buoyancy effects\noriginate from (i) the density difference between the droplet and the ambient\nliquid, which is characterized by Galileo number $Ga$, and (ii) the density\ndifference between the diffusing product (i.e. filled micelles) and the ambient\nliquid, characterized by a solutal Rayleigh number $Ra$. We analyze how the\nattracting and repulsing behavior depends on the control parameters $Pe$, $Ga$,\nand $Ra$. We find that while Marangoni flow causes repulsion, the buoyancy\neffect leads to attraction, and even collisions can take place at high Ra. We\nalso observe a delayed collision as $Ga$ increases. Moreover, we derive that\nthe attracting velocity, characterized by a Reynolds number $Re_d$, is\nproportional to $Ra^{1/4}/(l/R)$, where $l/R$ is the normalized distance by\nradius between neighboring droplets. Finally, we obtain repulsive velocity,\ncharacterized by $Re_{rep}$, as proportional to $PeRa^{-0.38}$. The balance of\nattractive and repulsive effects results in $Pe \\sim Ra^{0.63}$, which agrees\nwith the transition curve between regimes with and without collision.", "category": "physics_flu-dyn" }, { "text": "Pore-scale Mixing and the Evolution of Hydrodynamic Dispersion in Porous\n Media: We study the interplay of pore-scale mixing and network-scale advection\nthrough heterogeneous porous media, and its role for the evolution and\nasymptotic behavior of hydrodynamic dispersion. In a Lagrangian framework, we\nidentify three fundamental mechanisms of pore-scale mixing that determine large\nscale particle motion, namely, the smoothing of intra-pore velocity contrasts,\nthe increase of the tortuosity of particle paths, and the setting of a maximum\ntime for particle transitions. Based on these mechanisms, we derive a theory\nthat predicts anomalous and normal hydrodynamic dispersion based on the\ncharacteristic pore length, Eulerian velocity distribution and P\\'eclet number.", "category": "physics_flu-dyn" }, { "text": "Optimal sensor placement for reconstructing wind pressure field around\n buildings using compressed sensing: Deciding how to optimally deploy sensors in a large, complex, and spatially\nextended structure is critical to ensure that the surface pressure field is\naccurately captured for subsequent analysis and design. In some cases,\nreconstruction of missing data is required in downstream tasks such as the\ndevelopment of digital twins. This paper presents a data-driven sparse sensor\nselection algorithm, aiming to provide the most information contents for\nreconstructing aerodynamic characteristics of wind pressures over tall building\nstructures parsimoniously. The algorithm first fits a set of basis functions to\nthe training data, then applies a computationally efficient QR algorithm that\nranks existing pressure sensors in order of importance based on the state\nreconstruction to this tailored basis. The findings of this study show that the\nproposed algorithm successfully reconstructs the aerodynamic characteristics of\ntall buildings from sparse measurement locations, generating stable and optimal\nsolutions across a range of conditions. As a result, this study serves as a\npromising first step toward leveraging the success of data-driven and machine\nlearning algorithms to supplement traditional genetic algorithms currently used\nin wind engineering.", "category": "physics_flu-dyn" }, { "text": "Direct and inverse pumping in flows with homogeneous and non-homogeneous\n swirl: The conditions in which meridional recirculations appear in swirling flows\nabove a fixed wall are analysed. In the classical Bodew\\\"adt problem, where the\nswirl tends towards an aysmptotic value away from the wall, the well-known\n\"tea-cup effect\" drives a flow away from the plate at the centre of the vortex.\nSimple dimensional arguments applied to a single vortex show that if the\nintensity of the swirl decreases away from the wall, the sense of the\nrecirculation can be inverted, and that the associated flow rate scales with\nthe swirl gradient. Only if the flow is quasi-2D, does the classical tea-cup\neffect take place. This basic theory is confirmed by numerical simulations of a\nsquare array of steady, electrically driven vortices. Experiments in the\nturbulent regimes of the same configuration reveal that these mechanisms are\nactive in the average flow and in its fluctuating part. The mechanisms singled\nout in this letter provide an explanation for previously observed phenomena in\nelectrolyte flows. They also put forward a possible mechanism for the\ngeneration of helicity in flows close to two-dimensionality, which plays a key\nrole in the transition between 2D and 3D turbulence.", "category": "physics_flu-dyn" }, { "text": "Validity of Sound-Proof Approaches in Rapidly-Rotating Compressible\n Convection: Marginal Stability vs. Turbulence: The validity of the anelastic approximation has recently been questioned in\nthe regime of rapidly-rotating compressible convection in low Prandtl number\nfluids (Calkins et al. 2015). Given the broad usage and the high computational\nefficiency of sound-proof approaches in this astrophysically relevant regime,\nthis paper clarifies the conditions for a safe application. The potential of\nthe alternative pseudo-incompressible approximation is investigated, which in\ncontrast to the anelastic approximation is shown to never break down for\npredicting the point of marginal stability. Its accuracy, however, decreases\nclose to the parameters corresponding to the failure of the anelastic approach,\nwhich is shown to occur when the sound-crossing time of the domain exceeds a\nrotation time scale, i.e. for rotational Mach numbers greater than one.\nConcerning the supercritical case, which is naturally characterised by smaller\nrotational Mach numbers, we find that the anelastic approximation does not show\nunphysical behaviour. Growth rates computed with the linearised anelastic\nequations converge toward the corresponding fully compressible values as the\nRayleigh number increases. Likewise, our fully nonlinear turbulent simulations,\nproduced with our fully compressible and anelastic models and carried out in a\nhighly supercritical, rotating, compressible, low Prandtl number regime show\ngood agreement. However, this nonlinear test example is for only a moderately\nlow convective Rossby number of 0.14.", "category": "physics_flu-dyn" }, { "text": "Understanding how porosity gradients can make a better filter using\n homogenization theory: Filters whose porosity decreases with depth are often more efficient at\nremoving solute from a fluid than filters with a uniform porosity. We\ninvestigate this phenomenon via an extension of homogenization theory that\naccounts for a macroscale variation in microstructure. In the first stage of\nthe paper, we homogenize the problems of flow through a filter with a\nnear-periodic microstructure and of solute transport due to advection,\ndiffusion, and filter adsorption. In the second stage, we use the\ncomputationally efficient homogenized equations to investigate and quantify why\nporosity gradients can improve filter efficiency. We find that a porosity\ngradient has a much larger effect on the uniformity of adsorption than it does\non the total adsorption. This allows us to understand how a decreasing porosity\ncan lead to a greater filter efficiency, by lowering the risk of localized\nblocking while maintaining the rate of total contaminant removal.", "category": "physics_flu-dyn" }, { "text": "Dynamics of pulsatile flows through elastic microtubes: We investigate the dynamics of pressure driven transient flows of\nincompressible Newtonian fluids through circular microtubes having thin elastic\nwalls under the long-wavelength and small deformation assumptions, which are\nvalid for many industrial and biological processes. An analytical solution of\nthe coupled fluid and solid equations is found using Navier slip boundary\nconditions and is shown to include some existing Womersley solutions as\nlimiting cases. The effect of the slip length at the fluid-solid interface is\nanalyzed for oscillatory pressure gradients using a range of slip ratio and\nfrequency parameters. The solutions for elastic and rigid walls are compared\nfor the cases with and without slip boundary conditions for a broad range of\nthe relevant parameters. It is shown that the elastic behavior of the microtube\ncouples nonlinearly with the slip velocity, which greatly enhances the\nachievable flow rate and pumping efficacy compared to the inelastic case. In\naddition, it is observed that increasing the slip length produces less shear\nstress, which is consistent with the nearly frictionless interfaces observed in\nmany microscale experiments.", "category": "physics_flu-dyn" }, { "text": "Multiscale analysis of the topological invariants in the logarithmic\n region of turbulent channels at $Re_\u03c4=932$: The invariants of the velocity gradient tensor, $R$ and $Q$, and their\nenstrophy and strain components are studied in the logarithmic layer of an\nincompressible turbulent channel flow. The velocities are filtered in the three\nspatial directions and the results analyzed at different scales. We show that\nthe $R$--$Q$ plane does not capture the changes undergone by the flow as the\nfilter width increases, and that the enstrophy/enstrophy-production and\nstrain/strain-production planes represent better choices. We also show that the\nconditional mean trajectories may differ significantly from the instantaneous\nbehavior of the flow since they are the result of an averaging process where\nthe mean is 3-5 times smaller than the corresponding standard deviation. The\norbital periods in the $R$--$Q$ plane are shown to be independent of the\nintensity of the events, and of the same order of magnitude than those in the\nenstrophy/enstrophy-production and strain/strain-production planes. Our final\ngoal is to test whether the dynamics of the flow are self-similar in the\ninertial range, and the answer turns out to be no. The mean shear is found to\nbe responsible for the absence of self-similarity and progressively controls\nthe dynamics of the eddies observed as the filter width increases. However, a\nself-similar behavior emerges when the calculations are repeated for the\nfluctuating velocity gradient tensor. Finally, the turbulent cascade in terms\nof vortex stretching is considered by computing the alignment of the vorticity\nat a given scale with the strain at a different one. These results generally\nsupport a non-negligible role of the phenomenological energy-cascade model\nformulated in terms of vortex stretching.", "category": "physics_flu-dyn" }, { "text": "Methods for panoramic visualization and digital analysis of\n thermophysical flow fields. A review: The paper presents a review of modern methods in the registration,\nprocessing, and analysis of dynamic processes in liquids, gases, plasmas,\nmultiphase media, which are realized in researches and technology. The review\nauthor observes both the physical fundamentals of flow visualization and the\nbasics of modern technologies for the digital processing of recorded flow\nimages. A brief analysis of the panoramic visualization methods progress\nhistory covers a period of one and a half centuries. In the works of the last\ndecade, the focus is on the methods of computer processing, tools, technologies\nfor analyzing and recognizing panoramic thermophysical fields, which make it\npossible to obtain quantitative information about flows. The review contains an\nanalysis of publications describing the main modern methods of visualization of\nflows: methods based on the phenomenon of refraction; electroluminescence;\ndigital tracing (particle image velocimetry), visualization of surface flows\n(PSP and TSP coatings, liquid crystals; oil coatings). Particular attention is\npaid to methods using cross-correlation image processing algorithms. Those are:\ndigital tracing (PIV), shadow background method (BOS), seedless shadow methods,\nthermographic PIV, velocity measurement in viscous coatings, micro, tomographic\nmodifications of PIV, etc. The actual problem of digital data analyzing in a\npanoramic experiment is touched upon - the problem of big data. Examples of\nmachine learning use in the analysis of big data sets of shadow surveys are\ngiven. Some examples of numerical simulation data visualization (simulation of\nexperimental flowfields) are considered.", "category": "physics_flu-dyn" }, { "text": "Accuracy and stability analysis of horizontal discretizations used in\n unstructured grid ocean models: One important tool at our disposal to evaluate the robustness of Global\nCirculation Models (GCMs) is to understand the horizontal discretization of the\ndynamical core under a shallow water approximation. Here, we evaluate the\naccuracy and stability of different methods used in, or adequate for,\nunstructured ocean models considering shallow water models. Our results show\nthat the schemes have different accuracy capabilities, with the A- (NICAM) and\nB-grid (FeSOM 2.0) schemes providing at least 1st order accuracy in most\noperators and time integrated variables, while the two C-grid (ICON and MPAS)\nschemes display more difficulty in adequately approximating the horizontal\ndynamics. Moreover, the theory of the inertia-gravity wave representation on\nregular grids can be extended for our unstructured based schemes, where from\nleast to most accurate we have: A-, B, and C-grid, respectively. Considering\nonly C-grid schemes, the MPAS scheme has shown a more accurate representation\nof inertia-gravity waves than ICON. In terms of stability, we see that both A-\nand C-grid MPAS scheme display the best stability properties, but the A-grid\nscheme relies on artificial diffusion, while the C-grid scheme doesn't.\nAlongside, the B-grid and C-grid ICON schemes are within the least stable.\nFinally, in an effort to understand the effects of potential instabilities in\nICON, we note that the full 3D model without a filtering term does not\ndestabilize as it is integrated in time. However, spurious oscillations are\nresponsible for decreasing the kinetic energy of the oceanic currents.\nFurthermore, an additional decrease of the currents' turbulent kinetic energy\nis also observed, creating a spurious mixing, which also plays a role in the\nstrength decrease of these oceanic currents.", "category": "physics_flu-dyn" }, { "text": "Self-similar solution of a supercritical two-phase laminar mixing layer: Previous works for a liquid suddenly contacting a gas at a supercritical\npressure show the coexistence of both phases and the generation of diffusion\nlayers on both sides of the liquid-gas interface due to thermodynamic phase\nequilibrium. A related numerical study of a laminar mixing layer between the\nliquid and gas streams in the near field of the splitter plate suggests that\nmass, momentum and thermal diffusion layers evolve in a self-similar manner at\nvery high pressures. In this paper, the high-pressure, two-phase, laminar\nmixing-layer equations are recast in terms of a similarity variable. A liquid\nhydrocarbon and gaseous oxygen are considered. Freestream conditions and proper\nmatching conditions at the liquid-gas interface are applied. To solve the\nsystem of equations, a real-fluid thermodynamic model based on the\nSoave-Redlich-Kwong equation of state is selected. A comparison with results\nobtained by directly solving the laminar mixing-layer equations shows the\nvalidity of the similarity approach applied to non-ideal two-phase flows. Even\nwhen the gas is hotter than the liquid, condensation can occur at high\npressures while heat conducts into the liquid. Finally, a generalized\ncorrelation is proposed to represent the evolution of the mixing layer\nthickness for different problem setups.", "category": "physics_flu-dyn" }, { "text": "Challenges in Modelling and Simulation of Turbulent Flows with\n Spatially-Developing Coherent Structures: The objective of this work is to investigate the challenges encountered in\nScale-Resolving Simulations (SRS's) of turbulent wake flows driven by\nspatially-developing coherent structures. SRS's of practical interest are\nexpressly intended for efficiently computing such flows by resolving only the\nmost important features of the coherent structures and modelling the remainder\nas stochastic field. The success of SRS methods depends upon three important\nfactors: i) ability to identify key flow mechanisms responsible for the\ngeneration of coherent structures; ii) determine the optimum range of\nresolution required to adequately capture key elements of coherent structures;\nand iii) ensure that the modelled part is comprised nearly exclusively of\nfully-developed stochastic turbulence. This study considers the canonical case\nof the flow around a circular cylinder to address these three key issues. It is\nfirst demonstrated using experimental evidence that the vortex-shedding\ninstability and flow-structure development involves four important stages. A\nseries of SRS computations of progressively increasing resolution are\nperformed. An a priori basis for locating the origin of the coherent structures\ndevelopment is proposed and examined. The criterion is based on the fact that\nthe coherent structures are generated by the Kelvin-Helmholtz (KH) instability.\nThe most important finding is that the key aspects of coherent structures can\nbe resolved only if the effective computational Reynolds number exceeds the\ncritical value of the KH instability in laminar flows. Finally, a quantitative\ncriterion assessing the nature of the unresolved field based on the strain-rate\nratio of mean and unresolved fields is examined. The two proposed conditions\nand underlying rationale offer a quantitative basis for develop \"good practice\"\nguidelines for SRS of complex turbulent wake flows with coherent structures.", "category": "physics_flu-dyn" }, { "text": "Spatial deterministic wave forecasting for nonlinear sea-states: We derive a simple algebraic form of the nonlinear wavenumber correction of\nsurface gravity waves in deep water, based on temporal measurements of the\nwater surface and the spatial Zakharov equation. This allows us to formulate an\nimprovement over linear deterministic wave forecasting with no additional\ncomputational cost. Our new formulation is used to forecast both synthetically\ngenerated as well as experimentally measured seas, and shows marked\nimprovements over the linear theory.", "category": "physics_flu-dyn" }, { "text": "Ultra-fast escape maneuver of an octopus-inspired robot: We design and test an octopus-inspired flexible hull robot that demonstrates\noutstanding fast-starting performance. The robot is hyper-inflated with water,\nand then rapidly deflates to expel the fluid so as to power the escape\nmaneuver. Using this robot we verify for the first time in laboratory testing\nthat rapid size-change can substantially reduce separation in bluff bodies\ntraveling several body lengths, and recover fluid energy which can be employed\nto improve the propulsive performance. The robot is found to experience speeds\nover ten body lengths per second, exceeding that of a similarly propelled\noptimally streamlined rigid rocket. The peak net thrust force on the robot is\nmore than 2.6 times that on an optimal rigid body performing the same maneuver,\nexperimentally demonstrating large energy recovery and enabling acceleration\ngreater than 14 body lengths per second squared. Finally, over 53% of the\navailable energy is converted into payload kinetic energy, a performance that\nexceeds the estimated energy conversion efficiency of fast-starting fish. The\nReynolds number based on final speed and robot length is $Re \\approx 700,000$.\nWe use the experimental data to establish a fundamental deflation scaling\nparameter $\\sigma^*$ which characterizes the mechanisms of flow control via\nshape change. Based on this scaling parameter, we find that the fast-starting\nperformance improves with increasing size.", "category": "physics_flu-dyn" }, { "text": "Temporally sparse data assimilation for the small-scale reconstruction\n of turbulence: Previous works have shown that the small-scale information of incompressible\nhomogeneous isotropic turbulence (HIT) is fully recoverable as long as\nsufficient large-scale structures are continuously enforced through temporally\ncontinuous data assimilation (TCDA). In the current work, we show that the\nassimilation time step can be relaxed to values about 1 $\\sim$ 2 orders larger\nthan that for TCDA, using a temporally sparse data assimilation (TSDA)\nstrategy, while the accuracy is still maintained or even slightly better in the\npresence of non-negligible large-scale errors. The one-step data assimilation\n(ODA) is examined to unravel the mechanism of TSDA. It is shown that the\nrelaxation effect for errors above the assimilation wavenumber $k_a$ is\nresponsible for the error decay in ODA. Meanwhile, The errors contained in the\nlarge scales can propagate into small scales and make the high-wavenumber\n($k>k_a$) error noise decay slower with TCDA than TSDA. This mechanism is\nfurther confirmed by incorporating different levels of errors in the large\nscales of the reference flow field. The advantage of TSDA is found to grow with\nthe magnitude of the incorporated errors. Thus, it is potentially more\nbeneficial to adopt TSDA if the reference data contains non-negligible errors.\nFinally, an outstanding issue raised in previous works regarding the\npossibility of recovering the dynamics of sub-Kolmogorov scales using direct\nnumerical simulation (DNS) data at Kolmogorov scale resolution is also\ndiscussed.", "category": "physics_flu-dyn" }, { "text": "Influence of atmospheric conditions on the power production of\n utility-scale wind turbines in yaw misalignment: The intentional yaw misalignment of leading, upwind turbines in a wind farm,\ntermed wake steering, has demonstrated potential as a collective control\napproach for wind farm power maximization. The optimal control strategy, and\nresulting effect of wake steering on wind farm power production, are in part\ndictated by the power degradation of the upwind yaw misaligned wind turbines.\nIn the atmospheric boundary layer, the wind speed and direction may vary\nsignificantly over the wind turbine rotor area, depending on atmospheric\nconditions and stability, resulting in freestream turbine power production\nwhich is asymmetric as a function of the direction of yaw misalignment and\nwhich varies during the diurnal cycle. In this study, we propose a model for\nthe power production of a wind turbine in yaw misalignment based on aerodynamic\nblade elements which incorporates the effects of wind speed and direction\nchanges over the turbine rotor area in yaw misalignment. A field experiment is\nperformed using multiple utility-scale wind turbines to characterize the power\nproduction of yawed freestream operating turbines depending on the wind\nconditions, and the model is validated using the experimental data. The\nresulting power production of a yaw misaligned variable speed wind turbine\ndepends on a nonlinear interaction between the yaw misalignment, the\natmospheric conditions, and the wind turbine control system.", "category": "physics_flu-dyn" }, { "text": "Multiple droplets on a conical fiber: formation, motion, and droplet\n mergers: Small droplets on slender conical fibers spontaneously move along the fiber\ndue to capillary action. The droplet motion depends on the geometry of the\ncone, the surface wettability, the surface tension, the viscosity, and the\ndroplet size. Here we study with experiments and numerical simulations, the\nformation, spontaneous motion, and the eventual merger, of multiple droplets on\nslender conical fibers as they interact with each other. The droplet size and\ntheir spacing on the fibre is controlled by the Plateau-Rayleigh instability\nafter dip-coating the conical fiber. Once these droplets are formed on the\nfiber, they spontaneously start to move. Since droplets of different size move\nwith different speeds, they effectively coarsen the droplet patterning by\nmerging on the fiber. The droplet merging process affects locally the droplet\nspeed and alters the spatiotemporal film deposition on the fiber.", "category": "physics_flu-dyn" }, { "text": "A unified slip boundary condition for flow over a surface: Interface between two phases of matter are ubiquitous in nature and\ntechnology. Determining the correct velocity condition at an interface is\nessential for understanding and designing of flows over a surface. We\ndemonstrate that both the widely used no-slip and the Navier and Maxwell slip\nboundary conditions do not capture the complete physics associated with complex\nproblems, such as spreading of liquids or corner flows. Hence, we present a\nunified boundary condition that is applicable to a wide-range of flow problems.", "category": "physics_flu-dyn" }, { "text": "A stochastic view of isotropic turbulence decay: A stochastic EDQNM approach is used to investigate self-similar decaying\nisotropic turbulence at high Reynolds number ($400 \\leq Re_\\lambda \\leq 10^4$).\nThe realistic energy spectrum functional form recently proposed by Meyers &\nMeneveau is generalised by considering some of the model constants as random\nparameters, since they escape measure in most experimental set-ups. The induced\nuncertainty on the solution is investigated building response surfaces for\ndecay power-law exponents of usual physical quantities. Large-scale\nuncertainties are considered, the emphasis being put on Saffman and Batchelor\nturbulence. The sensitivity of the solution to initial spectrum uncertainties\nis quantified through probability density functions of the decay exponents. It\nis observed that initial spectrum shape at very large scales governs the\nlong-time evolution, even at high Reynolds number, a parameter which is not\nexplicitly taken into account in many theoretical works. Therefore, a universal\nasymptotic behavior in which kinetic energy decays as $t^{-1}$ is not detected.\nBut this decay law is observed at finite Reynolds number with low probability\nfor some initial conditions.", "category": "physics_flu-dyn" }, { "text": "Kolmogorov Turbulence Coexists with Pseudo-Turbulence in Buoyancy-Driven\n Bubbly Flows: We investigate spectral properties of buoyancy driven bubbly flows. Using\nhigh-resolution numerical simulations and phenomenology of homogeneous\nturbulence, we identify the relevant energy transfer mechanisms. We find: (a)\nAt high enough Galilei number (ratio of the buoyancy to viscous forces) the\nkinetic energy spectrum shows the Kolmogorov scaling with a power law exponent\n$-5/3$ for the range of scales between the bubble diameter and the dissipation\nscale ($\\eta$). (b) For scales smaller than $\\eta$, the physics of\npseudo-turbulence is recovered.", "category": "physics_flu-dyn" }, { "text": "Smoothed Particle Hydrodynamics Physically Reconsidered -- The Relation\n to Explicit Large Eddy Simulation and the Issue of Particle Duality: In this work we will identify a novel relation between Smoothed Particle\nHydrodynamics (SPH) and explicit Large Eddy Simulation (LES) using a\ncoarse-graining method from Non-Equilibrium Molecular Dynamics (NEMD). While\nthe current literature points at the conclusion that characteristic SPH issues\nbecome restrictive for subsonic turbulent flows, we see the potential to\nmitigate these SPH issues by explicit subfilter stress (SFS) modelling. We\nverify our theory by various simulations of homogeneous, isotropic turbulence\n(HIT) at $Re=10^4$ and compare the results to a Direct Numerical Simulation\n(DNS) reported by Dairay et al. (2017). Although the simulations substantiate\nour theory, we see another issue arising, which is conceptually rooted in the\nparticle itself, termed as Particle Duality. Finally, we conclude our work by\nacknowledging SPH as coarse-graining method for turbulent flows, highlighting\nits capabilities and limitations.", "category": "physics_flu-dyn" }, { "text": "Turbulent boundary layer response to uniform changes of the pressure\n force contribution: We investigate a turbulent boundary layer (TBL) with uniform pressure force\nvariations, focusing on understanding its response to local pressure force,\nlocal pressure force variation (local disequilibrating effect), and upstream\nhistory. The studied flow starts as a zero-pressure-gradient (ZPG) TBL,\nfollowed by a uniform increase in the ratio of pressure force to turbulent\nforce in the outer region and concludes with a uniform decrease of the same\nmagnitude. The second zone includes a subzone with a diminishing\nadverse-pressure-gradient (APG), followed by an increasing\nfavorable-pressure-gradient (FPG). In both subzones, the impact remains the\nsame: mean momentum gain in the boundary layer and reduced turbulence. In the\nouter region, the mean flow responds to force balance changes with a\nconsiderable delay. The accumulated flow history effects lead to a FPG TBL at\nthe domain's end with a momentum defect comparable to APG TBLs. Below $y^+=10$,\nthe mean flow responds almost instantaneously to pressure force changes. In the\noverlap layer, the profiles deviate from the conventional logarithmic law of\nthe ZPG TBL. Regarding the outer-layer turbulence, its subsequent decay is\nslower than its initial increase, the latter persisting even after the pressure\nforce begins to decrease. As a result of the slow turbulence decay, the FPG TBL\nat the domain's end exhibits unusually high outer turbulence levels. Near the\nwall, turbulence responds with a delay to changes in the pressure force partly\ndue to large-scale turbulence influence. This study underscores the complexity\nof the triple action of the pressure force.", "category": "physics_flu-dyn" }, { "text": "A mesh-free framework for high-order direct numerical simulations of\n combustion in complex geometries: The multiscale nature of turbulent combustion necessitates accurate and\ncomputationally efficient methods for direct numerical simulations (DNS). The\nfield has long been dominated by high-order finite differences, which lack the\nflexibility and adaptivity for simulations of complex geometries and\nflame-turbulence-structure interactions in realistic settings. In this work we\nintroduce a new approach to DNS of premixed combustion, based on a high-order\nmesh-free discretisation in combination with finite differences, enabling\nhigh-order simulations in non-trivial geometries. The approach is validated\nagainst a range of two- and three-dimensional flows, both laminar and\nturbulent, and reacting and inert. The present method a) has the resolving\npower for DNS of both laminar flames and inert turbulence with comparable\naccuracy to high-order finite differences, b) can capture the dynamics of\nunsteady bluff body stabilised flames, and c) is capable of simulating\nflame-turbulence interactions, with results comparing qualitatively well with\npublished data. This work paves the way for DNS of combustion in complex\ngeometries, offering an alternative approach to methods based on structured\ngrids with immersed boundaries, or unstructured meshes. Further studies with\nthe present method are proposed, which will aid understanding of fundamental\nflame dynamics in non-trivial geometries. Planned developments in adaptivity\nand extension of the mesh-free construction to all three dimensions will\nincrease the value of the method, and support the push towards DNS of real\ngeometries.", "category": "physics_flu-dyn" }, { "text": "On the shock wave boundary layer interaction in slightly-rarefied gas: The shock wave and boundary layer interaction (SWBLI) plays an important role\nin the design of hypersonic vehicles. However, discrepancies between the\nnumerical results of high-temperature gas dynamics and experiment data have not\nbeen fully addressed. It is believed that the rarefaction effects are important\nin SWBLI, but the systematic analysis of the temperature-jump boundary\nconditions and the role of translational/rotational/vibrational heat\nconductivities are lacking. In this paper, we derive the three-temperature\nNavier-Stokes-Fourier (NSF) equations from the gas kinetic theory, with special\nattention paid to the components of heat conductivity. With proper\ntemperature-jump boundary conditions, we simulate the SWBLI in the double cone\nexperiment. Our numerical results show that, when the three heat conductivities\nare properly recovered, the NSF equations can capture the position and peak\nvalue of the surface heat flux, in both low- and high-enthalpy inflow\nconditions. Moreover, the separation bubble induced by the separated shock and\nthe reattachment point induced by impact between transmitted shock and boundary\nlayer are found to agree with the experimental measurement.", "category": "physics_flu-dyn" }, { "text": "Breakup of oil slicks on wavy water surfaces: We hypothesize that the spread of oil slicks on the water's surface during\noil spills is significantly influenced by water wave motion at the initial or\nintermediate spreading stages, well before emulsification processes have a\nsubstantial impact on the oil film's state. We demonstrate that the spreading\ndynamics of an oil slick on the water surface are facilitated by water waves,\nemploying the thin film approximation. It is shown that water wave motion can\nrapidly deplete any oil slick, reducing the oil layer's thickness to nearly\nzero. This mechanism may act as a precursor to emulsification processes,\nleading to the accelerated depletion of oil spills into a distribution of\ndroplets that form an emulsion.", "category": "physics_flu-dyn" }, { "text": "The three-dimensional instabilities and destruction of the Hill's vortex: The Hill vortex is a three-dimensional vortex structure solution of the Euler\nequations. For small amplitude axisymmetric disturbances on the external\nsurface from the linear stability analysis by \\citet{moff78} emerged the\nformation of a tail. Successive studies confirmed this fact, and in addition\nobserved a different shape of the tail with azimuthal small amplitude\ndisturbances. In this paper the Navier-Stokes equations are solved at high\nvalues of the Reynolds number imposing large amplitude axisymmetric and\nthree-dimensional disturbances on the surface of the vortex. The conclusion is\nthat the azimuthal disturbances produce a hierarchy of structures inside the\nvortex maintaining the shape of the vortex. On the other hand the axisymmetric\ndisturbances are convected in the rear side, are dumped and form an\naxisymmetric wave longer higher the magnitude of the surface disturbance.\nSimulations in a moving frame allow to extend the evolution in time leading to\nthe conclusion that the Hill vortex is unstable and produces a wide range of\nenergy containing scales characteristic of three-dimensional flows.", "category": "physics_flu-dyn" }, { "text": "Generation and application of sub-kilohertz oscillatory flows in\n microchannels: We present a user-friendly and versatile experimental technique that\ngenerates sub-kilohertz sinusoidal oscillatory flows within microchannels. The\nmethod involves the direct interfacing of microfluidic tubing with a\nloudspeaker diaphragm to generate oscillatory flow in microchannels with\nfrequencies ranging from $10-1000$ Hz and amplitudes ranging from $10 - 600 \\\n\\mu$m. The speaker-based apparatus allows independent control of frequency and\namplitude that is unique to the speaker's manufacturing specifications. The\nperformance of our technique is evaluated by Fourier spectral analysis of\noscillatory motion of tracer particles, obtained by particle tracking\nvelocimetry, as well as by comparing oscillatory flow profiles against\ntheoretical benchmarks such as Stokes flow in a square channel and Stokes'\nsecond problem near a solid boundary. Applications that utilize both the\noscillatory flow and the associated steady rectified flows are demonstrated in\nprototypical microfluidic configurations. These include inertial focusing and\nmixing at low Reynolds numbers, respectively.", "category": "physics_flu-dyn" }, { "text": "Turbulent pair dispersion of inertial particles: The relative dispersion of pairs of inertial particles in incompressible,\nhomogeneous, and isotropic turbulence is studied by means of direct numerical\nsimulations at two values of the Taylor-scale Reynolds number $Re_{\\lambda}\n\\sim 200$ and 400. The evolution of both heavy and light particle pairs is\nanalysed at varying the particle Stokes number and the fluid-to-particle\ndensity ratio. For heavy particles, it is found that turbulent dispersion is\nschematically governed by two temporal regimes. The first is dominated by the\npresence, at large Stokes numbers, of small-scale caustics in the particle\nvelocity statistics, and it lasts until heavy particle velocities have relaxed\ntowards the underlying flow velocities. At such large scales, a second regime\nstarts where heavy particles separate as tracers particles would do. As a\nconsequence, at increasing inertia, a larger transient stage is observed, and\nthe Richardson diffusion of simple tracers is recovered only at large times and\nlarge scales. These features also arise from a statistical closure of the\nequation of motion for heavy particle separation that is proposed, and which is\nsupported by the numerical results. In the case of light particles with high\ndensity ratios, strong small-scale clustering leads to a considerable fraction\nof pairs that do not separate at all, although the mean separation increases\nwith time. This effect strongly alters the shape of the probability density\nfunction of light particle separations.", "category": "physics_flu-dyn" }, { "text": "Stability of Hartmann flow with the convective approximation: This research is focused on linear analysis of a plane-parallel flow\nstability in a transverse magnetic field (Hartmann flow) within a convective\napproximation. We derive and solve equations describing the perturbation\ngrowth. Perturbation modes and their nonexcitation conditions have been\ndetermined. An equation for the instability increment has been derived and it\nis shown that the equation has an unstable root. Additionally, we show that the\nresulting instabilities qualitatively agree with the experimental data.", "category": "physics_flu-dyn" }, { "text": "Folding of the frozen-in-fluid di-vorticity field in two-dimensional\n hydrodynamic turbulence: The vorticity rotor field ${\\bf B}=\\mbox{rot}\\,\\mathbf{\\omega}$\n(di-vorticity) for freely decaying two-dimensional hydrodynamic turbulence due\nto a tendency to breaking is concentrated in the vicinity of the lines\ncorresponding to the position of the vorticity quasi-shocks. The maximum value\nof the di-vorticity $B_{max}$ at the stage of quasi-shocks formation increases\nexponentially in time, while the thickness $\\ell(t)$ of the maximum area in the\ntransverse direction to the vector ${\\bf B}$ decreases in time also\nexponentially. It is numerically shown that $B_{max} (t)$ depends on the\nthickness according to the power law: $B_{max}(t)\\sim \\ell^{-\\alpha}(t)$, where\nthe exponent $\\alpha\\approx 2/3$. This behavior indicates in favor of folding\nfor the divergence-free vector field of the di-vorticity.", "category": "physics_flu-dyn" }, { "text": "The effect of a flight stream on subsonic turbulent jets: This study concerns a turbulent jet at Mach number $M_j=0.9$, subject to a\nuniform external flow stream at $M_f=0.15$. We assess the mechanisms that\nunderpin the reduction in fluctuation energy that is known to occur when a jet\nis surrounded by a flight stream. The analysis combines experimental and\nnumerical databases, spectral proper orthogonal decomposition (SPOD) and linear\nmodelling. The experiments involve Time-Resolved, Stereo PIV measurements at\ndifferent cross-sections of the jet. A companion large-eddy simulation was\nperformed with the same operating conditions using the \"CharLES\" solver by\nCascade Technologies in order to obtain a complete and highly resolved 3D\ndatabase. We show that the energy reduction is spread over a broad region of\nthe frequency-wavenumber space and involves, apart from the known stabilization\nof the modal Kelvin-Helmholtz (KH) instability, the attenuation of flow\nstructures associated with the non-modal Orr and lift-up mechanisms. Streaky\nstructures, associated with helical azimuthal wavenumbers and very slow time\nscales, are the most strongly affected by the flight stream, in terms of energy\nattenuation and spatial distortion. The energy reductions are accompanied by a\nweakening of the low-rank behaviour of the jet dynamics revealed by previous\nstudies. These trends are found to be consistent with results of a local linear\nmodel based on the modified mean flow in the flight stream case.", "category": "physics_flu-dyn" }, { "text": "Flow cross-overs under surface fluctuations in cylindrical nano-channel: We analyse surface-fluctuations-driven fluid flow through nano-channels to\ndifferentiate boundary layer flow structures from the bulk flow of fluid under\na pressure-head. Surface fluctuations of a wide range of frequencies (up to\nseveral thousands of Hertz) in a nano-channel keep the flow in the low Reynolds\nnumber regime. Using this advantage of low Reynolds number flow, we develop a\nperturbation analysis of the fluid flow that clearly distinguishes the bulk\nflow under a pressure head around the axis of a nano-tube from its surface flow\nstructure induced by fluctuations. In terms of particle transport under such\nflow conditions, there exists the opportunity to drag particles near the\nperiphery of the nano-tube in a direction opposite to the bulk flow near the\naxis. This can potentially find applications in the separation, trapping, and\nfiltration of particles under surface-driven flow through nano-tubes under\nwidely varying conditions.", "category": "physics_flu-dyn" }, { "text": "A consistent co-rotational formulation for aerodynamic nonlinear\n analysis of flexible frame structures: The design of structures submitted to aerodynamic loads usually requires the\ndevelopment of specific computational models considering fluid-structure\ninteractions. Models using structural frame elements are developed in several\nrelevant applications such as the design of advanced aircraft wings, wind\nturbine blades or power transmission lines. In the case of flexible frame\nstructures submitted to fluid flows, the consistent computation of inertial and\naerodynamic forces for large displacements is a challenging task. In this\narticle we present a novel formulation for the accurate computation of\naerodynamic forces for large displacements and rotations using the\nco-rotational approach, the quasi-steady theory and the principle of virtual\nwork. This formulation is coupled with a reference consistent co-rotational\nformulation for computing internal and inertial forces, providing a unified set\nof nonlinear balance equations. A numerical resolution procedure is proposed\nand implemented within the open-source library ONSAS. The proposed formulation\nand its implementation are validated through the resolution of five examples,\nincluding a realistic wind turbine analysis problem. The results provided by\nthe proposed formulation are compared with analytic solutions and solutions\nobtained using a lumped mass approach. The proposed formulation provides\naccurate solutions for challenging numerical problems with large displacements\nand rotations.", "category": "physics_flu-dyn" }, { "text": "Constraints on scalar diffusion anomaly in three-dimensional flows\n having bounded velocity gradients: This study is concerned with the decay behaviour of a passive scalar $\\theta$\nin three-dimensional flows having bounded velocity gradients. Given an\ninitially smooth scalar distribution, the decay rate $d<\\theta^2>/dt$ of the\nscalar variance $<\\theta^2>$ is found to be bounded in terms of controlled\nphysical parameters. Furthermore, in the zero diffusivity limit, $\\kappa\\to0$,\nthis rate vanishes as $\\kappa^{\\alpha_0}$ if there exists an $\\alpha_0\\in(0,1]$\nindependent of $\\kappa$ such that $<|(-\\Delta)^{\\alpha/2}\\theta|^2><\\infty$ for\n$\\alpha\\le\\alpha_0$. This condition is satisfied if in the limit $\\kappa\\to0$,\nthe variance spectrum $\\Theta(k)$ remains steeper than $k^{-1}$ for large wave\nnumbers $k$. When no such positive $\\alpha_0$ exists, the scalar field may be\nsaid to become virtually singular. A plausible scenario consistent with\nBatchelor's theory is that $\\Theta(k)$ becomes increasingly shallower for\nsmaller $\\kappa$, approaching the Batchelor scaling $k^{-1}$ in the limit\n$\\kappa\\to0$. For this classical case, the decay rate also vanishes, albeit\nmore slowly -- like $(\\ln P_r)^{-1}$, where $P_r$ is the Prandtl or Schmidt\nnumber. Hence, diffusion anomaly is ruled out for a broad range of scalar\ndistribution, including power-law spectra no shallower than $k^{-1}$. The\nimplication is that in order to have a $\\kappa$-independent and non-vanishing\ndecay rate, the variance at small scales must necessarily be greater than that\nallowed by the Batchelor spectrum. These results are discussed in the light of\nexisting literature on the asymptotic exponential decay $<\\theta^2>\\sim\ne^{-\\gamma t}$, where $\\gamma>0$ is independent of $\\kappa$.", "category": "physics_flu-dyn" }, { "text": "Wake-foil interactions and energy harvesting efficiency in tandem\n oscillating foils: Oscillating foils in synchronized pitch/heave motions can be used to harvest\nhydrokinetic energy. By understanding the wake structure and its correlation\nwith the foil kinematics, predictive models for how foils can operate in array\nconfigurations can be developed. To establish a relationship between foil\nkinematics and wake characteristics, a wide range of kinematics is explored in\na two-foil tandem configuration with interfoil spacing from four to nine chord\nlengths separation and multiple interfoil phases. Using data from experiments\nand simulations, an in-depth wake analysis is performed and the mean velocity\nand the turbulent kinetic energy are quantified in the wake. With this energy\nquantification, the trailing foil efficiency is modified to account for the\nmean flow in addition to the energy transported by the coherent leading edge\nvortices (LEVs) shed from the leading foil. With the mean wake velocity, a\npredictive wake model is able to distinguish three regimes through analyzing\ntrailing foil efficiency profiles and the strength of the primary LEV shed from\nthe leading foil. Dividing the wake into regimes is an insightful way to narrow\nthe range of foil kinematics and configurations and improve the energy\nharvesting in a two-tandem foil array.", "category": "physics_flu-dyn" }, { "text": "Electro-osmotic Instability of Concentration Enrichment in Curved\n Geometries for an Aqueous Electrolyte: We report that an electro-osmotic instability of concentration enrichment in\ncurved geometries for an aqueous electrolyte, as opposed to the well-known one,\nis initiated exclusively at the enriched interface (anode), rather than at the\ndepleted one (cathode). For this instability, the limitation of unrealistically\nhigh material Peclet number in planar geometry is eliminated by the strong\nelectric field arising from the line charge singularity. In a model setup of\nconcentric circular electrodes, we show by stability analysis, numerical\nsimulation, and experimental visualization that instability occurs at the inner\nanode, below a critical radius of curvature. The stability criterion is also\nformulated in terms of a critical electric field and extended to arbitrary (2d)\ngeometries by conformal mapping. This discovery suggests that transport may be\nenhanced in processes limited by salt enrichment, such as reverse osmosis, by\ntriggering this instability with needle-like electrodes.", "category": "physics_flu-dyn" }, { "text": "The compressible granular collapse in a fluid as a continuum: validity\n of a Navier-Stokes model with $\u03bc(J)$-$\u03c6(J)$-rheology: The incompressible $\\mu(I)$-rheology has been used to study subaerial\ngranular flows with remarkable success. For subaquatic granular flows, drag\nbetween grains and the pore fluid is substantially higher and the physical\nbehaviour is more complex. High drag forces constrain the rearrangement of\ngrains and dilatancy, leading to a considerable build-up of pore pressure. Its\ntransient and dynamic description is the key to modelling subaquatic granular\nflows but out of the scope of incompressible models. In this work, we advance\nfrom the incompressible $\\mu(I)$-rheology to the compressible\n$\\mu(J)$-$\\phi(J)$-rheology to account for pore pressure, dilatancy, and the\nscaling laws under subaquatic conditions. The model is supplemented with\ncritical state theory to yield the correct properties in the quasi-static\nlimit. The pore fluid is described by an additional set of conservation\nequations and the interaction with grains is described by a drag model. This\nnew implementation enables us to include most of the physical processes\nrelevant for submerged granular flows in a highly transparent manner. Both, the\nincompressible and compressible rheologies are implemented into OpenFOAM and\nvarious simulations at low and high Stokes numbers are conducted with both\nframeworks. We found a good agreement of the $\\mu(J)$-$\\phi(J)$-rheology with\nlow Stokes number experiments, that incompressible models fail to describe. The\ncombination of granular rheology, pore pressure, and drag model leads to\ncomplex phenomena such as apparent cohesion, remoulding, hydroplaning, and\nturbidity currents. The simulations give remarkable insights into these\nphenomena and increase our understanding of subaquatic mass transports.", "category": "physics_flu-dyn" }, { "text": "The effective slip length and vortex formation in laminar flow over a\n rough surface: The flow of viscous incompressible fluid over a periodically corrugated\nsurface is investigated numerically by solving the Navier-Stokes equation with\nthe local slip and no-slip boundary conditions. We consider the effective slip\nlength which is defined with respect to the level of the mean height of the\nsurface roughness. With increasing corrugation amplitude the effective no-slip\nboundary plane is shifted towards the bulk of the fluid, which implies a\nnegative effective slip length. The analysis of the wall shear stress indicates\nthat a flow circulation is developed in the grooves of the rough surface\nprovided that the local boundary condition is no-slip. By applying a local slip\nboundary condition, the center of the vortex is displaced towards the bottom\nthe grooves and the effective slip length increases. When the intrinsic slip\nlength is larger than the corrugation amplitude, the flow streamlines near the\nsurface are deformed to follow the boundary curvature, the vortex vanishes, and\nthe effective slip length saturates to a constant value. Inertial effects\npromote vortex flow formation in the grooves and reduce the effective slip\nlength.", "category": "physics_flu-dyn" }, { "text": "Computations of Nonlinear Propagation of Sound Emitted from High Speed\n Mixing Layers: Non-linear sound propagation is investigated computationally by simulating\ncompressible time-developing mixing layers using the Large Eddy Simulation\n(LES) approach and solving the viscous Burgers Equation. The mixing layers are\nof convective Mach numbers of 0.4, 0.8 and 1.2. The LES results agree\nqualitatively with known flow behavior. Mach waves are observed in the near\nsound field of the supersonic mixing layer computed by the LES. These waves\nshow steepening typical to non-linear propagation. Further calculations using\nthe Burgers equation support this finding, where the initial wave slope has a\nrole in kicking them. No visible non-linear propagation effects were found for\nthe subsonic mixing layers. The effects of geometrical spreading and viscosity\nare also considered.", "category": "physics_flu-dyn" }, { "text": "Turbulence comes in bursts in stably stratified flows: There is a clear distinction between simple laminar and complex turbulent\nfluids. But in some cases, as for the nocturnal planetary boundary layer, a\nstable and well-ordered flow can develop intense and sporadic bursts of\nturbulent activity which disappear slowly in time. This phenomenon is\nill-understood and poorly modeled; and yet, it is central to our understanding\nof weather and climate dynamics. We present here a simple model which shows\nthat in stably stratified turbulence, the stronger bursts can occur when the\nflow is expected to be more stable. The bursts are generated by a rapid\nnon-linear amplification of energy stored in waves, and are associated with\nenergetic interchanges between vertical velocity and temperature (or density)\nfluctuations. Direct numerical simulations on grids of 2048^3 points confirm\nthis somewhat paradoxical result of measurably stronger events for more stable\nflows, displayed not only in the temperature and vertical velocity derivatives,\nbut also in the amplitude of the fields themselves.", "category": "physics_flu-dyn" }, { "text": "Geometric derivation and structure-preserving simulation of\n quasi-geostrophy on the sphere: We present a geometric derivation of the quasi-geostrophic equations on the\nsphere, starting from the rotating shallow water equations. We utilise\nperturbation series methods in vorticity and divergence variables. The\nderivation employs asymptotic analysis techniques, leading to a global\nquasi-geostrophic potential vorticity model on the sphere without approximation\nof the Coriolis parameter. The resulting model forms a closed system for the\nevolution of potential vorticity with a rich mathematical structure, including\nLagrangian and Hamiltonian descriptions. Formulated using the Lie-Poisson\nbracket reveals the geometric invariants of the quasi-geostrophic model.\nMotivated by these geometric results, simulations of quasi-geostrophic flow on\nthe sphere are presented based on structure-preserving Lie-Poisson\ntime-integration. We explicitly demonstrate the preservation of Casimir\ninvariants and show that the hyperbolic quasi-geostrophic equations can be\nsimulated in a stable manner over long time. We show the emergence of\nlongitudonal jets, wrapped around the circumference of the sphere in a general\ndirection that is perpendicular to the axis of rotation.", "category": "physics_flu-dyn" }, { "text": "Fully 3D Rayleigh-Taylor Instability in a Boussinesq Fluid: Rayleigh-Taylor instability occurs when a heavier fluid overlies a lighter\nfluid, and the two seek to exchange positions under the effect of gravity. We\npresent linearized theory for arbitrary 3D initial disturbances that grow in\ntime, and calculate the evolution of the interface for early times. A new\nspectral method is introduced for the fully 3D non-linear problem in a\nBoussinesq fluid, where the interface between the light and heavy fluid is\napproximated with a smooth but rapid density change in the fluid. The results\nof large-scale numerical calculation are presented in fully 3D geometry,\ncompared and contrasted with the early-time linearized theory.", "category": "physics_flu-dyn" }, { "text": "The method to increase the thrust of high Mach number Scramjets: The problem of engine unstart of scramjets has not been resolved. In this\npaper, the mechanism of engine unstart is discussed from the point of view of\nshock/shock interaction and deflagration-to-detonation transition. The\nshock/shock interaction leads to the nonlinear, transient and discontinuous\nprocess of the supersonic combustion flow field. This process is similar to the\ndeflagration-to-detonation transition process. If the velocity of\npre-combustion shock wave is faster than the velocity in the isolator, it will\npropagate upstream and cause the engine unstart. The C-J detonation velocity is\ndefined as the stable operation boundary of scramjets, which is the maximum\nshock wave produced by combustion theoretically. The scramjets will work stable\nif the velocity in the isolator is faster than the corresponding C-J detonation\nvelocity. The combustion characteristics and propulsive performance of\nscramjets is theoretically analyzed by using C-J detonation theory. For high\nMach number scramjets, the velocity in the isolator is much faster than the C-J\ndetonation velocity. Therefore, extra fuel and oxygen can be injected into the\ncombustor to increase the thrust as long as the shock wave velocity driven by\nthe combustion products is slower than the air velocity in the isolator. The\ntheoretical results agree well with the existing experimental results, which\ncan be used as a baseline for the development of scramjets.", "category": "physics_flu-dyn" }, { "text": "Electrical effects in superfluid helium. I. Thermoelectric effect in\n Einstein's capacitor: The Einstein's research ideas on the thermodynamical fluctuational nature of\ncertain electrical phenomena [1] and the physical nature of the electric\npotentials difference U in an electric capacitor at the temperature of T [2]\nwere proposed in 1906-1907. On the base of Einstein's research ideas, we\nexplain the recent experimental results [3, 4] and propose the consistent\ntheory on the physical nature of the electric effects in an electric capacitor\nat an action of the second sound standing wave in the superfluid Helium (4He)\nand in the rotational torsional mechanical resonator in Helium II. The use of\nthe Einstein's research approach, based on the consideration of an\ninterconnection between the thermal, mechanical and electrical fluctuations,\nallows us to obtain the quantitative theoretical research results, which are in\na good agreement with the experimental data on the correlations of the\nalternate low temperatures difference in the second sound wave, and the\nalternate electric potentials difference between the capacitor plates in the\nsuperfluid Helium as well as in the rotational torsional mechanical resonator\nin Helium II.", "category": "physics_flu-dyn" }, { "text": "Coalescence of Liquid Drops: Different Models Versus Experiment: The process of coalescence of two identical liquid drops is simulated\nnumerically in the framework of two essentially different mathematical models,\nand the results are compared with experimental data on the very early stages of\nthe coalescence process reported recently. The first model tested is the\n`conventional' one, where it is assumed that coalescence as the formation of a\nsingle body of fluid occurs by an instant appearance of a liquid bridge\nsmoothly connecting the two drops, and the subsequent process is the evolution\nof this single body of fluid driven by capillary forces. The second model under\ninvestigation considers coalescence as a process where a section of the free\nsurface becomes trapped between the bulk phases as the drops are pressed\nagainst each other, and it is the gradual disappearance of this `internal\ninterface' that leads to the formation of a single body of fluid and the\nconventional model taking over. Using the full numerical solution of the\nproblem in the framework of each of the two models, we show that the recently\nreported electrical measurements probing the very early stages of the process\nare better described by the interface formation/disappearance model. New\ntheory-guided experiments are suggested that would help to further elucidate\nthe details of the coalescence phenomenon. As a by-product of our research, the\nrange of validity of different `scaling laws' advanced as approximate solutions\nto the problem formulated using the conventional model is established.", "category": "physics_flu-dyn" }, { "text": "Viscous damping of gravity-capillary waves: Dispersion relations and\n nonlinear corrections: We discuss the impact of viscosity on nonlinear propagation of surface waves\nat the interface of air and a fluid of large depth. After a survey of the\navailable approximations of the dispersion relation, we propose to modify the\nhydrodynamic boundary conditions to model both short and long waves. From them,\nwe derive a nonlinear Schr\\\"odinger equation where both linear and nonlinear\nparts are modified by dissipation and show that the former plays the main role\nin both gravity and capillary-gravity waves while, in most situations, the\nlatter represents only small corrections. This provides a justification of the\nconventional approaches to damped propagation found in the literature.", "category": "physics_flu-dyn" }, { "text": "On the analogy between streamlined magnetic and solid obstacles: Analogies are elaborated in the qualitative description of two systems: the\nmagnetohydrodynamic (MHD) flow moving through a region where an external local\nmagnetic field (magnetic obstacle) is applied, and the ordinary hydrodynamic\nflow around a solid obstacle. The former problem is of interest both\npractically and theoretically, and the latter one is a classical problem being\nwell understood in ordinary hydrodynamics. The first analogy is the formation\nin the MHD flow of an impenetrable region -- core of the magnetic obstacle --\nas the interaction parameter $N$, i.e. strength of the applied magnetic field,\nincreases significantly. The core of the magnetic obstacle is streamlined both\nby the upstream flow and by the induced cross stream electric currents, like a\nforeign insulated insertion placed inside the ordinary hydrodynamic flow. In\nthe core, closed streamlines of the mass flow resemble contour lines of\nelectric potential, while closed streamlines of the electric current resemble\ncontour lines of pressure. The second analogy is the breaking away of attached\nvortices from the recirculation pattern produced by the magnetic obstacle when\nthe Reynolds number $Re$, i.e. velocity of the upstream flow, is larger than a\ncritical value. This breaking away of vortices from the magnetic obstacle is\nsimilar to that occurring past a real solid obstacle. Depending on the inlet\nand/or initial conditions, the observed vortex shedding can be either symmetric\nor asymmetric.", "category": "physics_flu-dyn" }, { "text": "Mathematical reformulation of the Kolmogorov-Richardson energy cascade\n in terms of vortex stretching: In this paper, with the aid of direct numerical simulations (DNS) of forced\nturbulence in a periodic domain, we mathematically reformulate the\nKolmogorov-Richardson energy cascade in terms of vortex stretching. By using\nthe description, we prove that if the Navier-Stokes flow satisfies a new\nregularity criterion in terms of the enstrophy production rate, then the flow\ndoes not blow up. Our DNS results seem to support this regularity criterion.\nNext, we mathematically construct the hierarchy of tubular vortices, which is\nstatistically self-similar in the inertial range. Under the assumptions of the\nscale-locally of the vortex stretching/compressing (i.e. energy cascade)\nprocess and the statistical independence between vortices that are not directly\nstretched or compressed, we can derive the $-5/3$ power law of the energy\nspectrum of statistically stationary turbulence without directly using the\nKolmogorov hypotheses.", "category": "physics_flu-dyn" }, { "text": "Optimal Nonlinear Eddy Viscosity in Galerkin Models of Turbulent Flows: We propose a variational approach to identification of an optimal nonlinear\neddy viscosity as a subscale turbulence representation for POD models. The\nansatz for the eddy viscosity is given in terms of an arbitrary function of the\nresolved fluctuation energy. This function is found as a minimizer of a cost\nfunctional measuring the difference between the target data coming from a\nresolved direct or large-eddy simulation of the flow and its reconstruction\nbased on the POD model. The optimization is performed with a data-assimilation\napproach generalizing the 4D-VAR method. POD models with optimal eddy\nviscosities are presented for a 2D incompressible mixing layer at $Re=500$\n(based on the initial vorticity thickness and the velocity of the high-speed\nstream) and a 3D Ahmed body wake at $Re=300,000$ (based on the body height and\nthe free-stream velocity). The variational optimization formulation elucidates\na number of interesting physical insights concerning the eddy-viscosity ansatz\nused. The 20-dimensional model of the mixing-layer reveals a negative\neddy-viscosity regime at low fluctuation levels which improves the transient\ntimes towards the attractor. The 100-dimensional wake model yields more\naccurate energy distributions as compared to the nonlinear modal eddy-viscosity\nbenchmark {proposed recently} by \\\"Osth et al. (2014). Our methodology can be\napplied to construct quite arbitrary closure relations and, more generally,\nconstitutive relations optimizing statistical properties of a broad class of\nreduced-order models.", "category": "physics_flu-dyn" }, { "text": "Shock polars for ideal non-polytropic gas: We show that shock polars for ideal non-polytropic gas (thermally but not\ncalorically perfect) have a unique velocity angle maximum, the critical shock,\nassuming convex equation of state (positive fundamental derivative) and other\nstandard conditions. We also show that the critical shock is always transonic.\nIn the process we show that temperature, pressure, energy, enthalpy, normal\nmass flux and entropy are strictly increasing along the forward Hugoniot\ncurves, and hence along the polar from vanishing to normal shock; speed is\nstrictly decreasing along the entire polar, mass flux and importantly Mach\nnumber are decreasing on subsonic parts of the polar. If the equation of state\nis ideal but not convex, or convex but not ideal, counterexamples can be given\nwith multiple critical shocks, permitting more than two shocks attaining the\nsame velocity angle, in particular more than one shock of weak type.", "category": "physics_flu-dyn" }, { "text": "Collision of elastic drop with thin cylinder: The collision of water and elastic liquid drops with a thin cylinder (thread)\nis studied. The droplet flight trajectory and the cylinder axis are mutually\nperpendicular. Attention is focused on the difference between collisions of\nwater drops and drops of elastic fluids. In the experiments, the drop diameter\nwas 3 mm, the diameter of horizontal stainless steel cylinders was 0.4 and 0.8\nmm. The drops were formed by slowly pumping liquid through a vertical stainless\nsteel capillary with an outer diameter of 0.8 mm, from which droplets were\nperiodically detached under the action of gravity. The droplet velocity before\ncollision was defined by the distance between the capillary cut and the target\n(cylinder); in experiments, this distance was approximately 5, 10, and 20 mm.\nThe drop velocities before impact are estimated in the range of 0.2-0.5 m/s.\nThe collision process was monitored by high-speed video recording methods with\na frame rate of 240 and 960 Hz. The test liquids were water and aqueous\nsolutions of polyacrylamide of molecular weight 11 million and concentrations\nof 100 and 1000 ppm (PAM-100 and PAM-1k). Experiments have shown that,\ndepending on the drop impact height and polymer concentration, different\nscenarios of a drop collision with a thin cylinder are possible: a short-term\nrecoil of a drop from an obstacle, a drop flowing around a cylindrical obstacle\nwhile maintaining the continuity of the drop, the breakup of a drop into two\nsecondary drops, one of which can continue flight and the other one is captured\nby the cylinder, or both secondary droplets continue to fly, the drop also can\nbe captured by the cylinder, until the impact of the next drop(s) forces the\naccumulated drop detach from the cylinder. Numerical modeling satisfactorily\nreproduces the phenomena observed in the experiment.", "category": "physics_flu-dyn" }, { "text": "Finite-size anisotropy in statistically uniform porous media: Anisotropy of the permeability tensor in statistically uniform porous media\nof sizes used in typical computer simulations is studied. Although such systems\nare assumed to be isotropic by default, we show that de facto their anisotropic\npermeability can give rise to significant changes of transport parameters such\nas permeability and tortuosity. The main parameter controlling the anisotropy\nis $a/L$, being the ratio of the obstacle to system size. Distribution of the\nangle $\\alpha$ between the external force and the volumetric fluid stream is\nfound to be approximately normal, and the standard deviation of $\\alpha$ is\nfound to decay with the system size as $(a/L)^{d/2}$, where $d$ is the space\ndimensionality. These properties can be used to estimate both\nanisotropy-related statistical errors in large-scale simulations and the size\nof the representative elementary volume.", "category": "physics_flu-dyn" }, { "text": "3D Euler equations and ideal MHD mapped to regular systems: probing the\n finite-time blowup hypothesis: We prove by an explicit construction that solutions to incompressible 3D\nEuler equations defined in the periodic cube can be mapped bijectively to a new\nsystem of equations whose solutions are globally regular. We establish that the\nusual Beale-Kato-Majda criterion for finite-time singularity (or blowup) of a\nsolution to the 3D Euler system is equivalent to a condition on the\ncorresponding \\emph{regular} solution of the new system. In the hypothetical\ncase of Euler finite-time singularity, we provide an explicit formula for the\nblowup time in terms of the regular solution of the new system. The new system\nis amenable to being integrated numerically using similar methods as in Euler\nequations. We propose a method to simulate numerically the new regular system\nand describe how to use this to draw robust and reliable conclusions on the\nfinite-time singularity problem of Euler equations, based on the conservation\nof quantities directly related to energy and circulation. The method of mapping\nto a regular system can be extended to any fluid equation that admits a\nBeale-Kato-Majda type of theorem, e.g. 3D Navier-Stokes, 2D and 3D\nmagnetohydrodynamics, and 1D inviscid Burgers. We discuss briefly the case of\n2D ideal magnetohydrodynamics. In order to illustrate the usefulness of the\nmapping, we provide a thorough comparison of the analytical solution versus the\nnumerical solution in the case of 1D inviscid Burgers equation.", "category": "physics_flu-dyn" }, { "text": "Mixing in two-dimensional shear flow with smooth fluctuations: Chaotic variations in flow speed up mixing of scalar fields via intensified\nstirring. This paper addresses the statistical properties of a passive scalar\nfield mixing in a regular shear flow with random fluctuations against its\nbackground. We consider two-dimensional flow with shear component dominating\nover smooth fluctuations. Such flow is supposed to model passive scalar mixing\ne.g. inside a large-scale coherent vortex forming in two-dimensional turbulence\nor in elastic turbulence in a micro-channel. We examine both the decaying case\nand the case of the continuous forcing of the scalar variances. In both cases\ndynamics possesses strong intermittency, that can be characterized via the\nsingle-point moments and correlation functions calculated in our work. We\npresent general qualitative properties of pair correlation function as well as\ncertain quantitative results obtained in the framework of the model with\nfluctuations that are short correlated in time.", "category": "physics_flu-dyn" }, { "text": "Holographic shear rheology of viscoelastic fluids: In this study, we report the use of digital holography microscopy (DHM) for\n3D-resolved flow kinematics and shear rheometry of viscoelastic polymeric\nfluids. We computationally reconstruct the recorded holograms to visualize the\ntracer imbued flow volume in microchannels, followed by implementation of\nparticle tracking velocimetry (PTV) to quantitate spatially-resolved velocity\nfields in 3D. In order to select optimal parameters for DHM-PTV\ncharacterization of complex fluids, we studied the effect of hologram recording\ndistance, seeding density and particle size. Using the optimal parameters, we\nshow quantitative characterization of the shear rheology from the velocity\nfields without any a-priori assumptions of wall boundary condition or\nconstitutive equation. The viscosity versus shear rate data for Newtonian and\npolyethylene oxide solutions could be measured in the range of ~ 0.05 - 20,000\ns-1 with just four input flow rates. This data from holographic shear rheometry\nwas found to be in good agreement with computational fluid dynamics simulations\nand macrorheometry. The holographic shear rheology technique remained\nunaffected by wall-slip events and instead provided an avenue to quantitate\nslip severity. Finally, we discuss holographic visualization of particle\nmigration in microfluidic flows which can limit flow field access while at the\nsame time provide a fingerprint of the suspending fluid rheology.", "category": "physics_flu-dyn" }, { "text": "The time evolution equation for advective heat transport as a constraint\n for optimal bounds in Rayleigh-B\u00e9nard convection: Upper bounds on the heat transport and other quantities of interest in\nRayleigh-B\\'enard convection are derived in previous work from constraints\nresulting from the equations of time evolution for kinetic energy, the root\nmean square of temperature, and the temperature averaged over horizontal\nplanes. Here, we investigate the effect of a new constraint derived from the\ntime evolution equation for the advective heat transport. This additional\nconstraint leads to improved bounds on the toroidal dissipation.", "category": "physics_flu-dyn" }, { "text": "Cavitation nuclei regeneration in water-particle suspension: Bubble nucleation in water induced by boiling, gas supersaturation or\ncavitation usually originates from pre-existing gas cavities trapped into solid\ndefects. Even though the destabilization of such gas pockets, called nuclei,\nhas been extensively studied, little is known on the nuclei dynamic. Here,\nnuclei of water-particle suspensions are excited by acoustic cavitation, and\ntheir dynamic is investigated by monitoring the cavitation probability over\nseveral thousand pulses. A stable and reproducible cavitation probability\nemerges after a few thousand pulses and depends on particle concentration,\nhydrophobicity, and dissolved gas content. Our observations indicate that a\nstable nuclei distribution is reached at a later-time, different from\npreviously reported nuclei depletion in early-time. This apparent paradox is\nelucidated by varying the excitation rate, where the cavitation activity\nincreases with the repetition period, indicating that the nuclei depletion is\nbalanced by spontaneous nucleation or growth of nuclei. A model of this\nself-supporting generation of nuclei suggests an origin from dissolved gas\nadsorption on surfaces. The method developed can be utilized to further\nunderstand the spontaneous formation and distribution of nano-sized bubbles on\nheterogeneous surfaces.", "category": "physics_flu-dyn" }, { "text": "Modulation-resonance mechanism for surface waves in a two-layer fluid\n system: We propose a Boussinesq-type model to study the surface/interfacial wave\nmanifestation of an underlying, slowly-varying, long-wavelength, baroclinic\nflow in a two-layer, density-stratified system. The results of our model show\nnumerically that, under strong nonlinearity, surface waves, with their typical\nwavenumber being the resonant $k_{\\mathrm{res}}$, can be generated locally at\nthe leading edge of the underlying slowly-varying, long-wavelength baroclinic\nflow. Here, the resonant $k_{\\mathrm{res}}$ satisfies the class 3 triad\nresonance condition among two short-mode waves and one long-mode wave in which\nall waves propagate in the same direction. Moreover, when the slope of the\nbaroclinic flow is sufficiently small, only one spatially-localized\nlarge-amplitude surface wave packet can be generated at the leading edge. This\nlocalized surface wave packet becomes high in amplitude and large in group\nvelocity after the interaction with its surrounding waves. These results are\nqualitatively consistent with various experimental observations including\nresonant surface waves at the leading edge of an internal wave. Subsequently,\nwe propose a mechanism, referred to as the modulation-resonance mechanism,\nunderlying these surface phenomena, based on our numerical simulations. The\nproposed modulation-resonance mechanism combines the linear modulation\n(ray-based) theory for the spatiotemporal asymmetric behavior of surface waves\nand the nonlinear class 3 triad resonance theory for the energy focusing of\nsurface waves around the resonant wavenumber $k_{\\mathrm{res}}$ in Fourier\nspace.", "category": "physics_flu-dyn" }, { "text": "Wave propagation in stenotic vessels; theoretical analysis and\n comparison between 3D and 1D fluid-structure-interaction models: Using analytical expressions for the pressure and velocity waveforms in\ntapered vessels, we construct a linear 1D model for wave propagation in\nstenotic vessels in the frequency domain. We demonstrate that using only two\nparameters to approximate the exact geometry of the constriction (length and\ndegree of stenosis), we can construct a model that can be solved analytically\nand can approximate with excellent accuracy the response of the original vessel\nfor a wide range of physiologically relevant frequencies. We then proceed to\ncompare the 1D results with full 3D FSI results from the literature for\nparameters corresponding to an idealized stenotic carotid artery. We find\nexcellent matching with the volume flow rate over the cardiac cycle (less than\n$1\\%$ error). Using results from DNS simulations to parametrize the velocity\nprofile in the stenotic region, we manage to predict also the pressure\ndistribution with small error (a few percentage points). The method proposed in\nthe paper can be used to approximate vessels of arbitrary shape profile and can\nbe extended to cover the whole cardiovascular tree. Recursive expressions make\nthe solution very fast and open the possibility of carrying out sensitivity and\nuncertainty quantification studies that require thousands (or even millions) of\nsimulations with minimal cost.", "category": "physics_flu-dyn" }, { "text": "The influence of van der Waals interactions on a bubble moving in a tube: We consider the unsteady thin-film dynamics of a long bubble of negligible\nviscosity that advances at a uniform speed in a cylindrical capillary tube. The\nbubble displaces a viscous, nonwetting fluid, creating a thin film between its\ninterface and the tube walls. The film is considered thin enough that\nintermolecular forces in the form of van der Waals attractions are significant\nand thin-film rupture is possible. In the case of negligible intermolecular\nforces, a steady-state solution exits where a film of uniform thickness is\ndeposited in the annular region between the bubble interface and the tube\nwalls. However, once intermolecular interactions are important, the interface\nis perturbed out of its steady state and either (i) the perturbation grows\nsufficiently before reaching the rear meniscus of the bubble such that rupture\noccurs; or (ii) the perturbation remains small due to weak intermolecular\nforces until it leaves the bubble interface through the rear meniscus. We\nobtain, both numerically and asymptotically, the time-scale over which rupture\noccurs, and thus, we find a critical capillary number, depending on the bubble\nlength and the strength of the intermolecular forces, below which the film is\npredicted to rupture.", "category": "physics_flu-dyn" }, { "text": "Predominance of pressure transport in spatial energy budget for a mixing\n layer approaching absolute instability: In this letter, we report the outcome of a spatial energy budget performed\nfor the linear convective instability of the plane incompressible mixing layer\nwithin the inviscid framework. We find that, as the critical condition for the\nonset of absolute instability is approached, the integrated pressure-transport\nterm becomes increasingly more prominent as compared to the integrated\nproduction term and it dominates the energy budget completely at the critical\ncondition. This implies that, near the threshold of absolute instability, the\ngrowth of disturbances is almost entirely due to the pressure transport\nmechanism (rather than the production mechanism), which is a striking result.\nThe part of the pressure-transport term that represents the work done by the\nfluctuating pressure forces is seen to be primarily responsible for the\nobserved shift in the energy balance. These results can help us better\nunderstand the physical processes causing absolute instability in a mixing\nlayer. In particular, the redistribution of disturbance energy in streamwise\ndirection by fluctuating pressure, which is \"non-local\" in character for\nincompressible flows, seems to play a key role in this respect.", "category": "physics_flu-dyn" }, { "text": "Phase-Field Modeling of Selective Laser Brazing of Diamond Grits: Diamond grit is widely used in cutting, grinding, and polishing tools for its\nsuperior mechanical properties and performance in machining hard materials.\nSelective laser brazing (SLB) of diamond grits is a new additive manufacturing\ntechnique that has great potential to fabricate the next generation of\nhigh-performance diamond tools. However, fundamental understanding and\nquantitative analysis for the design and tuning of the SLB process and the\nresulting bonding efficiency are not yet established as the process is\ncomplicated by heating, fusion, wetting, solidification, grit migration,\nbonding, reaction, and the interplay between these effects. We present a\nthermodynamically consistent phase-field theoretical model for the prediction\nof melting and wetting of SLB on diamond grits using a powder-based additive\nmanufacturing technique. The melting dynamics is driven by laser heating in a\nchamber filled with argon gas and is coupled with the motion of multiple\nthree-phase contact lines. The relevant wetting dynamics, interfacial\nmorphology, and temperature distribution are computationally resolved in a\nsimplified 2D configuration.", "category": "physics_flu-dyn" }, { "text": "Modal analysis of high-fidelity simulations in turbomachinery: We revisit recently published high-fidelity implicit large eddy simulation\ndatasets obtained with a high-order discontinuous Galerkin spectral element\nmethod and analyse them using Proper Orthogonal Decomposition (POD) as well as\nSpectral Proper Orthogonal Decomposition (SPOD). The first configuration is the\nMTU T161 low-pressure turbine cascade with resolved end wall boundary layers in\na clean version and one with incoming turbulent wakes. We focus on the\nbehaviour of the laminar separation bubble and the secondary flow system and\nhow these phenomena are affected by incoming wakes. The second configuration is\na transonic compressor cascade in which we analyse the unsteady behaviour of\nthe shock wave boundary layer interaction. Throughout the paper, we try to\ndiscuss not only the flow physics but also insights into how the modal\ndecomposition techniques can help facilitate understanding and where their\nlimitations are.", "category": "physics_flu-dyn" }, { "text": "Vortex rings generated by a translating disk from start to stop: In this article, we investigate experimentally and numerically the time\nevolution of vortex rings generated by the translation of a rigid disk in a\nfluid initially at rest and submitted to an acceleration followed by a\ndeceleration. The size of the disk and its motion in terms of stroke length and\ntravel time are varied as control parameters. The start-up vortex ring created\nin the near wake of the disk is characterized experimentally by PIV, and the\nmeasurements agree quantitatively with axisymmetric numerical simulations\nperformed with the Basilisk flow solver. The maximum radius and circulation of\nthe annular vortex and its dynamics are shown to follow different power laws\nwith the control parameters. The modeling adapted from Wedemeyer's\ntwo-dimensional theoretical calculations [E. Wedemeyer, Ausbildung eines\nWirbelpaares an den Kanten einer Platte, Ingenieur-Archiv 30, (1961)] captures\nthe observed scaling laws. Besides, after the disk stops, a secondary\n``stopping\" vortex ring is generated, which is shown to affect the motion of\nthe main vortex ring.", "category": "physics_flu-dyn" }, { "text": "Characteristics of dynamic contact-angle in presence of surface-charge: We account for the presence of surface charges towards describing variations\nin the dynamic contact angle of an advancing liquid-gas meniscus. Starting from\nthe thin-film based formalism, we present closed-form analytical expressions\nrelating the dynamic contact-angle with the capillary number (essentially\nnormalized contact-line speed) and other interfacial parameters. Specifically,\nour analysis presents, within the realm of hydrodynamic paradigm, a connection\nbetween the micro- and macro-scale physics at the vicinity of the contact-line\nregion, taking the combined confluence of viscous and capillary forces along\nwith van der Waals and electrostatic interactions. This connection rationalizes\nthe hitherto reported anomalous window of the magnitude of the microscopic\nlength scales required to corroborate experimental data for ionic liquids.\nMoreover, our analysis shows the possibility of a transition from strong to\nweak influence of surface charge in a dynamic fashion with contact-line speed\nalong with other electrochemical parameters, leading towards substantiating the\nreported multiple asymptotic behavior of dynamic contact-angle over a wide\nrange of capillary numbers.", "category": "physics_flu-dyn" }, { "text": "A Transfer Matrix for the Input Impedance of weakly tapered Cones as of\n Wind Instruments: A formula for the local acoustical admittance in a conical waveguide with\nviscous and thermal losses given by Nederveen [(1969) Acoustical Aspects of\nWoodwind Instruments (Frits Knuf, Amsterdam)] is rewritten as an impedance\ntransmission matrix. Based on a self-consistent approximation for the cone, it\ndiffers from other one-dimensional transmission matrices used in musical\nacoustics, which implicitly include the loss model of a cylinder. The resonance\nfrequencies of air columns calculated with this transmission matrix are in\nbetter agreement with more comprehensive models. Even for long cones with a\nslight taper, there is no need to discretize along the axis.", "category": "physics_flu-dyn" }, { "text": "Induced-charge Electrophoresis of Metallo-dielectric Particles: The application of AC electric fields in aqueous suspensions of anisotropic\nparticles leads to unbalanced liquid flows and nonlinear, induced-charge\nelectrophoretic (ICEP) motion. We report experimental observations of the\nmotion of \"Janus\" microparticles with one dielectric and one metal-coated\nhemisphere induced by uniform fields of frequency 100 Hz - 10 kHz in NaCl\nsolutions. The motion is perpendicular to the field axis and persists after\nparticles are attracted to a glass wall. This phenomenon may find applications\nin microactuators, microsensors, and microfluidic devices.", "category": "physics_flu-dyn" }, { "text": "Turbulence Momentum Transport and Prediction of the Reynolds Stress in\n Canonical Flows: We present a unique method for solving for the Reynolds stress in turbulent\ncanonical flows, based on the momentum balance for a control volume moving at\nthe local mean velocity. A differential transform converts this momentum\nbalance to a solvable form. Comparisons with experimental and computational\ndata in simple geometries show quite good agreements. An alternate picture for\nthe turbulence momentum transport is offered, as verified with data, where the\nturbulence momentum is transported by the mean velocity while being dissipated\nby viscosity. The net momentum transport is the Reynolds stress. This\nturbulence momentum balance is verified using DNS and experimental data.\nImplications of this work are that the Reynolds stress can be written\nexplicitly in terms of basic turbulence parameters in a simple form, derived\nfrom pure fluid physics, and that potential exists for applications of this\nconcept to more complex turbulent flows.", "category": "physics_flu-dyn" }, { "text": "Effect of slip on circulation inside a droplet: Internal recirculation in a moving droplet plays an important role in several\ndroplet-based microfluidic devices as it enhances mixing, chemical reaction and\nheat transfer. The occurrence of fluid slip at the wall, which becomes\nprominent at high shear rates and lower length scales, results in a significant\nchange in droplet circulation. Using molecular dynamics (MD) simulations, the\npresence of circulation in droplets is demonstrated and quantified. Circulation\nis shown to vary inversely with slip length, which is a measure of interface\nwettability. A simple circulation model is established that captures the effect\nof slip on droplet circulation. Scaling parameters for circulation and slip\nlength are identified from the circulation model which leads to the collapse of\ndata for droplets with varying aspect ratio (AR) and slip length. The model is\nvalidated using continuum and MD simulations and is shown to be accurate for\ndroplets with high AR.", "category": "physics_flu-dyn" }, { "text": "Linear Analysis on Multiple-relaxation-time Lattice Boltzmann Method: The development of multiple-relaxation-time (MRT) Lattice Boltzmann method\n(LBM) is a significant contribution in improving the numerical behavior,\nrevealing the math and physics mechanism and extending the application of LBM.\nHowever, some of the MRT schemes proposed previously are not\nphysically-consistent. In this work, we take D2Q9 as a example to show how to\nderive physically-consistent MRT-LBM schemes by eigenvalue decomposition of the\ncollision operator. In addition, the scheme is validated by the equivalence to\nNavier-Stokes equations and numerical simulations.", "category": "physics_flu-dyn" }, { "text": "Gap Size Effects for the Kelvin-Helmholtz Instability in a Hele-Shaw\n Cell: We report experimental results for the Kelvin-Helmholtz instability between\ntwo immiscible fluids in parallel flow in a Hele-Shaw cell. We focus our\ninterest on the influence of the gap size between the walls on the instability\ncharacteristics. Experimental results show that the instability threshold, the\ncritical wavelength, the phase velocity and the spatial growth rate depend on\nthis gap size. These results are compared to both the previous two-dimensional\nanalysis of Gondret and Rabaud (1997) and the three-dimensional analysis\nrecently derived by Plourabou=E9 and Hinch (2001), showing that the agreement\nis still not complete especially when gap size increases.", "category": "physics_flu-dyn" }, { "text": "Drag Reduction in Flows Past 2D and 3D Circular Cylinders Through Deep\n Reinforcement Learning: We investigate drag reduction mechanisms in flows past two- and\nthree-dimensional cylinders controlled by surface actuators using deep\nreinforcement learning. We investigate 2D and 3D flows at Reynolds numbers up\nto 8,000 and 4,000, respectively. The learning agents are trained in planar\nflows at various Reynolds numbers, with constraints on the available actuation\nenergy. The discovered actuation policies exhibit intriguing generalization\ncapabilities, enabling open-loop control even for Reynolds numbers beyond their\ntraining range. Remarkably, the discovered two-dimensional controls, inducing\ndelayed separation, are transferable to three-dimensional cylinder flows. We\nexamine the trade-offs between drag reduction and energy input while discussing\nthe associated mechanisms. The present work paves the way for control of\nunsteady separated flows via interpretable control strategies discovered\nthrough deep reinforcement learning.", "category": "physics_flu-dyn" }, { "text": "On the kurtosis of deep-water gravity waves: In this paper, we revisit Janssen's (2003) formulation for the dynamic excess\nkurtosis of weakly nonlinear gravity waves at deep water. For narrowband\ndirectional spectra, the formulation is given by a sixfold integral that\ndepends upon the Benjamin-Feir index and the parameter\n$R=\\sigma_{\\theta}^{2}/2\\nu^{2}$, a measure of short-crestedness for the\ndominant waves with $\\nu$ and $\\sigma_{\\theta}$} denoting spectral bandwidth\nand angular spreading. Our refinement leads to a new analytical solution for\nthe dynamic kurtosis of narrowband directional waves described with a Gaussian\ntype spectrum. For multidirectional or short-crested seas initially homogenous\nand Gaussian, in a focusing (defocusing) regime dynamic kurtosis grows\ninitially, attaining a positive maximum (negative minimum) at the intrinsic\ntime scale \\[\n\\tau_{c}=\\nu^{2}\\omega_{0}t_{c}=1/\\sqrt{3R},\\qquad\\mathrm{or}\\qquad\nt_{c}/T_{0}\\approx0.13/\\nu\\sigma_{\\theta}, \\] where $\\omega_{0}=2\\pi/T_{0}$\ndenotes the dominant angular frequency. Eventually the dynamic excess kurtosis\ntends monotonically to zero as the wave field reaches a quasi-equilibrium state\ncharacterized with nonlinearities mainly due to bound harmonics. Quasi-resonant\ninteractions are dominant only in unidirectional or long-crested seas where the\nlonger-time dynamic kurtosis can be larger than that induced by bound\nharmonics, especially as the Benjamin-Feir index increases. Finally, we discuss\nthe implication of these results on the prediction of rogue waves.", "category": "physics_flu-dyn" }, { "text": "Statistics of the Navier-Stokes-alpha-beta regularization model for\n fluid turbulence: We explore one-point and two-point statistics of the Navier-Stokes-alpha-beta\nregularization model at moderate Reynolds number in homogeneous isotropic\nturbulence. The results are compared to the limit cases of the\nNavier-Stokes-alpha model and the Navier-Stokes-alpha-beta model without\nsubgrid-scale stress, as well as with high resolution direct numerical\nsimulation. After reviewing spectra of different energy norms of the\nNavier-Stokes-alpha-beta model, the Navier-Stokes-alpha model, and\nNavier-Stokes-alpha-beta model without subrid-scale stress, we present\nprobability density functions and normalized probability density functions of\nthe filtered and unfiltered velocity increments along with longitudinal\nvelocity structure functions of the regularization models and direct numerical\nsimulation results. We highlight differences in the statistical properties of\nthe unfiltered and filtered velocity fields entering the governing equations of\nthe Navier-Stokes-alpha and Navier-Stokes-alpha-beta models and discuss the\nusability of both velocity fields for realistic flow predictions. The influence\nof the modified viscous term in the Navier-Stokes-alpha-beta model is studied\nthrough comparison to the case where the underlying subgrid-scale stress tensor\nis neglected. The filtered velocity field is found to have physically more\nviable probability density functions and structure functions for the\napproximation of direct numerical simulation results, whereas the unfiltered\nvelocity field is found to have flatness factors close to direct numerical\nsimulation results.", "category": "physics_flu-dyn" }, { "text": "Transient structures in rupturing thin-films: Marangoni-induced\n symmetry-breaking pattern formation in viscous fluids: In the minutes immediately preceeding the rupture of a soap bubble,\ndistinctive and repeatable patterns can be observed. These quasi-stable\ntransient structures are associated with the instabilities of the complex\nMarangoni flows on the curved thin film in the presence of a surfactant\nsolution. Here, we report a generalised Cahn-Hilliard-Swift-Hohenberg model\nderived using asymptotic theory which describes the quasi-elastic wrinkling\npattern formation and the consequent coarsening dynamics in a curved\nsurfactant-laden thin film. By testing the theory against experiments on soap\nbubbles, we find quantitative agreement with the analytical predictions of the\nnucleation and the early coarsening phases associated with the patterns. Our\nfindings provide fundamental physical understanding that can be used to\n(de-)stabilise thin films in the presence of surfactants and have important\nimplications for both natural and industrial contexts, such as the production\nof thin coating films, foams, emulsions and sprays.", "category": "physics_flu-dyn" }, { "text": "Fluid flow structures in an evaporating sessile droplet depending on the\n droplet size and properties of liquid and substrate: We investigate numerically quasi-steady internal flows in an axially\nsymmetrical evaporating sessile droplet depending on the ratio of substrate to\nfluid thermal conductivities, fluid volatility, contact angle and droplet size.\nTemperature distributions and vortex structures are obtained for droplets of\n1-hexanol, 1-butanol and ethanol. To this purpose, the hydrodynamics of an\nevaporating sessile drop, effects of the thermal conduction in the droplet and\nsubstrate and diffusion of vapor in air have been jointly taken into account.\nThe equations have been solved by finite element method using ANSYS Fluent. The\nphase diagrams demonstrating the number and orientation of the vortices as\nfunctions of the contact angle and the ratio of substrate to fluid thermal\nconductivities, are obtained and analyzed for various values of parameters. In\nparticular, influence of gravity on the droplet shape and the effect of droplet\nsize have been considered. We have found that the phase diagrams of highly\nvolatile droplets do not contain a subregion corresponding to a reversed single\nvortex, and their single-vortex subregion becomes more complex. The phase\ndiagrams for droplets of larger size do not contian subregions corresponding to\na regular single vortex and to three vortices. We demonstrate how the\nsingle-vortex subregion disappears with a gradual increase of the droplet size.", "category": "physics_flu-dyn" }, { "text": "A MHD reverse flow in 90 degree bend: A two-dimensional flow in a 90 degree bent channel is considered. A magnetic\nfield is uniform and parallel to inlet branch of the channel. A spectral/hp\nelement method was used for liquid motion calculations. Three types of steady\nflows were detected. It was found that magnetic forces can suppress a pressure\ngradient and throw out liquid from the inlet, a relatively large reverse flow\nappears.", "category": "physics_flu-dyn" }, { "text": "New exact superposition solutions to KdV2 equation: New exact solutions to the KdV2 equation (known also as the extended KdV\nequation) are constructed. The KdV2 equation is a second order approximation of\nthe set of Boussinesq's equations for shallow water waves which in first order\napproximation yields KdV. The exact solutions\n~$\\frac{A}{2}\\left(\\dn^2[B(x-vt),m]\\pm \\sqrt{m}\\,\\cn [B(x-vt),m]\\dn\n[B(x-vt),m]\\right)+D$~ in the form of periodic functions found in the paper\ncomplement other forms of exact solutions to KdV2 obtained earlier, i.e., the\nsolitonic ones and periodic ones given by a single $\\cn^2$ or $\\dn^2$ Jacobi\nelliptic functions.", "category": "physics_flu-dyn" }, { "text": "Thermocapillary migration of a droplet with a thermal source at large\n Reynolds and Marangoni numbers: The {\\it unsteady} process for thermocapillary droplet migration at large\nReynolds and Marangoni numbers has been previously reported by identifying a\nnonconservative integral thermal flux across the surface in the {\\it steady}\nthermocapillary droplet migration, [Wu and Hu, J. Math. Phys. {\\bf 54} 023102,\n(2013)]. Here we add a thermal source in the droplet to preserve the integral\nthermal flux across the surface as conservative, so that thermocapillary\ndroplet migration at large Reynolds and Marangoni numbers can reach a {\\it\nquasi-steady} process. Under assumptions of {\\it quasi-steady} state and\nnon-deformation of the droplet, we make an analytical result for the {\\it\nsteady} thermocapillary migration of droplet with the thermal source at large\nReynolds and Marangoni numbers. The result shows that the thermocapillary\ndroplet migration speed slowly increases with the increase of Marangoni number.", "category": "physics_flu-dyn" }, { "text": "Multi-Scale Turbulence Injector: a new tool to generate intense\n homogeneous and isotropic turbulence for premixed combustion: Nearly homogeneous and isotropic, highly turbulent flow, generated by an\noriginal multi-scale injector is experimentally studied. This multi-scale\ninjector is made of three perforated plates shifted in space such that the\ndiameter of their holes and their blockage ratio increase with the downstream\ndistance. The Multi-Scale Turbulence Injector (hereafter, MuSTI) is compared\nwith a Mono-Scale Turbulence Injector (MoSTI), the latter being constituted by\nonly the last plate of MuSTI. This comparison is done for both cold and\nreactive flows. For the cold flow, it is shown that, in comparison with the\nclassical mono-scale injector, for the MuSTI injector: (i) the turbulent\nkinetic energy is roughly twice larger, and the kinetic energy supply is\ndistributed over the whole range of scales. This is emphasized by second and\nthird order structure functions. (ii) the transverse fluxes of momentum and\nenergy are enhanced, (iii) the homogeneity and isotropy are reached earlier\n($\\approx 50$%), (iv) the jet merging distance is the relevant scaling\nlength-scale of the turbulent flow, (v) high turbulence intensity ($\\approx\n15$%) is achieved in the homogeneous and isotropic region, although the\nReynolds number based on the Taylor microscale remains moderate ($Re_\\lambda\n\\approx 80$). In a second part, the interaction between the multi-scale\ngenerated turbulence and the premixed flame front is investigated by laser\ntomography. A lean V-shaped methane/air flame is stabilised on a heated rod in\nthe homogeneous and isotropic region of the turbulent flow. The main\nobservation is that the flame wrinkling is hugely amplified with the\nmulti-scale generated injector, as testified by the increase of the flame brush\nthickness.", "category": "physics_flu-dyn" }, { "text": "Super-resolving sparse observations in partial differential equations: A\n physics-constrained convolutional neural network approach: We propose the physics-constrained convolutional neural network (PC-CNN) to\ninfer the high-resolution solution from sparse observations of spatiotemporal\nand nonlinear partial differential equations. Results are shown for a chaotic\nand turbulent fluid motion, whose solution is high-dimensional, and has fine\nspatiotemporal scales. We show that, by constraining prior physical knowledge\nin the CNN, we can infer the unresolved physical dynamics without using the\nhigh-resolution dataset in the training. This opens opportunities for\nsuper-resolution of experimental data and low-resolution simulations.", "category": "physics_flu-dyn" }, { "text": "Wall-modeled large-eddy simulation of three-dimensional turbulent\n boundary layer in a bent square duct: We conduct wall-modeled LES (WMLES) of a pressure-driven three-dimensional\nturbulent boundary layer (3DTBL) developing on the floor of a bent square duct\nto investigate the predictive capability of three widely used wall models,\nnamely, a simple equilibrium stress model, an integral nonequilibrium model,\nand a PDE nonequilibrium model. The numerical results are compared with the\nexperiment of Schwarz and Bradshaw (J. Fluid Mech. (1994), vol. 272, pp.\n183-210). While the wall-stress magnitudes predicted by the three wall models\nare comparable, the PDE nonequilibrium wall model produces a substantially more\naccurate prediction of the wall-stress direction, followed by the integral\nnonequilibrium wall model. The wall-stress direction from the wall models is\nshown to have separable contributions from the equilibrium stress part and the\nintegrated nonequilibrium effects, where how the latter is modeled differs\namong the wall models. The triangular plot of the wall-model solution reveals\ndifferent capabilities of the wall models in representing variation of flow\ndirection along the wall-normal direction. On the contrary, the outer LES\nsolution is unaffected by the type of wall model used, resulting in nearly\nidentical predictions of the mean and turbulent statistics in the outer region\nfor all the wall models. This is explained by the vorticity dynamics and the\ninviscid skewing mechanism of generating the mean three-dimensionality.\nFinally, the LES solution in the outer layer is used to study the anisotropy of\nturbulence. In contrast to the canonical 2D wall turbulence, the Reynolds\nstress anisotropy exhibit strong non-monotonic behavior with increasing wall\ndistance.", "category": "physics_flu-dyn" }, { "text": "Wetting Splashing: We present fluid dynamics videos illustrating wetting splashing-produced by\nwater drop impact onto hydrophobic microstructures at high impact velocity\n($\\sim 3$ ms$^{-1}$). The substrate consists of regular and transparent\nmicrotextures in square or hexagonal lattice, yielding a high contact angle of\n$\\sim 150 ^\\circ$. Our high speed top-or-bottom view recordings through the\ntransparent surface shed light on the solid-liquid-air interfaces at impact.\nDespite the superhydrophobicity of the latticed micropillars (of a periodicity\nof $10 {\\mu}m), water droplet wets a certain central area and moreover entraps\nan air bubble beneath the droplet. Besides the central wet area, lamella surf\non air splashing outward at high impinging velocity. The effects of\nmicropatterns and air pressure on the impact outcome are also examined. We show\nthat microscopic boundary condition, imposed by the solid texture, profoundly\ninfluences the macroscopic flow dynamics upon superhydrophobic surfaces at high\nimpinging velocity. In addition, the intervening air between the liquid and the\nsolid plays a crucial role in directional splash, which can be eliminated by a\nreduced air pressure.", "category": "physics_flu-dyn" }, { "text": "Universality of oscillatory instabilities in fluid mechanical systems: Oscillatory instability (OI) emerges amidst turbulent states in experiments\nin various turbulent fluid and thermo-fluid systems such as aero-acoustic,\nthermoacoustic and aeroelastic systems. For the time series of the relevant\ndynamic variable at the onset of the OI, universal scaling behavior have been\ndiscovered in experiments via the Hurst exponent and certain spectral measures.\nBy means of a center manifold reduction, the spatiotemporal dynamics of these\nreal systems can be mapped to a complex Ginzburg-Landau equation with a linear\nglobal coupling (GCGLE). In this work, we show that the GCGLE is able to\ncapture the universal behavior of the route to OI, elucidating it as a\ntransition from defect to phase turbulence mediated by the global coupling.", "category": "physics_flu-dyn" }, { "text": "From the butterfly effect to intrinsic randomness: the spontaneous\n growth of singular shear flows: The butterfly effect is today commonly identified with the sensitive\ndependence of deterministic chaotic systems upon initial conditions. However,\nthis is only one facet of the notion of unpredictability pioneered by Lorenz,\nwho actually predicted that multiscale fluid flows could spontaneously lose\ntheir deterministic nature and become intrinsically random. This effect, which\nis radically different from chaos, have remained out of reach for detailed\nphysical observations. Here, we substantiate this scenario by showing that it\nis inherent to the elementary Kelvin--Helmholtz hydrodynamical instability of\nan initially singular shear layer. We moreover provide evidence that the\nresulting macroscopic flow displays universal statistical properties that are\ntriggered by, but independent of specific physical properties at micro-scales.\nThis spontaneous stochasticity is interpreted as an Eulerian counterpart to\nRichardson's relative dispersion of Lagrangian particles, giving substance to\nthe intrinsic nature of randomness in turbulence.", "category": "physics_flu-dyn" }, { "text": "Linking turbulent waves and bubble diffusion in self-aerated\n open-channel flows: Two-state air concentration: High Froude-number flows become self-aerated when the destabilizing effect of\nturbulence overcomes gravity and surface tension forces. Traditionally, the\nresulting air concentration profile has been explained using single-layer\napproaches that invoke solutions of the advection-diffusion equation for air in\nwater, i.e., bubbles' dispersion. Based on a wide range of experimental\nevidences, we argue that the complete air concentration profile shall be\nexplained through the weak interaction of different canonical turbulent flows,\nnamely a Turbulent Boundary Layer (TBL) and a Turbulent Wavy Layer (TWL).\nMotivated by a decomposition of the streamwise velocity into a pure wall flow\nand a free-stream flow [Krug et al., J. Fluid Mech. (2017), vol. 811, pp.\n421--435], we present a physically consistent two-state formulation of the\nstructure of a self-aerated flow. The air concentration is mathematically built\nupon a modified Rouse profile and a Gaussian error function, resembling\nvertical mass transport in the TBL and the TWL. We apply our air concentration\ntheory to over 500 profiles from different data sets, featuring excellent\nagreement. Finally, we show that the turbulent Schmidt number, characterizing\nthe momentum-mass transfer, ranges between 0.2 to 1, which is consistent with\nprevious mass-transfer experiments in TBLs. Altogether, the proposed flow\nconceptualization sets the scene for more physically-based numerical modelling\nof turbulent mass diffusion in self-aerated flows.", "category": "physics_flu-dyn" }, { "text": "A high-order diffused-interface approach for two-phase compressible flow\n simulations using a Discontinuous Galerkin framework: A diffused-interface approach based on the Allen-Cahn phase field equation is\ndeveloped within a high-order Discontinuous Galerkin framework. The interface\ncapturing technique is based on the balance between explicit diffusion and\nsharpening terms in the phase field equation, where the former term involves\nthe computation of the local interface normal vectors. Due to the well-known\nGibbs phenomenon encountered in high-order discretisations of steep profiles\nsuch as shocks and/or interfaces, the accurate evaluation of the normal vector\nrequires special consideration. To this end, a non-linear preconditioning\nstrategy is proposed in this work where an additional smooth level-set function\nadvected by the velocity field is used for the evaluation of the normal\nvectors. It is shown that for appropriate choices of numerical fluxes and\nparameters of the model, the phase field remains bounded without any need for\nexplicit regularisation. The proposed diffused-interface technique is\nimplemented within a five equation model for fully compressible two-phase\nflows. In order to preserve isolated interfaces, a quasi-conservative\ndiscretisation of the five equation model is employed. A series of numerical\nexperiments of increasing complexity are performed in order to assess the\naccuracy and robustness of the developed methodology, including two-phase flows\ninvolving viscous effects, gravitational forces, and surface tension.", "category": "physics_flu-dyn" }, { "text": "The physics of stripe patterns in turbulent channel flow determined by\n DNS results: The turbulent flow in an infinitely extended plane channel is analysed by\nsolving the Navier-Stokes equations with a DNS approach. Solutions are obtained\nin a numerical solution domain of finite size in the streamwise as well as in\nthe lateral direction setting periodic boundary conditions in both directions.\nTheir impact on large scale structures in the turbulent flow field is analysed\ncarefully in order to avoid their suppression. When this is done appropriately\nwell known stripe patterns in these flows can be observed and analysed\nespecially with respect to their relative motion compared to the mean flow\nvelocity. Various details of this stripe pattern dominated velocity field are\nshown. Also global parameters like the friction factor in the flow field and\nthe Nusselt number in the temperature field are determined based on the\nstatistics of the flow and temperature data in a very large time period that\nguarantees fully developed turbulent flow and heat transfer.", "category": "physics_flu-dyn" }, { "text": "Nonconforming Schwarz-Spectral Element Methods For Incompressible Flow: We present scalable implementations of spectral-element-based Schwarz\noverlapping (overset) methods for the incompressible Navier-Stokes (NS)\nequations. Our SEM-based overset grid method is implemented at the level of the\nNS equations, which are advanced independently within separate subdomains using\ninterdomain velocity and pressure boundary-data exchanges at each timestep or\nsub-timestep. Central to this implementation is a general, robust, and scalable\ninterpolation routine, {\\em gslib-findpts}, that rapidly determines the\ncomputational coordinates (processor $p$, element number $e$, and local\ncoordinates $(r,s,t) \\in \\hat{\\Omega} := [-1,1]^3$) for any arbitrary point\n$\\mathbf{x}^* =(x^*,y^*,z^*) \\in \\Omega \\subset {\\rm I\\!R}^3$. The\ncommunication kernels in $gslib$ execute with at most $\\log P$ complexity for\n$P$ MPI ranks, have scaled to $P > 10^6$, and obviate the need for development\nof any additional MPI-based code for the Schwarz implementation. The original\ninterpolation routine has been extended to account for multiple overlapping\ndomains. The new implementation discriminates the possessing subdomain by\ndistance to the domain boundary, such that the interface boundary data is taken\nfrom the inner-most interior points. We present application of this approach to\nseveral heat transfer and fluid dynamic problems, discuss the\ncomputation/communication complexity and accuracy of the approach, and present\nperformance measurements for $P > 12,000$.", "category": "physics_flu-dyn" }, { "text": "Two-dimensional turbulence in a finite box: We present theory of two-dimensional turbulence excited by an external force\nin thin fluid films on scales larger than the film thickness. The principal\nfeature of two-dimensional turbulence is the tendency of producing motions of\nlarger and larger scales thanks to the nonlinear interaction. The tendency\nleads to formation of the so-called inverse cascade and, at some conditions, of\nbig coherent vortices. We discuss the mean velocity profile of the coherent\nvortices and the flow fluctuations on the background of the mean velocity for\ndifferent regimes. We demonstrate that the regime of strongly interacting\nfluctuations leads to an anisotropic scaling inside the coherent vortices.", "category": "physics_flu-dyn" }, { "text": "Bounds on the mixing enhancement for a stirred binary fluid: The Cahn-Hilliard equation describes phase separation in binary liquids. Here\nwe study this equation with spatially-varying sources and stirring, or\nadvection. We specialize to symmetric mixtures and time-independent sources and\ndiscuss stirring strategies that homogenize the binary fluid. By measuring\nfluctuations of the composition away from its mean value, we quantify the\namount of homogenization achievable. We find upper and lower bounds on our\nmeasure of homogenization using only the Cahn-Hilliard equation and the\nincompressibility of the advecting flow. We compare these theoretical bounds\nwith numerical simulations for two model flows: the constant flow, and the\nrandom-phase sine flow. Using the sine flow as an example, we show how our\nbounds on composition fluctuations provide a measure of the effectiveness of a\ngiven stirring protocol in homogenizing a phase-separating binary fluid.", "category": "physics_flu-dyn" }, { "text": "The Effect of Fluid Properties on the Operation of Thermal Bubble jet: Over the last two decades, since explosive boiling has been widely used in\nindustry, research on it has been increased. Thermal bubble jet printer, micro\ninjectors and using in micro medicine for injection are some possible\napplications. The operation of thermal bubble jets consist of three stages: 1-\nHeat transfer process, 2- Bubble formation and growth in the microchannel, and\n3-Drop ejection. The effect of fluid properties on the operation of thermal\nbubble jet is investigated in this paper. Thus the effect of fluid properties\nis investigated in each above mentioned processes. Eventually, comparing\nexperimental results of drop ejection with the results from simulations, drop\nproperties such as volume and velocity are given.", "category": "physics_flu-dyn" }, { "text": "Three-dimensional Vortex-induced Reaction Hotspots at Flow Intersections: We show the emergence of reaction hotspots induced by three-dimensional (3D)\nvortices with a simple $A+B \\rightarrow C$ reaction. We conduct microfluidics\nexperiments to visualize the spatial map of the reaction rate with the\nchemiluminescence reaction and cross-validate the results with direct numerical\nsimulations. 3D vortices form at spiral saddle type stagnation points, and the\n3D vortex flow topology is essential for initiating reaction hotspots. The\neffect of vortices on mixing and reaction becomes more vigorous for\nrough-walled channels, and our findings are valid over wide ranges of channel\ndimensions and Damk\\\"{o}hler numbers.", "category": "physics_flu-dyn" }, { "text": "Dynamics of particle-laden turbulent Couette flow. Part2: Modified\n fluctuating force model (M-FFS): Two-way coupled DNS simulation of particle-laden turbulent Couette-flow [1],\nin the volume fraction regime $\\phi>10^{-4}$, showed a discontinuous decrease\nof turbulence intensity beyond a critical volume fraction\n$\\phi_{cr}\\sim7.875\\times10^{-4}$. Due to the presence of high inertial\nparticles, the drastic reduction of shear production of turbulence is found to\nbe the main cause for the discontinuous attenuation of turbulence. In this\narticle, particle-phase statistics is explored. The two-way coupled DNS reveal\nthat the mean-square velocity profiles in cross-stream (y) and span-wise (z)\ndirections are flat and increase with $\\phi$ as the higher frequency of\ncollision helps in transferring streamwise momentum to span-wise and\nwall-normal directions. Whereas, streamwise fluctuations decrease and tend\nbecome flatter with increase in loading. In the regime with $\\phi>\\phi_{cr}$,\nthe particle velocity fluctuations drive the fluid phase velocity fluctuations.\nAdditionally it is observed that one-way coupled DNS and Fluctuating Force\nSimulation (FFS) [2] are capable to predict the particle phase statistics with\nreasonable accuracy in the regime $\\phi<\\phi_{cr}$ where wall-particle\ncollision time and inter-particle collision time is lesser than viscous\nrelaxation time of the particles. For, $\\phi>\\phi_{cr}$, a significant error in\nthe prediction from one-way coupled DNS and FFS is observed due to the\nlimitation of FFS in capturing the turbulence attenuation and the change in\nmean fluid velocity profile. A modified FFS model (M-FFS) is successfully\ndeveloped in this article with modified mean fluid velocity profile and\nzero-diffusivity.", "category": "physics_flu-dyn" }, { "text": "Nonlinear variation of bedload thickness with fluid flow rate in laminar\n shearing flow: The movement of subaqueous sediment in laminar shearing flow is numerically\ninvestigated by the coupled lattice Boltzmann and discrete element methods.\nFirst, the numerical method is validated by comparing the phase diagram\nproposed by Ouriemi et al. ({\\it J. Fluid Mech}., vol. 636, 2009, pp. 321-336).\nSecond, a detailed study on sediment movement is performed for sediment with\nvarying solid volume fractions, and a nonlinear relationship between the\nnormalised thickness of the mobile layer and the normalised fluid flow rate is\nobserved for a densely-packed sediment. Third, an independent investigation on\nthe effective viscosity and friction coefficient of the sediment under\ndifferent fluid flow rates is conducted in a shear cell; and substitution of\nthese two critical parameters into a theoretical expression proposed by\nAussillous et al. ({\\it J. Fluid Mech}., vol. 736, 2013, pp. 594-615) provides\nconsistent predictions of bedload thickness with the simulation results of\nsediment movement. Therefore, we conclude that the non-Newtonian behaviour of\ndensely-packed sediment leads to the nonlinear relationship between the\nnormalised thickness of the mobile layer and the normalised fluid flow rate.", "category": "physics_flu-dyn" }, { "text": "A comparative study of two Allen-Cahn models for immiscible $N$-phase\n flows by using a consistent and conservative lattice Boltzmann method: In this work, we conduct a detailed comparison between two second-order\nconservative Allen-Cahn (AC) models [\\emph{Model A}: Zheng \\emph{et al.}, Phys.\nRev. E 101, 0433202 (2020) and \\emph{Model B}: Mirjalili and Mani, (2023)] for\nthe immiscible $N$-phase flows. Mathematically, these two AC equations can be\nproved to be equivalent under some approximate conditions. However, the effects\nof these approximations are unclear from the theoretical point of view, and\nwould be considered numerically. To this end, we propose a consistent and\nconservative lattice Boltzmann method for the AC models for $N$-phase flows,\nand present some numerical comparisons of accuracy and stability between these\ntwo AC models. The results show that both two AC models have good performances\nin accuracy, but the \\emph{Model B} is more stable for the realistic complex\n$N$-phase flows, although there is an adjustable parameter in the \\emph{Model\nA}.", "category": "physics_flu-dyn" }, { "text": "The transient swimming of a waving sheet: Small-scale locomotion plays an important role in biology. Different\nmodelling approaches have been proposed in the past. The simplest model is an\ninfinite inextensible two-dimensional waving sheet, {originally introduced by\nTaylor}, which serves as an idealized geometrical model for both spermatozoa\nlocomotion and ciliary transport in Stokes flow. Here we complement classic\nsteady-state calculations by deriving the transient low-Reynolds number\nswimming speed of such a waving sheet when starting from rest (small-amplitude\ninitial-value problem). We also determine the transient fluid flow in the\n`pumping' setup where the sheet is not free to move but instead generates a net\nfluid flow around it. The time scales for these two problems, which in general\ngovern transient effects in transport and locomotion at low Reynolds numbers,\nare also derived using physical arguments.", "category": "physics_flu-dyn" }, { "text": "Implications of inertial subrange scaling for stably stratified mixing: The effects of turbulent dynamic range on scalar mixing in stably stratified\nturbulence are investigated by an adaptation of the theoretical passive scalar\nmodelling arguments of Beguier et al. (1978) and demonstrated statistically\nusing direct numerical simulations of statistically stationary homogeneous\nstratified and sheared turbulence (SHSST). By analysis of inertial and\ninertial-convective subrange scaling, we show that the relationship between\nactive scalar and turbulence time scales is predicted by the ratio of the\nKolmogorov and Oboukhov-Corrsin constants provided there is sufficient scale\nseparation for inertial and inertial-convective subrange scalings to be valid.\nWith this analysis, we show that the turbulent mixing coefficient, $\\Gamma\n\\equiv \\chi/\\epsilon$, that is, within this context defined to be the ratio of\navailable potential energy ($E_p$) and turbulent kinetic energy ($E_k$)\ndissipation rates, can be estimated by $E_p,E_k$ and a universal constant\nprovided a Reynolds number is sufficiently high, observed here at $Re_b \\equiv\n\\epsilon / \\nu N^2 \\gtrapprox 300$ where $\\nu$ is the kinematic viscosity and\n$N$ is the characteristic buoyancy frequency. We propose a model for diapycnal\nmixing with robust theoretical parametrisation and asymptotic behaviour in this\nhigh-$Re_b$ limit.", "category": "physics_flu-dyn" }, { "text": "Benchmark modeling and 3D applications of solidification and\n macro-segregation based on an operator-splitting and fully decoupled scheme\n with term-wise matrix assembly: The solidification and macro-segregation problem involving unsteady\nmulti-physics and multi-phase fields is typically a complex process with mass,\nmomentum, heat, and species transfers among solid, mushy, and liquid phase\nregions. The quantitative prediction of phase change, chemical heterogeneities,\nand multi-phase and multi-component flows plays critical roles in many natural\nscenarios and industrial applications that involve many disciplines, like\nmaterial, energy, and even planet science. In view of this, some scholars and\nresearch institutions have called for more contributors to join the benchmark\nanalysis of solidification and segregation problems. Our work proposes an\noperator-splitting and matrix-based method to avoid non-linear systems. Also,\nthe combination of vectorization and forward equation-based matrix assembly\ntechniques enhances the implementability of extensions of 3D applications.\nLastly, the novel scheme is well validated through a bunch of 2D and 3D\nbenchmark cases. The numerical results also illustrate that this method can\nensure accurate prediction and adequately capture the physical details of\nphenomena caused by the solutally and thermally driven flow, which include\nchannel segregation, the formation of freckles, edge effect, aspect ratio\neffect, and 3D effect.", "category": "physics_flu-dyn" }, { "text": "Large Eddy Simulations of bubbly flows and breaking waves with Smoothed\n Particle Hydrodynamics: For turbulent bubbly flows, multi-phase simulations resolving both the liquid\nand bubbles are prohibitively expensive in the context of different natural\nphenomena. One example is breaking waves, where bubbles strongly influence wave\nimpact loads, acoustic emissions, and atmospheric-ocean transfer, but detailed\nsimulations in all but the simplest settings are infeasible. An alternative\napproach is to resolve only large scales, and model small scale bubbles\nadopting sub-resolution closures. Here we introduce a large eddy simulation\n(LES) Smoothed Particle Hydrodynamics (SPH) scheme for simulations of bubbly\nflows. The continuous liquid phase is resolved with a semi-implicit\nisothermally compressible SPH framework. This is coupled with a discrete\nLagrangian bubble model. Bubbles and liquid interact via exchanges of volume\nand momentum, through turbulent closures, bubble breakup and entrainment, and\nfree-surface interaction models. By representing bubbles as individual\nparticles, they can be tracked over their lifetimes, allowing closure models\nfor sub-resolution fluctuations, bubble deformation, breakup and free-surface\ninteraction in integral form, accounting for the finite timescales over which\nthese events occur. We investigate two flows: bubble plumes, and breaking\nwaves, and find close quantitative agreement with published experimental and\nnumerical data. In particular, for plunging breaking waves, our framework\naccurately predicts the Hinze scale, bubble size distribution, and growth rate\nof the entrained bubble population. This is the first coupling of an SPH\nframework with a discrete bubble model, with potential for cost effective\nsimulations of wave-structure interactions and more accurate predictions of\nwave impact loads.", "category": "physics_flu-dyn" }, { "text": "The folding motion of an axisymmetric jet of wormlike-micelles solution: The problem of buckling and coiling of jets of viscous, Newtonian liquids has\nreceived a substantial level of attention over the past two decades, both from\nexperimental and theoretical points of view. Nevertheless, many industrial\nfluids and consumer products are non-Newtonian, and their rheological\nproperties affect their flow behavior. The present work aims at studying the\ndynamics of cylindrical jets of a viscoelastic, shear-thinning solution of\ncetylpyridinium salt (CPyCl). In concentrated solutions, CPyCl surfactant\nmolecules have been shown to assemble in long wormlike micellar structures,\nwhich gives the fluid its non-Newtonian properties. Jets of this fluid show\nnovel features compared to their Newtonian counterparts, including a type of\nmotion, in which the jet folds back and forth on itself in a fashion similar to\nsheets of viscous fluids, instead of coiling around the vertical axis as\ncylindrical Newtonian jets do. Another novel feature of CPyCl micellar fluid\njets is a widening of the jet above the plate reminiscent of the die-swell\nphenomenon that we call \\emph{reverse swell}. We propose physical mechanisms\nfor both folding and reverse swell, and compare theoretical predictions to\nexperimental measurements. In addition, we systematically explore different\nflow regimes in the parameter space of the height of fall and flow rate and\ncompare regime maps of a CPyCl micellar solution and a Newtonian silicone oil.", "category": "physics_flu-dyn" }, { "text": "New approach for analysing dynamical processes on the surface of\n photospheric vortex tubes: The majority of studies on multi-scale vortex motions employ a\ntwo-dimensional geometry by using a variety of observational and numerical\ndata. This approach limits the understanding the nature of physical processes\nresponsible for vortex dynamics. Here we develop a new methodology to extract\nessential information from the boundary surface of vortex tubes. 3D\nhigh-resolution magnetoconvection MURaM numerical data has been used to analyse\nphotospheric intergranular velocity vortices. The Lagrangian Averaged Vorticity\nDeviation (LAVD) technique was applied to define the centers of vortex\nstructures and their boundary surfaces based on the advection of fluid\nelements. These surfaces were mapped onto a constructed envelope grid that\nallows the study of the key plasma parameters as functions of space and time.\nQuantities that help in understanding the dynamics of the plasma, e.g. Lorentz\nforce, pressure force, plasma-$\\beta$ were also determined. Our results suggest\nthat, while density and pressure have a rather global behaviour, the other\nphysical quantities undergo local changes, with their magnitude and orientation\nchanging in space and time. At the surface, the mixing in the horizontal\ndirection is not efficient, leading to appearance of localized regions with\nhigher/colder temperatures. In addition, the analysis of the MHD Poynting flux\nconfirms that the majority of the energy is directed in the horizontal\ndirection. Our findings also indicate that the pressure and magnetic forces\nthat drive the dynamics of the plasma on vortex surfaces are unbalanced and\ntherefore the vortices do not rotate as a rigid body.", "category": "physics_flu-dyn" }, { "text": "The onset of turbulence in particle-laden pipe flows: We propose a scaling law for the onset of turbulence in pipe flow of\nneutrally buoyant suspensions. This scaling law, based on a large set of\nexperimental data, relates the amplitude of the particle-induced perturbations\n($\\epsilon$) to the critical suspension Reynolds number, $Re_{s,c}$. Here\n$\\epsilon$ is a function of the particle-to-pipe diameter ratio and the volume\nfraction of the suspended particles, $\\epsilon = (d/D)^{1/2} \\phi^{1/6}$.\n$Re_{s,c}$ is found to scale as $\\epsilon^{-1}$. Furthermore, the perturbation\namplitude allows a distinction between classical, intermediate and\nparticle-induced transition.", "category": "physics_flu-dyn" }, { "text": "The deformation of a flexible fiber settling in a quiescent viscous\n fluid: The equilibrium state of a flexible fiber settling in a viscous fluid is\nexamined using a combination of macroscopic experiments, numerical simulations\nand scaling arguments. We identify three regimes having different signatures on\nthis equilibrium configuration of the elastic filament: weak and large\ndeformation regimes wherein the drag is proportional to the settling velocity\nas expected in Stokes flow and an intermediate elastic reconfiguration regime\nwhere the filament deforms to adopt a shape with a smaller drag which is no\nlonger linearly proportional to the velocity.", "category": "physics_flu-dyn" }, { "text": "Speech can produce jet-like transport relevant to asymptomatic spreading\n of virus: Many scientific reports document that asymptomatic and presymptomatic\nindividuals contribute to the spread of COVID-19, probably during conversations\nin social interactions. Droplet emission occurs during speech, yet few studies\ndocument the flow to provide the transport mechanism. This lack of\nunderstanding prevents informed public health guidance for risk reduction and\nmitigation strategies, e.g. the \"six-foot rule\". Here we analyze flows during\nbreathing and speaking, including phonetic features, using order-of-magnitudes\nestimates, numerical simulations, and laboratory experiments. We document the\nspatio-temporal structure of the expelled air flow. Phonetic characteristics of\nplosive sounds like 'P' lead to enhanced directed transport, including jet-like\nflows that entrain the surrounding air. We highlight three distinct temporal\nscaling laws for the transport distance of exhaled material including (i)\ntransport over a short distance ($<$ 0.5 m) in a fraction of a second, with\nlarge angular variations due to the complexity of speech, (ii) a longer\ndistance, approximately 1 m, where directed transport is driven by individual\nvortical puffs corresponding to plosive sounds, and (iii) a distance out to\nabout 2 m, or even further, where sequential plosives in a sentence,\ncorresponding effectively to a train of puffs, create conical, jet-like flows.\nThe latter dictates the long-time transport in a conversation. We believe that\nthis work will inform thinking about the role of ventilation, aerosol transport\nin disease transmission for humans and other animals, and yield a better\nunderstanding of linguistic aerodynamics, i.e., aerophonetics.", "category": "physics_flu-dyn" }, { "text": "Possibility of Turbulence from a Post-Navier-Stokes Equation: We introduce corrections to the Navier-Stokes equation arising from the\ntransitions between molecular states and the injection of external energy. In\nthe simplest application of the proposed post Navier-Stokes equation, we find a\nmulti-valued velocity field and the immediate possibility of velocity reversal,\nboth features of turbulence.", "category": "physics_flu-dyn" }, { "text": "Rapid growth of cloud droplets by turbulence: Assuming perfect collision efficiency, we demonstrate that turbulence can\ninitiate and sustain the rapid growth of very small water droplets in air even\nwhen these droplets are too small to cluster, and even without having to take\ngravity and small-scale intermittency into account. This is because the range\nof local Stokes numbers of identical droplets in the turbulent flow field is\nbroad enough even when small-scale intermittency is neglected. This\ndemonstration is given for turbulence which is one order of magnitude less\nintense than is typical in warm clouds but with a volume fraction which, even\nthough small, is nevertheless large enough for an estimated a priori frequency\nof collisions to be ten times larger than in warm clouds. However, the time of\ngrowth in these conditions turns out to be one order of magnitude smaller than\nin warm clouds.", "category": "physics_flu-dyn" }, { "text": "Amplification and Nonlinear Mechanisms in Plane Couette Flow: We study the input-output response of a streamwise constant projection of the\nNavier-Stokes equations for plane Couette flow, the so-called 2D/3C model.\nStudy of a streamwise constant model is motivated by numerical and experimental\nobservations that suggest the prevalence and importance of streamwise and\nquasi-streamwise elongated structures. Periodic spanwise/wall-normal (z-y)\nplane stream functions are used as input to develop a forced 2D/3C streamwise\nvelocity field that is qualitatively similar to a fully turbulent spatial field\nof DNS data. The input-output response associated with the 2D/3C nonlinear\ncoupling is used to estimate the energy optimal spanwise wavelength over a\nrange of Reynolds numbers. The results of the input-output analysis agree with\nprevious studies of the linearized Navier-Stokes equations. The optimal energy\ncorresponds to minimal nonlinear coupling. On the other hand, the nature of the\nforced 2D/3C streamwise velocity field provides evidence that the nonlinear\ncoupling in the 2D/3C model is responsible for creating the well known\ncharacteristic \"S\" shaped turbulent velocity profile. This indicates that there\nis an important tradeoff between energy amplification, which is primarily\nlinear and the seemingly nonlinear momentum transfer mechanism that produces a\nturbulent-like mean profile.", "category": "physics_flu-dyn" }, { "text": "Using multiscale norms to quantify mixing and transport: Mixing is relevant to many areas of science and engineering, including the\npharmaceutical and food industries, oceanography, atmospheric sciences, and\ncivil engineering. In all these situations one goal is to quantify and often\nthen to improve the degree of homogenisation of a substance being stirred,\nreferred to as a passive scalar or tracer. A classical measure of mixing is the\nvariance of the concentration of the scalar, which can be related to the $L^2$\nnorm of the concentration field. Recently other norms have been used to\nquantify mixing, in particular the mix-norm as well as negative Sobolev norms.\nThese norms have the advantage that unlike variance they decay even in the\nabsence of diffusion, and their decay corresponds to the flow being mixing in\nthe sense of ergodic theory. General Sobolev norms weigh scalar gradients\ndifferently, and are known as multiscale norms for mixing. We review the\napplications of such norms to mixing and transport, and show how they can be\nused to optimise the stirring and mixing of a decaying passive scalar. We then\nreview recent work on the less-studied case of a continuously-replenished\nscalar field --- the source-sink problem. In that case the flows that optimally\nreduce the norms are associated with transport rather than mixing: they push\nsources onto sinks, and vice versa.", "category": "physics_flu-dyn" }, { "text": "Microscopic statistical description of incompressible Navier-Stokes\n granular fluids: Based on the recently-established Master kinetic equation and related Master\nconstant H-theorem which describe the statistical behavior of the\nBoltzmann-Sinai classical dynamical system for smooth and hard spherical\nparticles, the problem is posed of determining a microscopic statistical\ndescription holding for an incompressible Navier-Stokes fluid. The goal is\nreached by introducing a suitable mean-field interaction in the Master kinetic\nequation. The resulting Modified Master Kinetic Equation (MMKE) is proved to\nwarrant at the same time the condition of mass-density incompressibility and\nthe validity of the Navier-Stokes fluid equation. In addition, it is shown that\nthe conservation of the Boltzmann-Shannon entropy can similarly be warranted.\nApplications to the plane Couette and Poiseuille flows are considered showing\nthat they can be regarded as final decaying states for suitable non-stationary\nflows. As a result, it is shown that an arbitrary initial stochastic $1-$body\nPDF evolving in time by means of MMKE necessarily exhibits the phenomenon of\nDecay to Kinetic Equilibrium (DKE), whereby the $1-$body PDF asymptotically\nrelaxes to a stationary and spatially-uniform Maxwellian PDF.", "category": "physics_flu-dyn" }, { "text": "Applications of biorthogonal decompositions in fluid-structure\n interactions: This paper is dedicated to the study of the orthogonal decomposition of\nspatially and temporally distributed signals in fluid-structure interaction\nproblems. First application is concerned with the analysis of wall-pressure\ndistributions over bluff bodies. The need for such a tool is increasing due to\nthe progress in data acquisition systems and in computational fluid dynamics.\nThe classical proper orthogonal decomposition (POD) method is discussed, and it\nis shown that heterogeneity of the mean pressure over the structure induces\ndifficulties in the physical interpretation. It is then proposed to use the\nbiorthogonal decomposition (BOD) technique instead; although it appears similar\nto POD, it is more general and fundamentally different since this tool is\ndeterministic rather than statistical. The BOD method is described and adapted\nto wall-pressure distribution, with emphasis on aerodynamic load decomposition.\nThe second application is devoted to the generation of a spatially correlated\nwind velocity field which can be used for the temporal calculation of the\naeroelastic behaviour of structures such as bridges. In this application, the\nspace-time symmetry of the BOD method is absolutely necessary. Examples are\nprovided in order to illustrate and show the satisfactory performance and the\ninterest of the method. Extensions to other fluid-structure problems are\nsuggested.", "category": "physics_flu-dyn" }, { "text": "Are Defect Profile Similarity Criteria Different Than Velocity Profile\n Similarity Criteria for the Turbulent Boundary Layer?: The use of the defect profile instead of the experimentally observed velocity\nprofile for the search for similarity parameters has become firmly imbedded in\nthe turbulent boundary layer literature. However, a search of the literature\nreveals that there are no theoretical reasons for this defect profile\npreference over the more traditional velocity profile. In the report herein, we\nuse the flow governing equation approach to develop similarity criteria for the\ntwo profiles. Results show that the derived similarity criteria are identical.\nTogether with previous work that found that defect profile similarity must be\naccompanied by velocity profile similarity, then ones expectations must be that\neither profile can be used to search for similarity in experimental datasets.\nThe choice should therefore be dictated by which one works best for\nexperimental investigations, which in this case is the velocity profile.", "category": "physics_flu-dyn" }, { "text": "Thermalization in the Lenard-Jones gas: In this letter, using energy transfers, we demonstrate a route to\nthermalization in an isolated ensemble of realistic gas particles. We performed\na grid-free classical molecular dynamics simulation of two-dimensional\nLenard-Jones gas. We start our simulation with a large-scale vortex akin to a\nhydrodynamic flow and study its non-equilibrium behavior till it attains\nthermal equilibrium. In the intermediate phases, small wavenumbers ($k$)\nexhibit $E(k) \\propto k^{-3}$ kinetic energy spectrum whereas large wavenumbers\nexhibit $E(k) \\propto k$ spectrum. Asymptotically, $E(k) \\propto k$ for the\nwhole range of $k$, thus indicating thermalization. These results are akin to\nthose of Euler turbulence despite complex collisions and interactions among the\nparticles.", "category": "physics_flu-dyn" }, { "text": "An acoustic space-time and the Lorentz transformation in aeroacoustics: In this paper we introduce concepts from relativity and geometric algebra to\naeroacoustics. We do this using an acoustic space-time transformation within\nthe framework of sound propagation in uniform flows. By using Geometric Algebra\nwe are able to provide a simple geometric interpretation to the space-time\ntransformation, and are able to give neat and lucid derivations of the\nfree-field Green's function for the convected wave equation and the Doppler\nshift for a stationary observer and a source in uniform rectilinear motion in a\nuniform flow.", "category": "physics_flu-dyn" }, { "text": "Near field magnetostatics and Neel Brownian interactions mediated\n magnetorheological characteristics of highly stable nano ferrocolloids: Magnetic nanocolloids with synthesized super paramagnetic Fe3O4 nanoparticles\n(SPION) (5 to 15 nm) dispersed in insitu developed Polyethylene Glycol (PEG\n400) and nano silica complex have been synthesized. The PEG nano Silica complex\nphysically encapsulates the SPIONs, ensuring no phase separation under high\nmagnetic fields (1.2 T). Exhaustive magnetorheological investigations have been\nperformed to comprehend the structural behavior and response of the\nferrocolloids. Remarkable stability and reversibility have been observed under\nmagnetic field for concentrated systems. The results exhibit the impact of\nparticle concentration, size and encapsulation efficiency on parameters such as\nshear viscosity, yield stress, viscoelastic moduli, magnetoviscous hysteresis\netc. Analytical models to reveal the system mechanism and mathematically\npredict the magnetoviscosity and magneto yield stress has been theorized. The\nmechanistic approach based on near field magnetostatics and Neel Brownian\ninteractivities can predict the colloidal properties under the effect of field\naccurately. The colloid exhibits amplified storage and loss moduli alongside\nhighly augmented linear viscoelastic region under magnetic stimuli. The\ntransition of the colloidal state from fluidic phase to soft condensed phase\nand its viscoelastic stimuli under the influence of magnetic field has been\nexplained based on the mathematical analysis. The remarkable stability,\nmagnetic properties and accurate physical models reveal good promise for the\ncolloids in transient situations viz. magneto MEMS/ NEMS devices, antiseismic\ndamping, biomedical invasive treatments etc.", "category": "physics_flu-dyn" }, { "text": "Spatio-temporal correlation functions in scalar turbulence from\n functional renormalization group: We provide the leading behavior at large wavenumbers of the two-point\ncorrelation function of a scalar field passively advected by a turbulent flow.\nWe first consider the Kraichnan model, in which the turbulent carrier flow is\nmodeled by a stochastic vector field with a Gaussian distribution, and then a\nscalar advected by a homogeneous and isotropic turbulent flow described by the\nNavier-Stokes equation, under the assumption that the scalar is passive, i.e.\nthat it does not affect the carrier flow. We show that at large wavenumbers,\nthe two-point correlation function of the scalar in the Kraichnan model decays\nas an exponential in the time delay, in both the inertial and dissipation\nranges. We establish the expression, both from a perturbative and from a\nnonperturbative calculation, of the prefactor, which is found to be always\nproportional to $k^2$. For a real scalar, the decay is Gaussian in $t$ at small\ntime delays, and it crosses over to an exponential only at large $t$. The\nassumption of delta-correlation in time of the stochastic velocity field in the\nKraichnan model hence significantly alters the statistical temporal behavior of\nthe scalar at small times.", "category": "physics_flu-dyn" }, { "text": "Influence of adversarial training on super-resolution turbulence\n reconstruction: Supervised super-resolution deep convolutional neural networks (CNNs) have\ngained significant attention for their potential in reconstructing velocity and\nscalar fields in turbulent flows. Despite their popularity, CNNs currently lack\nthe ability to accurately produce high-frequency and small-scale features, and\ntests of their generalizability to out-of-sample flows are not widespread.\nGenerative adversarial networks (GANs), which consist of two distinct neural\nnetworks (NNs), a generator and discriminator, are a promising alternative,\nallowing for both semi-supervised and unsupervised training. The difference in\nthe flow fields produced by these two NN architectures has not been thoroughly\ninvestigated, and a comprehensive understanding of the discriminator's role has\nyet to be developed. This study assesses the effectiveness of the unsupervised\nadversarial training in GANs for turbulence reconstruction in forced\nhomogeneous isotropic turbulence. GAN-based architectures are found to\noutperform supervised CNNs for turbulent flow reconstruction for in-sample\ncases. The reconstruction accuracy of both architectures diminishes for\nout-of-sample cases, though the GAN's discriminator network significantly\nimproves the generator's out-of-sample robustness using either an additional\nunsupervised training step with large eddy simulation input fields and a\ndynamic selection of the most suitable upsampling factor. These enhance the\ngenerator's ability to reconstruct small-scale gradients, turbulence\nintermittency, and velocity-gradient probability density functions. The\nextrapolation capability of the GAN-based model is demonstrated for\nout-of-sample flows at higher Reynolds numbers. Based on these findings,\nincorporating discriminator-based training is recommended to enhance the\nreconstruction capability of super-resolution CNNs.", "category": "physics_flu-dyn" }, { "text": "Anomalous Statistics and Large Deviations of Turbulent Water Waves past\n a Step: A computational strategy based on large deviation theory (LDT) is used to\nstudy the anomalous statistical features of turbulent surface waves propagating\npast an abrupt depth change created via a step in the bottom topography. The\ndynamics of the outgoing waves past the step are modeled using the truncated\nKorteweg-de Vries (TKdV) equation with random initial conditions at the step\ndrawn from the system's Gibbs invariant measure of the incoming waves. Within\nthe LDT framework, the probability distributions of the wave height can be\nobtained via the solution of a deterministic optimization problem. Detailed\nnumerical tests show that this approach accurately captures the non-Gaussian\nfeatures of the wave height distributions, in particular their asymmetric tails\nleading to high skewness. These calculations also give the spatio-temporal\npattern of the anomalous waves most responsible for these non-Gaussian\nfeatures. The strategy shows potential for a general class of nonlinear\nHamiltonian systems with highly non-Gaussian statistics.", "category": "physics_flu-dyn" }, { "text": "Phenomenology of disruptive breakup mechanism of a levitated evaporating\n emulsion droplet: Atomization of emulsion droplets is ubiquitous across a variety of\napplication domains ranging from NextGen combustors to fabrication of\nbiomedical implants. An understanding of the atomization mechanism in emulsions\ncan result in a paradigm shift in customized designs of efficient systems, be\nit in energy or biotechnology sectors. In this paper, we specifically study the\nbreakup mechanism of an evaporating contact-free (acoustic levitation) emulsion\ndroplet (water-oil) under external heating. Three distinct regimes are observed\nduring the lifespan of the evaporating droplet. Initially, the droplet diameter\nregresses linearly with time, followed by vapor bubble nucleation due to a\nsignificant difference in the boiling temperature among the components of the\nemulsion. The collapse of this bubble results in a high-intensity breakup of\nthe droplet leading to the propulsion of residual liquid in the form of a\ncrown-like sheet. The area of the expanding crown varies linearly with the\nsquare of the time. It is hypothesized that the expansion of the liquid sheet\ncentrifuges the larger water sub-droplets towards the edge, resulting in unique\nspatial segregation. Subsequently, we report the first observation of complex\npatches (representing water sub-droplets) and the rupture of the thin sheet\nadjacent to patches into holes (with hole growth rate ranging from 1.2 to 1.4\nm/s) in the context of an evaporating isolated emulsion droplet. The hole\nformation results in the creation of ligaments which undergo breakup into\nsecondary droplets with Sauter mean diameter (SMD) ~ 50 {\\mu}m.", "category": "physics_flu-dyn" }, { "text": "Evolution of inverse cascades and formation of precondensate in\n Gross-Pitaevskii turbulence in two dimensions: Here we study how coherence appears in a system driven by noise at small\nscales. In the wave turbulence modeled by the Gross-Pitaevskii / nonlinear\nSchr\\\"odinger equation, we observe states with correlation scales less than the\nsystem size but much larger than the excitation scale. We call such state\nprecondensate to distinguish it from condensate defined as a system-wide\ncoherent state. Both condensate and precondensate are characterized by large\nscale phase coherence and narrow distribution of amplitudes. When one excites\nsmall scales, precondensate is achieved relatively quickly by an inverse\ncascade heating quasi-equilibrium distribution of large-scale modes. The\ntransition from the precondensate to the system-wide condensate requires much\nlonger time. The spectra of precondensate differ from quasi-equilibrium and are\ncharacterized by two bending points, one on the scale of the average distance\nbetween vortex pairs, and the other on the scale of the distance between\nvortices in a pair. We suggest temporal evolution laws for both lengths and use\nthem to predict the probability of the transition to condensate.", "category": "physics_flu-dyn" }, { "text": "Roughness effects in turbulent forced convection: We conducted direct numerical simulations (DNSs) of turbulent flow over\nthree-dimensional sinusoidal roughness in a channel. A passive scalar is\npresent in the flow with Prandtl number $Pr=0.7$, to study heat transfer by\nforced convection over this rough surface. The minimal channel is used to\ncircumvent the high cost of simulating high Reynolds number flows, which\nenables a range of rough surfaces to be efficiently simulated. The near-wall\ntemperature profile in the minimal channel agrees well with that of the\nconventional full-span channel, indicating it can be readily used for\nheat-transfer studies at a much reduced cost compared to conventional DNS. As\nthe roughness Reynolds number, $k^+$, is increased, the Hama roughness\nfunction, $\\Delta U^+$, increases in the transitionally rough regime before\ntending towards the fully rough asymptote of $\\kappa_m^{-1}\\log(k^+)+C$, where\n$C$ is a constant that depends on the particular roughness geometry and\n$\\kappa_m\\approx0.4$ is the von K\\'arm\\'an constant. In this fully rough\nregime, the skin-friction coefficient is constant with bulk Reynolds number,\n$Re_b$. Meanwhile, the temperature difference between smooth- and rough-wall\nflows, $\\Delta\\Theta^+$, appears to tend towards a constant value,\n$\\Delta\\Theta^+_{FR}$. This corresponds to the Stanton number (the temperature\nanalogue of the skin-friction coefficient) monotonically decreasing with $Re_b$\nin the fully rough regime. Using shifted logarithmic velocity and temperature\nprofiles, the heat transfer law as described by the Stanton number in the fully\nrough regime can be derived once both the equivalent sand-grain roughness\n$k_s/k$ and the temperature difference $\\Delta \\Theta^+_{FR}$ are known. In\nmeteorology, this corresponds to the ratio of momentum and heat transfer\nroughness lengths, $z_{0m}/z_{0h}$, being linearly proportional to $z_{0m}^+$,\nthe momentum roughness length [continued]...", "category": "physics_flu-dyn" }, { "text": "Experimental Study on the Aerodynamic Sealing of Air Curtains: Controlling the air quality is of the utmost importance in today buildings.\nVertical air curtains are often used to separate two different climatic zones\nwith a view to reduce heat transfer. In fact, this research work proposes an\nair curtain aimed to ensure a proper separation between two zones, a clean one\nand a contaminated one. The methodology of this research includes: (i)\nsmall-scale tests on water models to ensure that the contamination does not\npass through the air curtain, and (ii) an analytical development integrating\nthe main physical characteristics of plane jets. In the solution developed, the\nairflow is extracted from the contaminated compartment to reduce the curtain\nairflow rejected to the exterior of the compartment. In this research work, it\nwas possible to determine the minimum exhaust flow necessary to ensure the\naerodynamic sealing of the air curtain. This article addresses the methodology\nused to perform the small-scale water tests and the corresponding results.", "category": "physics_flu-dyn" }, { "text": "Settling velocity of quasi-neutrally-buoyant inertial particles: We investigate the sedimentation properties of quasi-neutrally buoyant\ninertial particles carried by incompressible zero-mean fluid flows. We obtain\ngeneric formulae for the terminal velocity in generic space-and-time periodic\n(or steady) flows, along with further information for flows endowed with some\ndegree of spatial symmetry such as odd parity in the vertical direction. These\nexpressions consist in space-time integrals of auxiliary quantities which\nsatisfy partial differential equations of the advection--diffusion--reaction\ntype, that can be solved at least numerically since our scheme implies a huge\nreduction of the problem dimensionality from the full phase space to the\nclassical physical space.", "category": "physics_flu-dyn" }, { "text": "The fluctuation energy balance in non-suspended fluid-mediated particle\n transport: Here we compare two extreme regimes of non-suspended fluid-mediated particle\ntransport, transport in light and heavy fluids (\"saltation\" and \"bedload\",\nrespectively), regarding their particle fluctuation energy balance. From direct\nnumerical simulations, we surprisingly find that the ratio between collisional\nand fluid drag dissipation of fluctuation energy is significantly larger in\nsaltation than in bedload, even though the contribution of interparticle\ncollisions to transport of momentum and energy is much smaller in saltation due\nto the low concentration of particles in the transport layer. We conclude that\nthe much higher frequency of high-energy particle-bed impacts (\"splash\") in\nsaltation is the cause for this counter-intuitive behavior. Moreover, from a\ncomparison of these simulations to Particle Tracking Velocimetry measurements\nwhich we performed in a wind tunnel under steady transport of fine and coarse\nsand, we find that turbulent fluctuations of the flow produce particle\nfluctuation energy at an unexpectedly high rate in saltation even under\nconditions for which the effects of turbulence are usually believed to be\nsmall.", "category": "physics_flu-dyn" }, { "text": "Vortex Induced Oscillations of Cylinders: This article submitted to the APS-DFD 2008 conference, accompanies the fluid\ndynamics video depicting the various orientational dynamics of a hinged\ncylinder suspended in a flow tank. The different behaviors displayed by the\ncylinder range from steady orientation to periodic oscillation and even\nautorotation. We illustrate these features using a phase diagram which captures\nthe observed phenomena as a function of Reynolds number and reduced inertia. A\nhydrogen bubble flow visualization technique is also used to show vortex\nshedding structure in the cylinder's wake which results in these oscillations.", "category": "physics_flu-dyn" }, { "text": "Inhibition of water vapor condensation by dipropylene glycol droplets on\n hydrophobic surfaces via vapor sink strategy: Condensation, frosting and icing are natural phenomena, and have been\nenduring challenges for human society and modern engineering. These phenomena\npose a range of issues, from hazardous icy road surfaces, damage to electronic\ndevices due to condensation, to frost accumulation on power lines and aircraft.\nDipropylene glycol (DG) is a non-toxic, non-corrosive, and biologically safe\nsubstance with excellent hygroscopicity, capable of inhibiting condensation and\nice formation through the vapor sink strategy. However, there has been limited\nresearch on DG. Herein, we provides a detailed study of the performance of DG\ndroplets in suppressing condensation with experiments and theoretical analysis.\nResults found that a dry zone ring would form around the DG droplet on a\ncooling solid surface. The ratio of the dry-zone radius to the DG droplet\nradius increases with the solid surface temperature and scales as 1/(Tdew-Tc).\nResults indicated that the cooling time of substrate surface does not affect\nthe ratio in a short period. When the temperature is higher, the ratio\ndecreases slightly with DG droplet volume; while it keeps almost still, when\nthe temperature is lower. A theoretical model is also proposed to reveal the\nrelationship between the ratio and the substrate surface temperature. These\nfindings hold the promise of not only enhancing our understanding of\ncondensation suppression but also advancing the development of innovative\nsolutions for various industries, including transportation, electronics,\naviation, and power distribution.", "category": "physics_flu-dyn" }, { "text": "Energy Fluxes during Dynamo Reversals: Using direct numerical simulations of the equations of magnetohydrodynamics,\nwe study reversals of the magnetic field generated by the flow of an\nelectrically conducting fluid in a sphere. We show that at low magnetic Prandtl\nnumbers, Pm=0.5, the decrease of magnetic energy, ohmic dissipation and power\nof the Lorentz force during a reversal is followed by an increase of the power\ninjected by the force driving the flow and an increase of viscous dissipation.\nCross correlations show that the Lorentz energy flux is in advance with respect\nto the other energy fluxes. We also observe that during a reversal, the maximum\nof the magnetic energy density migrates from one hemisphere to the other and\ncomes back to its initial position, in agreement with recent experimental\nobservations. For larger magnetic Prandtl numbers (Pm= 1, 2), the magnetic\nfield reversals do not display these trends and strongly differ one from\nanother.", "category": "physics_flu-dyn" }, { "text": "Autophoretic locomotion from geometric asymmetry: Among the few methods which have been proposed to create small-scale\nswimmers, those relying on self-phoretic mechanisms present an interesting\ndesign challenge in that chemical gradients are required to generate net\npropulsion. Building on recent work, we propose that asymmetries in geometry\nare sufficient to induce chemical gradients and swimming. We illustrate this\nidea using two different calculations. We first calculate exactly the\nself-propulsion speed of a system composed of two spheres of unequal sizes but\nidentically chemically homogeneous. We then consider arbitrary,\nsmall-amplitude, shape deformations of a chemically-homogeneous sphere, and\ncalculate asymptotically the self-propulsion velocity induced by the shape\nasymmetries. Our results demonstrate how geometric asymmetries can be tuned to\ninduce large locomotion speeds without the need of chemical patterning.", "category": "physics_flu-dyn" }, { "text": "Exponential asymptotics for steady parasitic capillary ripples on steep\n gravity waves: In this paper we develop an asymptotic theory for steadily travelling\ngravity-capillary waves under the small-surface tension limit. In an\naccompanying work [Shelton et al. (2021), J. Fluid Mech., vol 922] it was\ndemonstrated that solutions associated with a perturbation about a\nleading-order gravity wave (a Stokes wave) contain surface-tension-driven\nparasitic ripples with an exponentially-small amplitude. Thus a naive\nPoincar\\'e expansion is insufficient for their description. Here, we shall\ndevelop specialised methodologies in exponential asymptotics for derivation of\nthe parasitic ripples on periodic domains. The ripples are shown to arise in\nconjunction with Stokes lines and the Stokes phenomenon. The analysis relies\ncrucially upon the derivation and analysis of singularities in the analytic\ncontinuation of the classic Stokes wave. A solvability condition is derived,\nshowing that solutions of this type do not exist at certain values of the Bond\nnumber. The asymptotic results are compared to full numerical solutions and\nshow excellent agreement. The work provides corrections and insight of a\nseminal theory on parasitic capillary waves first proposed by Longuet-Higgins\n[J. Fluid Mech., vol. 16 (1), 1963, pp. 138-159].", "category": "physics_flu-dyn" }, { "text": "Generating a tide-like flow in a cylindrical vessel by electromagnetic\n forcing: We show and compare numerical and experimental results on the electromagnetic\ngeneration of a tide-like flow structure in a cylindrical vessel which is\nfilled with the eutectic liquid metal alloy GaInSn. Fields of various strengths\nand frequencies are applied to drive liquid metal flows. The impact of the\nfield variations on amplitude and structure of the flows is investigated. The\nresults represent the basis for a future Rayleigh-B\\'enard experiment, in which\na modulated tide-like flow perturbation is expected to synchronize the typical\nsloshing mode of the large-scale circulation. A similar entrainment mechanism\nfor the helicity in the Sun may be responsible for the synchronization of the\nsolar dynamo with the alignment cycle of the tidally dominant planets Venus,\nEarth and Jupiter.", "category": "physics_flu-dyn" }, { "text": "Suppression of Rayleigh-Taylor turbulence by time-periodic acceleration: The dynamics of Rayleigh-Taylor turbulence convection in presence of an\nalternating, time periodic acceleration is studied by means of extensive direct\nnumerical simulations of the Boussinesq equations. Within this framework, we\ndiscover a new mechanism of relaminarization of turbulence: The alternating\nacceleration, which initially produces a growing turbulent mixing layer, at\nlonger times suppresses turbulent fluctuation and drives the system toward an\nasymptotic stationary configuration. Dimensional arguments and linear stability\ntheory are used to predict the width of the mixing layer in the asymptotic\nstate as a function of the period of the acceleration. Our results provide an\nexample of simple control and suppression of turbulent convection with\npotential applications in different fields.", "category": "physics_flu-dyn" }, { "text": "Solutions to aliasing in time-resolved flow data: Avoiding aliasing in time-resolved flow data obtained through high fidelity\nsimulations while keeping the computational and storage costs at acceptable\nlevels is often a challenge. Well-established solutions such as increasing the\nsampling rate or low-pass filtering to reduce aliasing can be prohibitively\nexpensive for large data sets. This paper provides a set of alternative\nstrategies for identifying and mitigating aliasing that are applicable even to\nlarge data sets. We show how time-derivative data, which can be obtained\ndirectly from the governing equations, can be used to detect aliasing and to\nturn the ill-posed problem of removing aliasing from data into a well-posed\nproblem, yielding a prediction of the true spectrum. Similarly, we show how\nspatial filtering can be used to remove aliasing for convective systems. We\nalso propose strategies to prevent aliasing when generating a database,\nincluding a method tailored for computing nonlinear forcing terms that arise\nwithin the resolvent framework. These methods are demonstrated using a\nnon-linear Ginzburg-Landau model and large-eddy simulation (LES) data for a\nsubsonic turbulent jet.", "category": "physics_flu-dyn" }, { "text": "Neural network-based modelling of unresolved stresses in a turbulent\n reacting flow with mean shear: Data-driven methods for modelling purposes in fluid mechanics are a promising\nalternative given the continuous increase of both computational power and\ndata-storage capabilities. Highly non-linear flows including turbulence and\nreaction are challenging to model, and accurate closures for the unresolved\nterms in large eddy simulations of such flows are difficult to obtain. In this\nstudy, we investigate the use of artificial neural networks for modelling an\nimportant unclosed term namely the unresolved stress tensor, in a highly\ndemanding turbulent and reacting flow, which additionally includes mean shear.\nThe performance of the neural network-based modelling approach is conducted a\npriori following a coarsened mesh approach, and compared against the\npredictions of eight other classic models in the literature, which include both\nstatic and dynamic formulations.", "category": "physics_flu-dyn" }, { "text": "Collective locomotion of two-dimensional lattices of flapping plates: We study the propulsive properties of rectangular and rhombic lattices of\nflapping plates at O(10--100) Reynolds numbers in incompressible flow. We vary\nfive parameters: flapping amplitude, frequency (or Reynolds number), horizontal\nand vertical spacings between plates, and oncoming fluid stream velocity.\nLattices that are closely spaced in the streamwise direction produce intense\nvortex dipoles between adjacent plates. The lattices transition sharply from\ndrag- to thrust-producing as these dipoles switch from upstream to downstream\norientations at critical flow speeds. Near these transitions the flows assume a\nvariety of periodic and nonperiodic states, with and without up-down symmetry,\nand multiple stable self-propelled speeds can occur. As the streamwise spacing\nincreases, the plates may shed typical vortex wakes that impinge on downstream\nneighbors. With small lateral spacing, rectangular lattices yield net drag,\nwhile rhombic lattices may generate net thrust, sometimes with high efficiency.\nAs lateral spacing increases, rectangular lattices begin to generate thrust,\neventually with slightly higher efficiencies than rhombic lattices, as the two\ntypes of lattice flows converge. At Re = 70, the maximum Froude efficiencies of\ntime-periodic lattice flows are about twice those of an isolated plate. At\nlower Re, the lattices' efficiency advantage increases until the isolated\nflapping plate no longer generates thrust. The mean input power needed to\ngenerate the lattice flows can be estimated in the limits of small and large\nstreamwise spacings, with small-gap and Poiseuille-like flows between the\nplates respectively in the two cases. For both lattices, the mean input power\nsaturates as the lateral spacing becomes large (and thrust occurs). At small\nlateral spacings, the rhombic lattices' input power may be much larger when the\nplates overlap, leading to a decrease in Froude efficiency.", "category": "physics_flu-dyn" }, { "text": "Effect of P\u00e9clet number on miscible rectilinear displacement in a\n Hele-Shaw cell: Influence of fluid dispersion on the Saffman-Taylor instability in miscible\nfluids has been investigated both in the linear and nonlinear regimes. The\nconvective characteristic scales are used for the dimensionless formulation\nthat incorporates P\\'{e}clet number (Pe) into the governing equations as a\nmeasure for the fluid dispersion. A linear stability analysis (LSA) has been\nperformed in a similarity transformation domain using the quasi-steady-state\napproximation. LSA results show that systems with large Pe become more unstable\nand the onset of instability occurs earlier compared to the case when Pe is\nsmaller. Variations of the most unstable wave number and the cut-off wave\nnumber with Pe have been analyzed. Fourier spectral method has been used for\nthe numerical simulations of the fully nonlinear system. The results indicate\nthat the wave numbers of the unstable modes increase with Pe. Influence of the\nanisotropic dispersion on the onset in both the linear and nonlinear regimes\nhas been analyzed. Large transverse diffusivity increases the length scale of\nthe fingers quickly and merges the fingers to generate coarser fingers. Finally\nthe combined effect of the Korteweg stress and Pe in the linear regime has been\nperused. In the presence of the Korteweg stresses and depending upon various\nflow parameters, a fluid system with larger Pe exhibits smaller instantaneous\ngrowth rate than with smaller Pe, which is a counter-intuitive result.", "category": "physics_flu-dyn" }, { "text": "Unsupervised Machine Learning to Teach Fluid Dynamicists to Think in 15\n Dimensions: An autoencoder is used to compress and then reconstruct three-dimensional\nstratified turbulence data in order to better understand fluid dynamics by\nstudying the errors in the reconstruction. The original single data set is\nresolved on approximately $6.9\\times10^{10}$ grid points, and 15 fluid\nvariables in three spatial dimensions are used, for a total of about $10^{12}$\ninput quantities in three dimensions. The objective is to understand which of\nthe input variables contains the most relevant information about the local\nturbulence regimes in stably stratified turbulence (SST). This is accomplished\nby observing flow features that appear in one input variable but then `bleed\nover' to multiple output variables. The bleed over is shown to be robust with\nrespect to the number of layers in the autoencoder. In this proof of concept,\nthe errors in the reconstruction include information about the spatial\nvariation of vertical velocity in most of the components of the reconstructed\nrate-of-strain tensor and density gradient, which suggests that vertical\nvelocity is an important marker for turbulence features of interest in SST.\nThis result is consistent with what fluid dynamicists already understand about\nSST and, therefore, suggests an approach to understanding turbulence based on\nmore detailed analyses of the reconstruction on errors in an autoencoding\nalgorithm.", "category": "physics_flu-dyn" }, { "text": "Evidence for Bolgiano-Obukhov scaling in rotating stratified turbulence\n using high-resolution direct numerical simulations: We report results on rotating stratified turbulence in the absence of\nforcing, with large-scale isotropic initial conditions, using direct numerical\nsimulations computed on grids of up to 4096^3 points. The Reynolds and Froude\nnumbers are respectively equal to Re=5.4 x 10^4 and Fr=0.0242. The ratio of the\nBrunt-V\\\"ais\\\"al\\\"a to the inertial wave frequency, N/f, is taken to be equal\nto 4.95, a choice appropriate to model the dynamics of the southern abyssal\nocean at mid latitudes. This gives a global buoyancy Reynolds number\nR_B=ReFr^2=32, a value sufficient for some isotropy to be recovered in the\nsmall scales beyond the Ozmidov scale, but still moderate enough that the\nintermediate scales where waves are prevalent are well resolved. We concentrate\non the large-scale dynamics, for which we find a spectrum compatible with the\nBolgiano-Obukhov scaling, and confirm that the Froude number based on a typical\nvertical length scale is of order unity, with strong gradients in the vertical.\nTwo characteristic scales emerge from this computation, and are identified from\nsharp variations in the spectral distribution of either total energy or\nhelicity. A spectral break is also observed at a scale at which the partition\nof energy between the kinetic and potential modes changes abruptly, and beyond\nwhich a Kolmogorov-like spectrum recovers. Large slanted layers are ubiquitous\nin the flow in the velocity and temperature fields, with local overturning\nevents indicated by small Richardson numbers, and a small large-scale\nenhancement of energy directly attributable to the effect of rotation is also\nobserved.", "category": "physics_flu-dyn" }, { "text": "Scaling behavior of optimally structured catalytic microfluidic reactors: In this study of catalytic microfluidic reactors we show that, when optimally\nstructured, these reactors share underlying scaling properties. The scaling is\npredicted theoretically and verified numerically. Furthermore, we show how to\nincrease the reaction rate significantly by distributing the active porous\nmaterial within the reactor using a high-level implementation of topology\noptimization.", "category": "physics_flu-dyn" }, { "text": "Long-time fractalisation of liquid films undergoing thermocapillary\n instability: The study of viscous thin film flow has led to the development of highly\nnonlinear partial differential equations that model how the evolution of the\nfilm height is affected by different forces. We investigate a model of\ninteraction between surface tension and the thermocapillary Marangoni effect,\nwith a particular focus on the long-time limit. In this limit, the model\npredicts patterns that repeat on successively smaller spatial scales near\npoints where the film thickness vanishes. By solving the governing equation in\na space- and time-rescaled coordinate system, we observe the dynamics much\nfurther in time than has previously been achieved. Our numerical results\nsuggest that patterns form on a close to, but not exactly, geometrically\ndecreasing scale, potentially approaching a geometric scale very slowly as time\ntends to infinity.", "category": "physics_flu-dyn" }, { "text": "An integral model based on slender body theory, with applications to\n curved rigid fibers: We propose a novel integral model describing the motion of curved slender\nfibers in viscous flow, and develop a numerical method for simulating dynamics\nof rigid fibers. The model is derived from nonlocal slender body theory (SBT),\nwhich approximates flow near the fiber using singular solutions of the Stokes\nequations integrated along the fiber centerline. In contrast to other models\nbased on (singular) SBT, our model yields a smooth integral kernel which\nincorporates the (possibly varying) fiber radius naturally. The integral\noperator is provably negative definite in a non-physical idealized geometry, as\nexpected from PDE theory. This is numerically verified in physically relevant\ngeometries. We propose a convergent numerical method for solving the integral\nequation and discuss its convergence and stability. The accuracy of the model\nand method is verified against known models for ellipsoids. Finally, a fast\nalgorithm for computing dynamics of rigid fibers with complex geometries is\ndeveloped.", "category": "physics_flu-dyn" }, { "text": "Equatorially Trapped Convection in a Rapidly Rotating Shallow Shell: Motivated by the recent discovery of subsurface oceans on planetary moons and\nthe interest they have generated, we explore convective flows in shallow\nspherical shells of dimensionless gap width $\\varepsilon^2\\ll 1$ in the rapid\nrotation limit $\\mathrm{E}\\ll1$, where $\\mathrm{E}$ is the Ekman number. We\nemploy direct numerical simulations (DNS) of the Boussinesq equations to\ncompute the local heat flux $\\mathrm{Nu}(\\lambda)$ as a function of the\nlatitude $\\lambda$ and use the results to characterize the trapping of\nconvection at low latitudes, around the equator. We show that these results are\nquantitatively reproduced by an asymptotically exact nonhydrostatic equatorial\n$\\beta$-plane convection model at a much more modest computational cost than\nDNS. We identify the trapping parameter $\\beta=\\varepsilon \\mathrm{E}^{-1}$ as\nthe key parameter that controls the vigor and latitudinal extent of convection\nfor moderate thermal forcing when $\\mathrm{E}\\sim\\varepsilon$ and\n$\\varepsilon\\downarrow 0$. This model provides a new theoretical paradigm for\nnonlinear investigations.", "category": "physics_flu-dyn" }, { "text": "Momentum transfer by linearised eddies in channel flows: The presence and structure of an Orr-like inviscid mechanism is studied in\nfully developed, large-scale turbulent channel flow. Orr-like `bursts' are\ndefined by the relation between the amplitude and local tilting angle of the\nwall-normal velocity perturbations, and extracted by means of wavelet-based\nfilters. They span the shear-dominated region of the flow, and their sizes and\nlifespans are proportional to the distance from the wall in the logarithmic\nlayer, forming a self-similar eddy hierarchy consistent with Townsend's\nattached-eddy model. Except for their amplitude, which has to be determined\nnonlinearly, linearised transient growth represents their evolution reasonably\nwell. Conditional analysis, based on wavelet-filtered and low-pass-filtered\nvelocity fields, reveals that bursts of opposite sign pair side-by-side to form\ntilted quasi-streamwise rollers, which align along the streaks of the\nstreamwise velocity with the right sign to reinforce them, and that they\npreferentially cluster along pre-existing streak inhomogeneities. On the other\nhand, temporal analysis shows that consecutive rollers do not form\nsimultaneously, suggesting that they incrementally trigger each other. This\npicture is similar to that of the streak-vortex cycle of the buffer layer, and\nthe properties of the bursts suggest that they are different manifestations of\nthe well-known attached Q$_2$-Q$_4$ events of the Reynolds stress.", "category": "physics_flu-dyn" }, { "text": "A Set of Statistical Variables for Hydrodynamic Flow: Through a discussion of some typical unsteady hydrodynamic flows, we argue\nthat the time averaged hydrodynamic functions at each point give a rather\nsparse filling of the local jet space. This situation then suggests a set of\ntime dependent probability functions that are shown to give evolution uniquely\ndefined by the Navier-Stokes equations through a set of \"differential\ndistribution equations.\" The closure relations are therefore unique and have no\nad hoc characteristics. Annealing methods are proposed as a way to arrive at\nthe stable stationary solutions corresponding to time averaged fluid flow with\nconstant driving forces and fixed boundary conditions. Some applications of\nthis method to quantum statistical mechanics and kinetic theory to higher\norders are suggested.", "category": "physics_flu-dyn" }, { "text": "Delayed coalescence of surfactant containing sessile droplets: When two sessile drops of the same liquid touch, they merge into one drop,\ndriven by capillarity. However, the coalescence can be delayed, or even\ncompletely stalled for a substantial period of time, when the two drops have\ndifferent surface tensions, despite being perfectly miscible. A temporary state\nof non-coalescence arises, during which the drops move on their substrate, only\nconnected by a thin neck between them. Existing literature covers pure liquids\nand mixtures with low surface activities. In this paper, we focus on the case\nof large surface activities, using aqueous surfactant solutions with varying\nconcentrations. It is shown that the coalescence behavior can be classified\ninto three regimes that occur for different surface tensions and contact angles\nof the droplets at initial contact. However, not all phenomenology can be\npredicted from surface tension contrast or contact angles alone, but strongly\ndepends on the surfactant concentrations as well. This reveals that the merging\nprocess is not solely governed by hydrodynamics and geometry, but also depends\non the molecular physics of surface adsorption.", "category": "physics_flu-dyn" }, { "text": "Competition between Rayleigh--B\u00e9nard and horizontal convection: We investigate the dynamics of a fluid layer subject to an imposed bottom\nheat flux and a top monotonically-increasing temperature profile driving\nhorizontal convection. We use direct numerical simulations and consider a large\nrange of flux-based Rayleigh numbers $10^6 \\leq Ra_F \\leq 10^9$ and imposed top\nhorizontal to bottom vertical heat flux ratios $0 \\leq \\Lambda \\leq 1$. The\nfluid domain is a closed two-dimensional box with aspect ratio $4\\leq \\Gamma\n\\leq 16$ and we consider no-slip boundaries and adiabatic side walls. We\ndemonstrate a regime transition from Rayleigh--B\\'enard convection (RB) to\nhorizontal convection (HC) at $\\Lambda\\approx 10^{-2}$, which is independent of\n$Ra_F$ and $\\Gamma$. At small $\\Lambda$, the flow is organized in multiple\noverturning cells with approximately unit aspect ratio, while at large\n$\\Lambda$ a single cell is obtained. The RB-relevant Nusselt number scaling\nwith $Ra_F$ and the HC-relevant Nusselt number scaling with the horizontal\nRayleigh number $Ra_L=Ra_F\\Lambda\\Gamma^4$ are in good agreement with previous\nresults from classical RB convection and HC studies in the limit $\\Lambda \\ll\n10^{-2}$ and $\\Lambda \\gg 10^{-2}$, respectively. We demonstrate that the\nsystem is multi-stable near the transition $\\Lambda\\approx10^{-2}$, i.e. the\nexact number of cells not only depends on $\\Lambda$ but also on the system's\nhistory. Our results suggest that subglacial lakes, which motivated this study,\nare likely to be dominated by RB convection, unless the slope of the ice-water\ninterface, which controls the horizontal temperature gradient via the\npressure-dependence of the freezing point, is greater than unity.", "category": "physics_flu-dyn" }, { "text": "Transition to turbulence in slowly divergent pipe flow: The results of a combined experimental and numerical study of the flow in\nslowly diverging pipes are presented. Interestingly, an axisymmetric conical\nrecirculation cell has been observed. The conditions for its existence and the\nlength of the cell are simulated for a range of diverging angles and expansion\nratios. There is a critical velocity for the appearance of this state. When the\nflow rate increases further, a subcritical transition for localized turbulence\narises. The transition and relaminarization experiments described here quantify\nthe extent of turbulence. The findings suggest that the transition scenario in\nslowly diverging pipes is a combination of stages similar to those observed in\nsudden expansions and in straight circular pipe flow.", "category": "physics_flu-dyn" }, { "text": "Emergent scar lines in chaotic advection of passive directors: We examine the spatial field of orientations of slender fibers that are\nadvected by a two-dimensional fluid flow. The orientation field of these\npassive directors are important in a wide range of industrial and geophysical\nflows. We introduce emergent scar lines as the dominant coherent structures in\nthe orientation field of passive directors in chaotic flows. Previous work has\nidentified the existence of scar lines where the orientation rotates by {\\pi}\nover short distances, but the lines that were identified disappeared as time\nprogressed. As a result, earlier work focused on topological singularities in\nthe orientation field which we find to play a negligible role at long times. We\nuse the standard map as a simple time-periodic two-dimensional (2D) flow that\nproduces Lagrangian chaos. This class of flows produce persistent patterns in\npassive scalar advection, and we find that a different kind of persistent\npattern develops in the passive director orientation field. We identify the\nmechanism by which emergent scar lines grow to dominate these patterns at long\ntimes in complex flows. Emergent scar lines form where the recent stretching of\nthe fluid element is perpendicular to earlier stretching. Thus these scar lines\ncan be labeled by their age, defined as the time since their stretching reached\na maximum.", "category": "physics_flu-dyn" }, { "text": "Layering and vertical transport in sheared double diffusive convection\n in the diffusive regime: A sequence of two and three-dimensional simulations is conducted for the\ndouble diffusive convection (DDC) flows in the diffusive regime subjected to an\nimposed shear. The flow is confined between two horizontal plates which are\nmaintained at different constant temperature, salinity, and different velocity,\nthus setting up a shear across the flow. The lower plate is fixed at higher\ntemperature and salinity, while the overall (unperturbed) density gradient is\nstatically stable. For a wide range of control parameters, and for sufficiently\nstrong perturbation of the conductive initial state, we find that\nstaircase-like structures spontaneously develop, with relatively well-mixed\nlayers separated by sharp interfaces of enhanced scalar gradient. Such\nstaircases appear to be robust even in the presence of strong shear over very\nlong times, although we typically observe early time coarsening of the number\nof observed layers. For the same set of control parameters, different\nasymptotic layered states, with markedly different vertical scalar fluxes, can\narise for different initial perturbation structures. The imposed shear does\nsignificantly spatio-temporally modify the vertical transport of the various\nscalars. The flux ratio (i.e., the ratio between the density fluxes due to the\ntotal (convective and diffusive) salt flux and the total heat flux) is found,\nat steady state, to be essentially equal to the square root of the ratio of the\nsalt diffusivity to the thermal diffusivity, consistently with the physical\nmodel originally proposed by Linden and Shirtcliffe (1978) and the variational\narguments presented by Stern (1982) for unsheared double diffusive convection.", "category": "physics_flu-dyn" }, { "text": "On the sub-shock formation in extended thermodynamics: In hyperbolic dissipative systems, the solution of the shock structure is not\nalways continuous and a discontinuous part (sub-shock) appears when the\nvelocity of the shock wave is greater than a critical value. In principle, the\nsub-shock may occur when the shock velocity $s$ reaches one of the\ncharacteristic eigenvalues of the hyperbolic system. Nevertheless, Rational\nExtended Thermodynamics (ET) for a rarefied monatomic gas predicts the\nsub-shock formation only when $s$ exceeds the maximum characteristic velocity\nof the system evaluated in the unperturbed state $\\lambda^{\\max}_0$. This fact\nagrees with a general theorem asserting that continuous shock structure cannot\nexist for $s >\\lambda^{\\max}_0 $. In the present paper, first, the shock\nstructure is numerically analyzed on the basis of ET for a rarefied polyatomic\ngas with $14$ independent fields. It is shown that, also in this case, the\nshock structure is still continuous when $s$ meets characteristic velocities\nexcept for the maximum one and therefore the sub-shock appears only when $s\n>\\lambda^{\\max}_0 $. This example reinforces the conjecture that, the\ndifferential systems of ET theories have the special characteristics such that\nthe sub-shock appears only for $s$ greater than the unperturbed maximum\ncharacteristic velocity. However, in the second part of the paper, we construct\na counterexample of this conjecture by using a simple $2 \\times 2$ hyperbolic\ndissipative system which satisfies all requirements of ET. In contrast to\nprevious results, we show the clear sub-shock formation with a slower shock\nvelocity than the maximum unperturbed characteristic velocity.", "category": "physics_flu-dyn" }, { "text": "Development of a coupling between a system thermal-hydraulic code and a\n reduced order CFD model: The nuclear community has coupled several three-dimensional Computational\nFluid Dynamics (CFD) solvers with one-dimensional system thermal-hydraulic\n(STH) codes. This work proposes to replace the CFD solver by a reduced order\nmodel (ROM) to reduce the computational cost. The system code RELAP5-MOD3.3 and\na ROM of the finite volume CFD solver OpenFOAM are coupled by a partitioned\ndomain decomposition coupling algorithm using an implicit coupling scheme. The\nvelocity transported over a coupling boundary interface is imposed in the ROM\nusing a penalty method. The coupled models are evaluated on open and closed\npipe flow configurations. The results of the coupled simulations with the ROM\nare close to those with the CFD solver. Also for new parameter sets, the\ncoupled RELAP5/ROM models are capable of predicting the coupled RELAP5/CFD\nresults with good accuracy. Finally, coupling with the ROM is 3-5 times faster\nthan coupling with the CFD solver.", "category": "physics_flu-dyn" }, { "text": "Relaminarising pipe flow by wall movement: Following the recent observation that turbulent pipe flow can be\nrelaminarised by a relatively simple modification of the mean velocity profile,\nwe here carry out a quantitative experimental investigation of this phenomenon.\nOur study confirms that a flat velocity profile leads to a collapse of\nturbulence and in order to achieve the blunted profile shape, we employ a\nmoving pipe segment that is briefly and rapidly shifted in the streamwise\ndirection. The relaminarisation threshold and the minimum shift length and\nspeeds are determined as a function of Reynolds number. Although turbulence is\nstill active after the acceleration phase, the modulated profile possesses a\nseverely decreased lift-up potential as measured by transient growth. As shown,\nthis results in an exponential decay of fluctuations and the flow\nrelaminarises. While this method can be easily applied at low to moderate flow\nspeeds, the minimum streamwise length over which the acceleration needs to act\nincreases linearly with the Reynolds number.", "category": "physics_flu-dyn" }, { "text": "Capillary filling in drop merging: dynamics of the four-phase contact\n point: The merging of immiscible drops differs significantly from the merging of\nmiscible drops due to the formation of a liquid-liquid interface between drops.\nThe immiscibility requires the formation of a four-phase contact point, where\nthe drops, the gas and the substrate meet. We show that this point has its own\nunique dynamics, never studied beforehand. For very different scenarios, the\npropagation distance of this point follows scales with time like t^{0.5}. A\nmodel balancing the driving and dissipative forces agrees with our experiments.", "category": "physics_flu-dyn" }, { "text": "Direct numerical simulations of incompressible multiphase\n electrohydrodynamic flow with single-phase transportation schemes: In the present study, two schemes named face discernment and flux correction\nare proposed to achieve single-phase transportation of free charge in\nmultiphase electrohydrodynamic(EHD) problems. Many EHD phenomena occur between\nair and another liquid while the free charge can only be transported in the\nliquid phase through ohmic conduction and convection due to the poor\nconductivity of air. However, the charge may be leaked into the dielectric air\nduring the simulation due to the asynchronous transportation between interface\nand free charge. To avoid this unphysical error, a face discernment method is\ndesigned to produce an accurate ohmic conduction of free charge by providing a\nsuperior physical properties distribution at the interface. Subsequently the\nflux correction method is developed to correct the advection flux of charge\ndensity to prevent ions crossing the interface. These two schemes are based on\nthe Volume of Fluid (VOF) model and independent with the specific interface\nupdating method. The performance of the proposed methods are carefully\nvalidated with several test cases. The algorithms are implemented as an\nOpenFOAM extension and are published as open source.", "category": "physics_flu-dyn" }, { "text": "Resonant triad interactions in stably-stratified uniform shear flow: We investigate exact and near resonant triad interactions (RTI) in a\ntwo-dimensional stably stratified uniform shear flow confined between two\ninfinite parallel walls in the absence of viscous and diffusive effects. RTI\noccur when three interacting waves satisfy the resonance conditions of the form\n$k_1 \\pm k_2 = k_3$ and $\\omega_1 \\pm \\omega_2 = \\omega_3$ with $k_i$ and\n$\\omega_i$ being the wavenumber and frequency of the ith wave ($i \\in\n{1,2,3}$), respectively. The linear stability problem is solved analytically,\nwhich gives the eigenfunctions in the form of the modified Bessel functions. It\nis identified that an interaction between two primary modes having the same\nfrequency $\\omega$ but different wavenumbers $k_m$ and $k_n$ produces two\ndifferent secondary modes: one time-dependent (superharmonic) mode having\nfrequency $2\\omega$ and wavenumber $k_m +k_n$, and the other time-independent\n(subharmonic) mode with $\\omega = 0$ and wavenumber $k_m - k_n$. The\ndifferential equation governing the spatial amplitude of the superharmonic mode\nis solved numerically as well as analytically using the method of variation of\nparameters. It turns out that the linear operator associated with the\ndifferential equation of the superharmonic mode is the same as the linear\nstability operator and that the solvability condition of the differential\nequation is found to be associated with the existence of RTI. The existence of\nresonant triad interactions predicted by the dispersion relation, are justified\nby showing the divergence of the spatial amplitude of superharmonic mode.\nVarious cases of wave interactions in a stably stratified shear flow are\nanalysed in the presence of a resonant triad for various frequencies and linear\nstratifications.", "category": "physics_flu-dyn" }, { "text": "Liquid ropes: a geometrical model for thin viscous jets instabilities: Thin viscous fluid threads falling onto a moving belt behave in a way\nreminiscent of a sewing machine, generating a rich variety of periodic\nstitch-like patterns including meanders, W-patterns, alternating loops, and\ntranslated coiling. These patterns form to accommodate the difference between\nthe belt speed and the terminal velocity at which the falling thread strikes\nthe belt. Using direct numerical simulations, we show that inertia is not\nrequired to produce the aforementioned patterns. We introduce a quasi-static\ngeometrical model which captures the patterns, consisting of three coupled ODEs\nfor the radial deflection, the orientation and the curvature of the path of the\nthread's contact point with the belt. The geometrical model reproduces well the\nobserved patterns and the order in which they appear as a function of the fall\nheight.", "category": "physics_flu-dyn" }, { "text": "Drop impact on viscous liquid films: When a liquid drop falls on a solid substrate, the air layer in between them\ndelays the occurrence of liquid--solid contact. For impacts on smooth\nsubstrates, the air film can even prevent wetting, allowing the drop to bounce\noff with dynamics identical to that observed for impacts on superamphiphobic\nmaterials. In this article, we investigate similar bouncing phenomena,\noccurring on viscous liquid films, that mimic atomically smooth substrates,\nwith the goal to probe their effective repellency. We elucidate the mechanisms\nassociated to the bouncing to non-bouncing (floating) transition using\nexperiments, simulations, and a minimal model that predicts the main\ncharacteristics of drop impact, the contact time, and the coefficient of\nrestitution. In the case of highly viscous or very thin films, the impact\ndynamics is not affected by the presence of the viscous film. Within this\nsubstrate--independent limit, bouncing is suppressed once the drop viscosity\nexceeds a critical value as on superamphiphobic substrates. For thicker or less\nviscous films, both the drop and film properties influence the rebound dynamics\nand conspire to inhibit bouncing above a critical film thickness. This\nsubstrate--dependent regime also admits a limit, for low viscosity drops, in\nwhich the film properties alone determine the limits of repellency.", "category": "physics_flu-dyn" }, { "text": "MHD Pulsatile Two-Phase Blood Flow Through a Stenosed Artery with Heat\n and Mass Transfer: In this paper, effects of heat and mass transfer on two-phase pulsatile blood\nflow through a narrowed stenosed artery with radiation and the chemical\nreaction have been investigated. A vertical artery is assumed in which magnetic\nfield is applied along the radial direction of the artery. The characteristics\nof blood in narrow arteries are analyzed by considering blood as Newtonian\nfluid in both core as well as in plasma regions. Exact solutions have been\nfound for velocity, energy and concentration equations of the blood flow. To\nunderstand the behavior of blood flow, graphs of the velocity profile, wall\nshear stress, flow rate, flow impedance and concentration profile have been\nportrayed for different values of the magnetic and radiation parameter. In\norder to validate our result, a comparative study has been presented between\nthe single-phase and two-phase model of the blood flow and it is observed that\nthe two-phase model fits more accurately with the experimental data than the\nsingle phase model. For pulsatile flow, the phase difference between the\npressure gradient and flow rate has been displayed with magnetic field\nparameter and height of the stenosis. Contour plots have been plotted for 10%,\n20% and 30% case of an arterial blockage.", "category": "physics_flu-dyn" }, { "text": "Generation and Breakdown of Surface Streaks in Wind-Driven Aqueous Flow: A bypass transition scenario in a wind-stress driven aqueous flow is analysed\nusing a temporally developing boundary layer model with accelerating surface\ndrift velocity. The parameters of the model are selected to mimic a wave-tank\nexperiment with a reference wind speed of 5 m/s. To study the boundary layer\nprocesses in isolation, a flat free surface is adapted, which inhibits the\ninitiation of waves. First, preferred initial perturbations to which the\nboundary layer is the most sensitive are identified using linear non-normal\ngrowth theory. These perturbations are arranged as streamwise-constant vortex\npairs located adjacent to the free surface. Subsequently, direct numerical\nsimulations are initialized with these optimal perturbations and streamwise\nstreaks are generated. High-speed streaks penetrate into deeper water layers\nand undergo sinuous instabilities reminiscent of the instabilities developing\non low-speed streaks in wall-bounded flows. Streak instabilities induce lateral\nundulations at the free surface, which closely resemble the dye patterns before\nthe onset of waves in wind-wave-tank experiments. The present analysis provides\na theoretical background for these experimental observations.", "category": "physics_flu-dyn" }, { "text": "Size-dependent transient nature of localized turbulence in transitional\n channel flow: It has been reported that a fully localized turbulent band in channel flow\nbecomes sustained when the Reynolds number is above a threshold. Here we show\nevidences that turbulent bands are of a transient nature instead. When the band\nlength is controlled to be fixed, lifetimes of turbulent bands appear to be\nstochastic and exponentially distributed, a sign of a memoryless transient\nnature. Besides increasing with the Reynolds number, the mean lifetime also\nstrongly increases with the band length. Given that the band length always\nchanges over time in real channel flow, this size dependence may translate into\na time dependence, which needs to be taken into account when clarifying the\nrelationship between channel flow transition and the directed percolation\nuniversality class.", "category": "physics_flu-dyn" }, { "text": "A two-dimensional analytical model of vertical water entry for\n asymmetric bodies with flow separation: The vertical water entry of asymmetric two-dimensional bodies with flow\nseparation is considered. As long as there is no flow separation, linearised\nWagner's theory combined with the Modified Logvinovich Model has been shown to\nprovide computationally fast and reliable estimates of slamming loads during\nwater entry. Tassin et al. (2014) introduced the Fictitious Body Continuation\n(FBC) concept as a way to extend the use of Wagner's model to separated flow\nconfigurations, but they only considered symmetric bodies. In the present\nstudy, we investigate the ability of the FBC concept to provide accurate\nestimates of slamming loads for asymmetric bodies. In this case, flow\nseparation may not occur simultaneously on both sides of the body. During an\nintermediate phase, slamming loads are governed by a competition between the\nlocal drop in pressure due to partial flow separation and the ongoing expansion\nof the wetted area. As a first benchmark for the model, we consider the water\nentry of an inclined flat plate and compare the FBC estimates with the results\nof a nonlinear model. Then, we consider the case of a foil and compare the FBC\nresults with Computational Fluid Dynamics predictions. In both cases, we find\nthat the FBC model is able to provide reliable estimates of the slamming loads.", "category": "physics_flu-dyn" }, { "text": "Nonlinear water waves in shallow water in the presence of constant\n vorticity: A Whitham approach: Two-dimensional nonlinear gravity waves travelling in shallow water on a\nvertically sheared current of constant vorticity are considered. Using Euler\nequations, in the shallow water approximation, hyperbolic equations for the\nsurface elevation and the horizontal velocity are derived. Using Riemann\ninvariants of these equations, that are obtained analytically, a closed-form\nnonlinear evolution equation for the surface elevation is derived. A dispersive\nterm is added to this equation using the exact linear dispersion relation. With\nthis new single first-order partial differential equation, vorticity effects on\nundular bores are studied. Within the framework of weakly nonlinear waves, a\nKdV-type equation and a Whitham equation with constant vorticity are derived\nfrom this new model and the effect of vorticity on solitary waves and periodic\nwaves is considered. Futhermore, within the framework of the new model and the\nWhitham equation a study of the effect of vorticity on the breaking time of\ndispersive waves and hyperbolic waves as well is carried out.", "category": "physics_flu-dyn" }, { "text": "Three-Dimensionality in the flow of an elastically mounted circular\n cylinder with two-degree-of-freedom vortex-induced-vibrations: The study numerically investigates the three-dimensionality in the flow and\ntwo-degree-of-freedom (2 DOF) vortex-induced-vibrations (VIV) characteristics\nof an elastically mounted circular cylinder. The cylinder is allowed to vibrate\nin both streamwise and transverse directions. A low value of mass-ratio with\nthe zero damping coefficient is taken for the simulations. The primary aim is\nto understand the vortex shedding behind the cylinder and the transition\ncharacteristics of the wake-flow from two-dimensional (2D) to three-dimensional\n(3D). The Reynolds number (Re) is varied from 150 (fully 2D flow) to 1000\n(fully 3D flow), which lies inside the laminar range. The reduced velocity is\nvaried which covers all three major VIV branches (Initial Branch (IB), Upper\nBranch (UB), and the Lower Branch (LB)). The oscillating cylinder sweeps the\nfigure-eight trajectory. Two branches (IB, LB) and three branches (IB, UB, LB)\namplitude responses are obtained for the low and high Re values, respectively.\nThe wake behind the cylinder with 2-DOF VIV undergoes the mode-C transition of\n2D to 3D flow as opposed to the direct mode-B transition observed for\ntransverse only VIV in the literature. The critical Re range of the 2D to 3D\ntransition for the 2-DOF VIV cylinder at a reduced velocity of 6 is around 250,\nless than the 1-DOF VIV. Also, this range varies with the variation in and the\nstreamwise to transverse oscillation frequency ratio. A map is proposed for the\n2-DOF VIV, highlighting the different modes of transition obtained for\ncombinations of reduced frequency and Re.", "category": "physics_flu-dyn" }, { "text": "A microfluidic device to sort capsules by deformability: A numerical\n study: Guided by extensive numerical simulations, we propose a microfluidic device\nthat can sort elastic capsules by their deformability. The device consists of a\nduct embedded with a semi-cylindrical obstacle, and a diffuser which further\nenhances the sorting capability. We demonstrate that the device can operate\nreasonably well under changes in the initial position of the the capsule. The\nefficiency of the device remains essentially unaltered under small changes of\nthe obstacle shape (from semi-circular to semi-elliptic cross-section).\nConfinement along the direction perpendicular to the plane of the device\nincreases its efficiency. This work is the first numerical study of cell\nsorting by a realistic microfluidic device.", "category": "physics_flu-dyn" }, { "text": "Hydrodynamic interaction between particles near elastic interfaces: We present an analytical calculation of the hydrodynamic interaction between\ntwo spherical particles near an elastic interface such as a cell membrane. The\ntheory predicts the frequency dependent self- and pair-mobilities accounting\nfor the finite particle size up to the 5th order in the ratio between particle\ndiameter and wall distance as well as between diameter and interparticle\ndistance. We find that particle motion towards a membrane with pure bending\nresistance always leads to mutual repulsion similar as in the well-known case\nof a hard-wall. In the vicinity of a membrane with shearing resistance,\nhowever, we observe an attractive interaction in a certain parameter range\nwhich is in contrast to the behavior near a hard wall. This attraction might\nfacilitate surface chemical reactions. Furthermore, we show that there exists a\nfrequency range in which the pair-mobility for perpendicular motion exceeds its\nbulk value, leading to short-lived superdiffusive behavior. Using the\nanalytical particle mobilities we compute collective and relative diffusion\ncoefficients. The appropriateness of the approximations in our analytical\nresults is demonstrated by corresponding boundary integral simulations which\nare in excellent agreement with the theoretical predictions.", "category": "physics_flu-dyn" }, { "text": "Corrugated interfaces in multiphase core-annular flow: Microfluidic devices can be used to produce highly controlled and\nmonodisperse double or multiple emulsions. The presence of inner drops inside a\njet of the middle phase introduces deformations in the jet, which leads to\nbreakup into monodisperse double emulsions. However, the ability to generate\ndouble emulsions can be compromised when the interfacial tension between the\nmiddle and outer phases is low, leading to flow with high capillary and Weber\nnumbers. In this case, the interface between the fluids is initially deformed\nby the inner drops but the jet does not break into drops. Instead, the jet\nbecomes highly corrugated, which prevents formation of controlled double\nemulsions. We show using numerical calculations that the corrugations are\ncaused by the inner drops perturbing the interface and the perturbations are\nthen advected by the flow into complex shapes.", "category": "physics_flu-dyn" }, { "text": "Synchronized switch harvesting applied to piezoelectric flags: In this article the energy transfer between a flow and a fluttering\npiezoelectric plate is investigated. In particular, the benefits of the use of\na Synchronized Switch Harvesting on Inductor (SSHI) circuit are studied. Both\nwind tunnel experiments and numerical simulations are conducted in order to\nanalyse the influence of the switching process on the dynamics and the\nefficiency of the system. Numerical simulations consist of a weakly non-linear\nmodel of a plate in axial flow equipped with a single pair of piezoelectric\npatches, discretized using a Galerkin method where basis functions are the\nmodes of the plate in vacuum. The discretized model is then integrated in time.\nThe results presented in this paper show that a significant improvement of the\nharvested energy can be obtained using SSHI circuits compared to basic\nresistive circuits. It is also shown that for strongly coupled systems, the\nswitching process inherent to he SSHI circuit has a significant impact on the\ndynamics of the flag, which tends to decrease the relative efficiency gain.", "category": "physics_flu-dyn" }, { "text": "Controlling inertial focussing using rotational motion: In inertial microfluidics lift forces cause a particle to migrate across\nstreamlines to specific positions in the cross section of a microchannel. We\ncontrol the rotational motion of a particle and demonstrate that this allows to\nmanipulate the lift-force profile and thereby the particle's equilibrium\npositions. We perform two-dimensional simulation studies using the method of\nmulti-particle collision dynamics. Particles with unconstrained rotational\nmotion occupy stable equilibrium positions in both halfs of the channel while\nthe center is unstable. When an external torque is applied to the particle, two\nequilibrium positions annihilate by passing a saddle-node bifurcation and only\none stable fixpoint remains so that all particles move to one side of the\nchannel. In contrast, non-rotating particles accumulate in the center and are\npushed into one half of the channel when the angular velocity is fixed to a\nnon-zero value.", "category": "physics_flu-dyn" }, { "text": "Fundamental solutions of an extended hydrodynamic model in two\n dimensions: Derivation, theory, and applications: The inability of the Navier-Stokes-Fourier equations to capture rarefaction\neffects motivates us to adopt the extended hydrodynamic equations. In the\npresent work, a hydrodynamic model, which consists of the conservation laws\nclosed with the recently propounded coupled constitutive relations (CCR), is\nutilized. This model is referred to as the CCR model and is adequate for\ndescribing moderately rarefied gas flows. A numerical framework based on the\nmethod of fundamental solutions is developed to solve the CCR model for\nrarefied gas flow problems in quasi two dimensions. To this end, the\nfundamental solutions of the linearized CCR model are derived in two\ndimensions. The significance of deriving the two-dimensional fundamental\nsolutions is that they cannot be deduced from their three-dimensional\ncounterparts that do exist in literature. As applications, the developed\nnumerical framework based on the derived fundamental solutions is used to\nsimulate (i) a rarefied gas flow between two coaxial cylinders with evaporating\nwalls and (ii) a temperature-driven rarefied gas flow between two non-coaxial\ncylinders. The results for both problems have been validated against those\nobtained with the other classical approaches. Through this, it is shown that\nthe method of fundamental solutions is an efficient tool for addressing quasi\ntwo-dimensional multiphase microscale gas flow problems at a low computational\ncost. Moreover, the findings also show that the CCR model solved with the\nmethod of fundamental solutions is able to describe rarefaction effects, like\ntranspiration flows and thermal stress, generally well.", "category": "physics_flu-dyn" }, { "text": "Realising large areal capacities in liquid metal batteries: a battery\n design concept for mass transfer enhancement: Liquid metal batteries (LMBs) are a promising grid-scale storage device\nhowever, the scalability of this technology and its electrochemical performance\nis limited by mass transport overpotentials. In this work, a numerical model of\na three-layer LMB was developed using a multi-region approach. An alternative\ndesign concept for the battery aimed at reducing mass transport overpotentials,\nincreasing cell capacity, and improving electrochemical cell performance was\nimplemented and evaluated. The design consisted of a coil implanted in the\ncathode, which induced mixing in the layer. Four cases were compared: three in\na 241 Ah LMB at 0.3, 0.5 and 1 A/cm$^{2}$, and one in a larger 481 Ah LMB at\n0.5 A/cm$^{2}$. LMB performance was determined by comparison against baseline\ndiffusion cases and a change in molar fraction of 0.1. The modified LMB\nexhibited dramatic performance increases with a 78% and 85% reduction in\nmass-transport overpotentials at 0.3 A/cm$^{2}$ and 0.5 A/cm$^{2}$,\nrespectively. The improved performance of the battery was directly attributed\nto the flow generated in the cathode. It was found that the coil substantially\nincreased the poloidal volumetric average velocity. Periodically, vortices\nformed that removed concentration gradients from the cathode-electrolyte\ninterface, minimising concentration polarisation. The viability of the design\nwas tested in a lab-scale prototype using Galinstan as the working fluid. The\nvelocity of the flow was determined using particle image velocimetry (PIV), and\nthe results compared to the numerical model. There was a close match between\nthe experimental and numerical results, validating the numerical model and the\nviability of the design. Implementation of this design concept in future LMBs\ncould lead to the realisation of extended discharge capacities and improved\nvoltages. Future work is planned to test the coil in a working battery.", "category": "physics_flu-dyn" }, { "text": "Self-similar motions and related relative equilibria in the $N$-point\n vortex system: We study self-similar solutions of the point-vortex system. The explicit\nformula for self-similar solutions has been obtained for the three point-vortex\nproblem and for a specific example of the four and five point-vortex problems.\nWe see that the families consisting of these self-similar collapsing solutions\nare described by one-parameter families, and their collapse time and\nHamiltonian are also expressed by functions of the same parameter. Then, the\nconfigurations at limit points of the parameter are in relative equilibria. For\nthe many-vortex problem, we investigate the point-vortex system with the help\nof numerical computations. In particular, considering the case that $N - 1$\npoint vortices have a uniform vortex strength, we show that families of\nself-similar collapsing solutions continuously depend on the Hamiltonian and\nthe self-similar solutions asymptotically approach relative equilibria as the\nHamiltonian gets close to certain values. In addition, we prove the existence\nof relative equilibria for the four point-vortex system. We also investigate an\nexample of seven point vortices with non-uniform vortex strengths and give\nnumerical results for it.", "category": "physics_flu-dyn" }, { "text": "An optimisation approach for analysing nonlinear stability with\n transition to turbulence in fluids as an exemplar: This article introduces, and reviews recent work using, a simple optimisation\ntechnique for analysing the nonlinear stability of a state in a dynamical\nsystem. The technique can be used to identify the most efficient way to disturb\na system such that it transits from one stable state to another. The key idea\nis introduced within the framework of a finite-dimensional set of ordinary\ndifferential equations (ODEs) and then illustrated for a very simple system of\n2 ODEs which possesses bistability. Then the transition to turbulence problem\nin fluid mechanics is used to show how the technique can be formulated for a\nspatially-extended system described by a partial differential equation (the\nwell-known Navier-Stokes equation). Within that context, the optimisation\ntechnique bridges the gap between (linear) optimal perturbation theory and the\n(nonlinear) dynamical systems approach to fluid flows. The fact that the\ntechnique has now been recently shown to work in this very high dimensional\nsetting augurs well for its utility in other physical systems.", "category": "physics_flu-dyn" }, { "text": "Electro-Orientation and Electro-Rotation of Metallodielectric Janus\n Particles: The electro-rotation (EROT) and electro-orientation (EOR) behavior of\nmetallodielectric (MD) spherical Janus particles (JP) are studied analytically\nand verified experimentally. This stands in contrast to previous either\nheuristic or numerically computed models of JP dipoles. First, we obtain\nfrequency-dependent analytic expressions for the corresponding dipole terms for\na JP composed of a dielectric and metallic hemispheres, by applying the\nstandard (weak-field) electrokinetic model and using a Fourier-Legendre\ncollocation method for solving two sets of linear equations. EROT and EOR\nspectra, describing the variation of the JP angular velocity on the forcing\nfrequency of a rotating and non-rotating spatially uniform electric field,\nrespectively, are explicitly determined and compared against experiments\n(different JP size and solution conductivity). A favorably good qualitative\nagreement between theory and experimental measurements was found.", "category": "physics_flu-dyn" }, { "text": "Taylor-Couette Instability in General Manifolds: A Lattice Kinetic\n Approach: We present a new lattice kinetic method to simulate fluid dynamics in\ncurvilinear geometries. A suitable discrete Boltzmann equation is solved in\ncontravariant coordinates, and the equilibrium distribution function is\nobtained by a Hermite polynomials expansion of the Maxwell-Boltzmann\ndistribution, expressed in terms of the contravariant coordinates and the\nmetric tensor. To validate the model, we calculate the critical Reynolds number\nfor the onset of the Taylor-Couette instability between two concentric\ncylinders, obtaining excellent agreement with the theory. In order to extend\nthis study to more general geometries, we also calculate the critical Reynolds\nnumber for the case of two concentric spheres, finding good agreement with\nexperimental data. In the case of two concentric tori, we have found that the\ncritical Reynolds is about 10% larger than the respective value for the two\nconcentric cylinders.", "category": "physics_flu-dyn" }, { "text": "Effects of a temperature dependent viscosity on thermal convection in\n binary mixtures: We investigate the effect of a temperature dependent viscosity on the onset\nof thermal convection in a horizontal layer of a binary fluid mixture that is\nheated from below. For an exponential temperature dependence of the viscosity,\nwe find in binary mixtures as a function of a positive separation ratio and\nbeyond a certain viscosity contrast a discontinuous transition between two\nstationary convection modes having a different wavelength. In the range of\nnegative values of the separation ratio, a (continuous or discontinuous)\ntransition from an oscillatory to a stationary onset of convection occurs\nbeyond a certain viscosity contrast, and for large values of the viscosity\nratio, the oscillatory onset of convection is suppressed.", "category": "physics_flu-dyn" }, { "text": "High-Speed Acquisition of Free Vortex Formation: The formation of a free-vortex has been captured by using a high-speed camera\n(Y3, IDTVision, Inc.). The experiment is conducted using a rectangular tank,\nwhich is filled with tap water. The water free surface is open to atmospheric\npressure and is at room temperature, 25\\textcelsius. Water occupies a volume of\n$25\\times 25\\times 10$cm$^3$. By using a stirring-spoon, the stagnant water is\nforced to rotate at a rate of $2\\pi$/sec. Once all the points in the water is\nrotating, it will be drained from a ball valve, with a diameter of 5mm, from\nthe bottom of the tank and the acquisition starts. The formation of the vortex\nis captured with a resolution of $352\\times 824$ pixels at 200 frames per\nseconds (fps) and is exported at 5fps and with a resolution of $1280\\times 720$\nin a \"fluid dynamics video\". The duration of the video in real time is 3.9\nseconds. The slow motion video is 160 seconds. The height of the water remains\nalmost unchanged while acquiring the images.", "category": "physics_flu-dyn" }, { "text": "Data-driven modeling of the chaotic thermal convection in an annular\n thermosyphon: dentifying accurate and yet interpretable low-order models from data has\ngained a renewed interest over the past decade. In the present work, we\nillustrate how the combined use of dimensionality reduction and sparse system\nidentification techniques allows us to obtain an accurate model of the chaotic\nthermal convection in a two-dimensional annular thermosyphon. Taking as\nguidelines the derivation of the Lorenz system, the chaotic thermal convection\ndynamics simulated using a high-fidelity computational fluid dynamics solver\nare first embedded into a low-dimensional space using dynamic mode\ndecomposition. After having reviewed the physical properties the reduced-order\nmodel should exhibit, the latter is identified using SINDy, an increasingly\npopular and flexible framework for the identification of nonlinear\ncontinuous-time dynamical systems from data. The identified model closely\nresembles the canonical Lorenz system, having a similar structure and\nexhibiting the same physical properties. Finally, extensions to other flow\nconfigurations with or without control are discussed.", "category": "physics_flu-dyn" }, { "text": "GFEM study of magnetohydrodynamics thermo-diffusive effect on nanofluid\n flow over power-law stretching sheet along with regression analysis: The present paper uses the Galerkin Finite Element Method to numerically\nstudy the triple diffusive boundary layer flow of homogenous nanofluid over\npower-law stretching sheet with the effect of external magnetic field. The\nfluid is composed of nanoparticles along with dissolved solutal particles in\nthe base fluid. The chief mechanisms responsible for enhancement of convective\ntransport phenomenon in nanofluids - Brownian Motion, Diffusiophoresis and\nThermophoresis have been considered. The simulations performed in this study\nare based on the boundary layer approach. Recently proposed heat flux and\nnanoparticle mass flux boundary conditions have been imposed. Heat transfer,\nsolutal mass transfer and nanoparticle mass transfer are investigated for\ndifferent values of controlling parameters i.e. Brownian-motion parameter,\nThermophoresis parameter, magnetic influence parameter and stretching\nparameter. Multiple regression analysis has been performed to verify the\nrelationship among transfer rate parameters and controlling parameters. The\npresent study finds application in insulation of wires, manufacture of tetra\npacks, production of glass fibres, fabrication of various polymer and plastic\nproducts, rubber sheets etc. where the quality merit of desired product depends\non the rate of stretching, external magnetic field and composition of materials\nused.", "category": "physics_flu-dyn" }, { "text": "Competition is the underlying mechanism controlling viscous fingering\n and wormhole growth: Viscous fingering and wormhole growth are complex nonlinear unstable\nphenomena. We view both as the result of competition for water in which the\ncapacity of an instability to grow depends on its ability to carry water. We\nderive empirical solutions to quantify the finger/wormhole flow rate in\nsingle-, two-, and multiple-finger systems. We use these solutions to show that\nfingering and wormhole patterns are a deterministic result of competition. For\nwormhole growth, controlled by dissolution, we solve re-active transport\nanalytically within each wormhole to compute dissolution at the wormhole walls\nand tip. The generated patterns (both for viscous fingering and wormhole growth\nunder moderate Damk ohler values) follow a power law decay of the number of\nfingers/wormholes with depth with an exponent of -1 consistent with field\nobservations", "category": "physics_flu-dyn" }, { "text": "Learning of viscosity functions in rarefied gas flows with\n physics-informed neural networks: The prediction non-equilibrium transport phenomena in disordered media is a\ndifficult problem for conventional numerical methods. An example of a\nchallenging problem is the prediction of gas flow fields through porous media\nin the rarefied regime, where resolving the six-dimensional Boltzmann equation\nor its numerical approximations is computationally too demanding.\nPhysics-informed neural networks (PINNs) have been recently proposed as an\nalternative to conventional numerical methods, but remain very close to the\nBoltzmann equation in terms of mathematical formulation. Furthermore, there has\nbeen no systematic study of neural network designs on the performance of PINNs.\nIn this work, PINNs are employed to predict the velocity field of a rarefied\ngas flow in a slit at increasing Knudsen numbers according to a generalized\nStokes phenomenological model using an effective viscosity function. We found\nthat activation functions with limited smoothness result in orders of magnitude\nlarger errors than infinitely differentiable functions and that the AdamW is by\nfar the best optimizer for this inverse problem. The design was found to be\nrobust from Knudsen numbers ranging from 0.1 to 10. Our findings stand as a\nfirst step towards the use of PINNs to investigate the dynamics of\nnon-equilibrium flows in complex geometries.", "category": "physics_flu-dyn" }, { "text": "Rayleigh-Brillouin scattering in binary mixtures of disparate-mass\n constituents: SF$_6-$He, SF$_6-$D$_2$ and SF$_6-$H$_2$: The spectral distribution of light scattered by microscopic thermal\nfluctuations in binary mixture gases was investigated experimentally and\ntheoretically. Measurements of Rayleigh-Brillouin spectral profiles were\nperformed at a wavelength of 532 nm and at room temperature, for mixtures of\nSF$_6-$He, SF$_6-$D$_2$ and SF$_6-$H$_2$. In these measurements, the pressure\nof the gases with heavy molecular mass (SF$_6$) is set at 1 bar, while the\npressure of the lighter collision partner was varied. In view of the large\npolarizability of SF$_6$ and the very small polarizabilities of He, H$_2$ and\nD$_2$, under the chosen pressure conditions these low mass species act as\nspectators and do not contribute to the light scattering spectrum, while they\ninfluence the motion and relaxation of the heavy SF$_6$ molecules. A\ngeneralized hydrodynamic model was developed that should be applicable for the\nparticular case of molecules with heavy and light disparate masses, as is the\ncase for the heavy SF$_6$ molecule, and the lighter collision partners. Based\non the kinetic theory of gases, our model replaces the classical\nNavier-Stokes-Fourier relations with constitutive equations having an\nexponential memory kernel. The energy exchange between translational and\ninternal modes of motion is included and quantified with a single parameter $z$\nthat characterizes the ratio between the mean elastic and inelastic molecular\ncollision frequencies. The model is compared with the experimental\nRayleigh-Brillouin scattering data, where the value of the parameter $z$ is\ndetermined in a least-squares procedure. Where very good agreement is found\nbetween experiment and the generalized hydrodynamic model, the computations in\nthe framework of classical hydrodynamics strongly deviate. Only in the\nhydrodynamic regime both models are shown to converge.", "category": "physics_flu-dyn" }, { "text": "Active-Subspace Analysis of Exceedance Probability for Shallow-Water\n Waves: We model shallow-water waves using a one-dimensional Korteweg-de Vries\nequation with the wave generation parameterized by random wave amplitudes for a\npredefined sea state. These wave amplitudes define the high-dimensional\nstochastic input vector for which we estimate the short-term wave crest\nexceedance probability at a reference point. For this high-dimensional and\ncomplex problem, most reliability methods fail, while Monte Carlo methods\nbecome impractical due to the slow convergence rate. Therefore, first within\noffshore applications, we employ the dimensionality reduction method called\n\\textit{Active-Subspace Analysis}. This method identifies a low-dimensional\nsubspace of the input space that is most significant to the input-output\nvariability. We exploit this to efficiently train a Gaussian process that\nmodels the maximum 10-minute crest elevation at the reference point, and to\nthereby efficiently estimate the short-term wave crest exceedance probability.\nThe active low-dimensional subspace for the Korteweg-de Vries model also\nexposes the expected incident wave groups associated with extreme waves and\nloads. Our results show the advantages and the effectiveness of the\nactive-subspace analysis against the Monte Carlo implementation for offshore\napplications.", "category": "physics_flu-dyn" }, { "text": "Influence of surface tension on two fluids shearing instability: Using extended Layzer's potential flow model, we investigate the effects of\nsurface tension on the growth of the bubble and spike in combined\nRayleigh-Taylor and Kelvin-Helmholtz instability. The nonlinear asymptotic\nsolutions are obtained analytically for the velocity and curvature of the\nbubble and spike tip. We find that the surface tension decreases the velocity\nbut does not affect the curvature, provided surface tension is greater than a\ncritical value. For a certain condition, we observe that surface tension\nstabilizes the motion. Any perturbation, whatever its magnitude, results stable\nwith nonlinear oscillations. The nonlinear oscillations depend on surface\ntension and relative velocity shear of the two fluids.", "category": "physics_flu-dyn" }, { "text": "An empirical mean-field model of symmetry-breaking in a turbulent wake: This work develops a low-dimensional nonlinear stochastic model of\nsymmetry-breaking coherent structures from experimental measurements of a\nturbulent axisymmetric bluff body wake. Traditional model reduction methods\ndecompose the field into a set of modes with fixed spatial support but\ntime-varying amplitudes. However, this fixed basis cannot resolve the mean flow\ndeformation due to variable Reynolds stresses, a central feature of Stuart's\nnonlinear stability mechanism, without the further assumption of weakly\nnonlinear interactions. Here, we introduce a parametric modal basis that\ndepends on the instantaneous value of the unsteady aerodynamic center of\npressure, which quantifies the degree to which the rotational symmetry of the\nwake is broken. Thus, the modes naturally interpolate between the unstable\nsymmetric state and the nonlinear equilibrium. We estimate the modes from\nexperimental measurements of the base pressure distribution by reducing the\nsymmetry via phase alignment and averaging conditioned on the center of\npressure. The amplitude dependence of the symmetric mode deviates significantly\nfrom the polynomial scaling predicted by weakly nonlinear analysis, confirming\nthat the parametric basis is crucial for capturing the effect of strongly\nnonlinear interactions. We also introduce a second model term capturing\naxisymmetric fluctuations associated with the mean-field deformation. We then\napply the Langevin regression system identification method to construct a\nstochastically forced nonlinear model for these two generalized mode\ncoefficients. The resulting model reproduces empirical power spectra and\nprobability distributions, suggesting a path towards developing interpretable\nlow-dimensional models of globally unstable turbulent flows from experimental\nmeasurements.", "category": "physics_flu-dyn" }, { "text": "Internal heating driven convection at infinite Prandtl number: We derive an improved rigorous bound on the space and time averaged\ntemperature $$ of an infinite Prandtl number Boussinesq fluid contained\nbetween isothermal no-slip boundaries thermally driven by uniform internal\nheating. A novel approach is used wherein a singular stable stratification is\nintroduced as a perturbation to a non-singular background profile, yielding the\nestimate $\\geq 0.419[R\\log(R)]^{-1/4}$ where $R$ is the heat Rayleigh\nnumber. The analysis relies on a generalized Hardy-Rellich inequality that is\nproved in the appendix.", "category": "physics_flu-dyn" }, { "text": "Zeitlin truncation of a Shallow Water Quasi-Geostrophic model for\n planetary flow: In this work, we consider a Shallow-Water Quasi Geostrophic equation on the\nsphere, as a model for global large-scale atmospheric dynamics. This equation,\nrecently studied by Verkley (2009) and Schubert et al. (2009), possesses a rich\ngeometric structure, called Lie-Poisson, and admits an infinite number of\nconserved quantities, called Casimirs. In this paper, we develop a Casimir\npreserving numerical method for long-time simulations of this equation. The\nmethod develops in two steps: firstly, we construct an N-dimensional\nLie-Poisson system that converges to the continuous one in the limit $N \\to\n\\infty$; secondly, we integrate in time the finite-dimensional system using an\nisospectral time integrator, developed by Modin and Viviani (2020). We\ndemonstrate the efficacy of this computational method by simulating a flow on\nthe entire sphere for different values of the Lamb parameter. We particularly\nfocus on rotation-induced effects, such as the formation of jets. In agreement\nwith shallow water models of the atmosphere, we observe the formation of robust\nlatitudinal jets and a decrease in the zonal wind amplitude with latitude.\nFurthermore, spectra of the kinetic energy are computed as a point of reference\nfor future studies.", "category": "physics_flu-dyn" }, { "text": "Data-driven subgrid-scale modeling of forced Burgers turbulence using\n deep learning with generalization to higher Reynolds numbers via transfer\n learning: Developing data-driven subgrid-scale (SGS) models for large eddy simulations\n(LES) has received substantial attention recently. Despite some success,\nparticularly in a priori (offline) tests, challenges have been identified that\ninclude numerical instabilities in a posteriori (online) tests and\ngeneralization (i.e., extrapolation) of trained data-driven SGS models, for\nexample to higher Reynolds numbers. Here, using the stochastically forced\nBurgers turbulence as the test-bed, we show that deep neural networks trained\nusing properly pre-conditioned (augmented) data yield stable and accurate a\nposteriori LES models. Furthermore, we show that transfer learning enables\naccurate/stable generalization to a flow with 10x higher Reynolds number.", "category": "physics_flu-dyn" }, { "text": "Three-dimensional low Reynolds number flows near biological filtering\n and protective layers: Mesoscale filtering and protective layers are replete throughout the natural\nworld. Within the body, arrays of extracellular proteins, microvilli, and cilia\ncan act as both protective layers and mechanosensors. For example, blood flow\nprofiles through the endothelial surface layer determine the amount of shear\nstress felt by the endothelial cells and may alter the rates at which molecules\nenter and exit the cells. Characterizing the flow profiles through such layers\nis therefore critical towards understanding the function of such arrays in cell\nsignaling and molecular filtering. External filtering layers are also important\nto many animals and plants. Trichomes (the hairs or fine outgrowths on plants)\ncan drastically alter both the average wind speed and profile near the leaf's\nsurface, affecting the rates of nutrient and heat exchange. In this paper,\ndynamically scaled physical models are used to study the flow profiles outside\nof arrays of cylinders that represent such filtering and protective layers. In\naddition, numerical simulations using the Immersed Boundary Method are used to\nresolve the 3D flows within the layers. The experimental and computational\nresults are compared to analytical results obtained by modeling the layer as a\nhomogeneous porous medium with free flow above the layer. The experimental\nresults show that the bulk flow is well described by simple analytical models.\nThe numerical results show that the spatially averaged flow within the layer is\nwell described by the Brinkman model. The numerical results also demonstrate\nthat the flow can be highly 3D with fluid moving into and out of the layer.\nThese effects are not described by the Brinkman model and may be significant\nfor biologically relevant volume fractions. The results of this paper can be\nused to understand how variations in density and height of such structures can\nalter shear stresses and bulk flows.", "category": "physics_flu-dyn" }, { "text": "Spatio-temporal spectral analysis of a forced cylinder wake: The wake of a circular cylinder performing rotary oscillations is studied\nusing hydrodynamic tunnel experiments at $Re=100$. Two-dimensional particle\nimage velocimetry on the mid-plane perpendicular to the axis of cylinder is\nused to characterize the spatial development of the flow and its stability\nproperties. The lock-in phenomenon that determines the boundaries between\nregions of the forcing parameter space were the wake is globally unstable or\nconvectively unstable is scrutinized using the experimental data. A novel\nmethod based on the analysis of power density spectra of the flow allows us to\ngive a detailed description of the forced wake, shedding light on the energy\ndistribution in the different frequency components and in particular on a\ncascade-like mechanism evidenced for a high amplitude of the forcing\noscillation. In addition, a calculation of the drag from the velocity field is\nperformed, allowing us to relate the resulting force on the body to the wake\nproperties.", "category": "physics_flu-dyn" }, { "text": "Performance Analysis of Vertical Axis Wind Turbine Clusters: Effect of\n Inter-Turbine Spacing and Turbine Rotation: Wind energy has emerged as a viable alternative to fossil fuels, with\nvertical axis wind turbines (VAWTs) gaining popularity due to their efficiency\nand adaptability. Combining the Actuator Line Method (ALM) with Large Eddy\nSimulation (LES) enables accurate performance evaluations, facilitating the\ndesign and optimization of wind turbines. The present study invokes ALM-based\nmethodology to perform calculations for the VAWTs. The results of the LES\nsimulations of the VAWTs have been extensively validated against the available\nexperimental and numerical data. The study further explores a VAWT cluster of\nthree turbines by investigating the influence of turbine spacing (in both\nin-line and staggered configuration) on cluster performance. The study shows\nthat the configuration with a streamwise separation (Xsep) of 0.34D and a\ntransverse separation (Ysep) of 2.5D exhibits superior performance to other\ncombinations owing to increased kinetic energy in the wake for the downstream\nturbines. Further, we have presented the effect of varying the rotation\ndirection (in combinations of Clockwise and Counter-Clockwise rotation) for the\nindividual turbines in the 3-turbine cluster for the two configurations:\nin-line (Xsep = 0D, Ysep = 2.5D) and staggered (Xsep = 0.34D, Ysep = 2.5D).\nStaggered counter-rotating turbine cases show reduced performance compared to\nco-rotating cases, specifically, the clockwise co-rotating (C-C-C)\nconfiguration. In the in-line configuration, counter-rotating setups outperform\nco-rotating ones. Counter-rotation analysis reveals that reducing streamwise\nseparation allows turbines to align in line without sacrificing performance,\nthereby increasing the power density of the turbine cluster", "category": "physics_flu-dyn" }, { "text": "Stability Criteria and Turbulence Paradox Problem For Type II 3D Shears: There are two types of 3D shears in channel flows: ($U(y,z),0,0$) and\n($U(y),0,W(y)$). Both are important in organizing the phase space structures of\nthe channel flows. Stability criteria of the type I 3D shears were studied in\n[Li, 2010]. Here we study the stability criteria of the type II 3D shears. We\nalso provide more support to the idea of resolution of a turbulence paradox,\nintroduced in [Li and Lin, 2010], by studying a sequence of type II 3D shears.", "category": "physics_flu-dyn" }, { "text": "Spatial extreme values of vorticity and velocity gradients in\n two-dimensional turbulent flows: We study the distribution of spatial extrema of vorticity and the determinant\nof the strain rate tensor for a two-dimensional turbulent flow forced by a\nKolmogorov forcing. The distribution of these quantities follow non-Gaussian\nbehaviour and they do not fall into the Generalised Extreme value\ndistributions. It is found that for the truncated Euler equations the spatial\nextrema of vorticity and strain rate tensor are well described by the Gumbel\ndistribution. The spatial extrema for the vorticity is found to be at the core\nof the vortices while the velocity gradients are found near the edges of the\nvortices or at the shear layers in the regions between the vortices. Temporal\ncorrelations of the velocity gradients shed light on the extreme value\ndistributions obtained for turbulence and the truncated Euler equations.", "category": "physics_flu-dyn" }, { "text": "The structure of the blue whirl revealed: The blue whirl is a small, stable, spinning blue flame that evolved\nspontaneously in laboratory experiments while studying, violent, turbulent fire\nwhirls. The blue whirl cleanly burns heavy, liquid hydrocarbon fuels with no\nsoot production, presenting a new potential way for low-emission combustion. It\nis reproducible, appears for a range of different fuels and initial conditions,\nis quiet, appears laminar, and has characteristics which led to the idea that\nit results from vortex breakdown in whirling, reacting fluid. Since its\ndiscovery, considerable effort has been put into measurements, which have shown\nits temperature structure and sensitivity to the boundary layer near the\nsurface. This has led to considerable speculation about the type of flames that\ncomprise it. Simultaneously, there was a numerical effort to study its\nstructure by performing simulations of vortex breakdown in gaseous reactive\nflows. The simulations described in this paper show that the stable blue whirl\nis composed of three different flame structures -- a diffusion flame and a\npremixed rich and lean flame -- all of which meet in a fourth structure, a\ntriple flame which appears as a whirling blue ring. In addition, the blue whirl\nstructure emerges as the result of vortex breakdown in a swirling reactive\nflow, as evidenced in the simulation by a bubble mode that is usually invisible\nin the experiments but at the center of the whirl. This paper also presents the\ntool used for the study and discusses how this might be used for future\ninvestigations.", "category": "physics_flu-dyn" }, { "text": "Extreme flow simulations reveal skeletal adaptations of deep-sea sponges: Since its discovery, the deep-sea glass sponge Euplectella aspergillum has\nattracted interest in its mechanical properties and beauty. Its skeletal system\nis composed of amorphous hydrated silica and is arranged in a highly regular\nand hierarchical cylindrical lattice that begets exceptional flexibility and\nresilience to damage. Structural analyses dominate the literature, but\nhydrodynamic fields that surround and penetrate the sponge have remained\nlargely unexplored. Here we address an unanswered question: whether, besides\nimproving its mechanical properties, the skeletal motifs of E. aspergillum\nunderlie the optimization of the flow physics within and beyond its body\ncavity. We use extreme flow simulations based on the 'lattice Boltzmann'\nmethod, featuring over fifty billion grid points and spanning four spatial\ndecades. These in silico experiments reproduce the hydrodynamic conditions on\nthe deep-sea floor where E. aspergillum lives. Our results indicate that the\nskeletal motifs reduce the overall hydrodynamic stress and support coherent\ninternal recirculation patterns at low flow velocity. These patterns are\narguably beneficial to the organism for selective filter feeding and sexual\nreproduction11,12. The present study reveals mechanisms of extraordinary\nadaptation to live in the abyss, paving the way towards further studies of this\ntype at the intersection between fluid mechanics, organism biology and\nfunctional ecology.", "category": "physics_flu-dyn" }, { "text": "Mathematical modelling and computational reduction of molten glass fluid\n flow in a furnace melting basin: In this work, we present the modelling and numerical simulation of a molten\nglass fluid flow in a furnace melting basin. We first derive a model for a\nmolten glass fluid flow and present numerical simulations based on the Finite\nElement Method (FEM). We further discuss and validate the results obtained from\nthe simulations by comparing them with experimental results. Finally, we also\npresent a non-intrusive Proper Orthogonal Decomposition (POD) based on\nArtificial Neural Networks (ANN) to efficiently handle scenarios which require\nmultiple simulations of the fluid flow upon changing parameters of relevant\nindustrial interest. This approach lets us obtain solutions of a complex 3D\nmodel, with good accuracy with respect to the FEM solution, yet with negligible\nassociated computational times.", "category": "physics_flu-dyn" }, { "text": "Mean flow in hexagonal convection: stability and nonlinear dynamics: Weakly nonlinear hexagon convection patterns coupled to mean flow are\ninvestigated within the framework of coupled Ginzburg-Landau equations. The\nequations are in particular relevant for non-Boussinesq Rayleigh-B\\'enard\nconvection at low Prandtl numbers. The mean flow is found to (1) affect only\none of the two long-wave phase modes of the hexagons and (2) suppress the\nmixing between the two phase modes. As a consequence, for small Prandtl numbers\nthe transverse and the longitudinal phase instability occur in sufficiently\ndistinct parameter regimes that they can be studied separately. Through the\nformation of penta-hepta defects, they lead to different types of transient\ndisordered states. The results for the dynamics of the penta-hepta defects shed\nlight on the persistence of grain boundaries in such disordered states.", "category": "physics_flu-dyn" }, { "text": "Uniqueness of Landau-Lifshitz Energy Frame in Relativistic Dissipative\n Hydrodynamics: We show that the relativistic dissipative hydrodynamic equation derived from\nthe relativistic Boltzmann equation by the renormalization-group method\nuniquely leads to the one in the energy frame proposed by Landau and Lifshitz,\nprovided that the macroscopic-frame vector, which defines the local rest frame\nof the fluid velocity, is independent of the momenta of constituent particles,\nas it should. We argue that the relativistic hydrodynamic equations for viscous\nfluids must be defined on the energy frame if it is consistent with the\nunderlying relativistic kinetic equation.", "category": "physics_flu-dyn" }, { "text": "Swimming of a sphere in a viscous incompressible fluid with inertia: The swimming of a sphere immersed in a viscous incompressible fluid with\ninertia is studied for surface modulations of small amplitude on the basis of\nthe Navier-Stokes equations. The mean swimming velocity and the mean rate of\ndissipation are expressed as quadratic forms in term of the surface\ndisplacements. With a choice of a basis set of modes the quadratic forms\ncorrespond to two hermitian matrices. Optimization of the mean swimming\nvelocity for given rate of dissipation requires the solution of a generalized\neigenvalue problem involving the two matrices. It is found for surface\nmodulations of low multipole order that the optimal swimming efficiency depends\nin intricate fashion on a dimensionless scale number involving the radius of\nthe sphere, the period of the cycle, and the kinematic viscosity of the fluid.", "category": "physics_flu-dyn" }, { "text": "Some exact solutions to the Lighthill Whitham Richards Payne traffic\n flow equations II: moderate congestion: We find a further class of exact solutions to the Lighthill Whitham Richards\nPayne (LWRP) traffic flow equations. As before, using two consecutive\nLagrangian transformations, a linearization is achieved. Next, depending on the\ninitial density, we either obtain exact formulae for the dependence of the car\ndensity and velocity on x, t, or else, failing that, the same result in a\nparametric representation. The calculation always involves two possible\nfactorizations of a consistency condition. Both must be considered. In physical\nterms, the lineup usually separates into two offshoots at different velocities.\nEach velocity soon becomes uniform. This outcome in many ways resembles not\nonly Rowlands, Infeld and Skorupski J. Phys. A: Math. Theor. 46 (2013) 365202\n(part I) but also the two soliton solution to the Korteweg-de Vries equation.\nThis paper can be read independently of part I. This explains unavoidable\nrepetitions. Possible uses of both papers in checking numerical codes are\nindicated at the end. Since LWRP, numerous more elaborate models, including\nmultiple lanes, traffic jams, tollgates etc. abound in the literature. However,\nwe present an exact solution. These are few and far between, other then found\nby inverse scattering. The literature for various models, including ours, is\ngiven. The methods used here and in part I may be useful in solving other\nproblems, such as shallow water flow.", "category": "physics_flu-dyn" }, { "text": "Solidification of a rivulet: shape and temperature fields: The freezing of a water rivulet begins with a water thread flowing over a\nvery cold surface, is naturally followed by the growth of an ice layer and ends\nup with a water rivulet flowing on a static thin ice wall. The structure of\nthis final ice layer presents a surprising linear shape that thickens with the\ndistance. This paper presents a theoretical model and experimental\ncharacterisation of the ice growth dynamics, the final ice shape and the\ntemperature fields. In a first part, we establish a 2D model, based on the\nadvection-diffusion heat equations, that allows us to predict the shape of the\nice structure and the temperature fields in both the water and the ice. Then,\nwe study experimentally the formation of the ice layer and we show that both\nthe transient dynamics and the final shape are well captured by the model. In a\nlast part, we characterise experimentally the temperature fields in the ice and\nin the water, using an infrared camera. The model shows an excellent agreement\nwith the experimental fields. In particular, it predicts well the linear\ndecrease of the water surface temperature observed along the plane, confirming\nthat the final ice shape is a consequence of the interaction between the\nthermal boundary layer and the free surface.", "category": "physics_flu-dyn" }, { "text": "Repetitive acoustic streaming patterns in sinusoidal shaped\n microchannels: Geometry of the fluid container plays a key role in the shape of acoustic\nstreaming patterns. Inadvertent vortices can be troublesome in some cases, but\nif treated properly, the problem turns into a very useful parameter in acoustic\ntweezing or micromixing applications. In this paper, the effects of sinusoidal\nboundaries of a microchannel on acoustic streaming patterns are studied.\nResults show that while top and bottom sinusoidal walls are vertically actuated\nat the resonance frequency of basic hypothetical rectangular microchannel, some\nrepetitive acoustic streaming patterns are recognized in classifiable cases.\nSuch patterns can never be produced in rectangular geometry with flat\nboundaries. Relations between geometrical parameters and emerging acoustic\nstreaming patterns lead us to propose formulas in order to predict more cases.\nSuch results and formulations were not trivial at a glance.", "category": "physics_flu-dyn" }, { "text": "Purely elastic turbulence in pressure-driven channel flows: Solutions of long, flexible polymer molecules are complex fluids that\nsimultaneously exhibit fluid-like and solid-like behaviour. When subjected to\nan external flow, dilute polymer solutions exhibit elastic turbulence - a\nunique, chaotic flow state absent in Newtonian fluids, like water. Unlike its\nNewtonian counterpart, elastic turbulence is caused by polymer molecules\nstretching and aligning in the flow, and can occur at vanishing inertia. While\nexperimental realisations of elastic turbulence are well-documented, there is\ncurrently no understanding of its mechanism. Here, we present large-scale\ndirect numerical simulations of elastic turbulence in pressure-driven flows\nthrough straight channels. We demonstrate that the transition to elastic\nturbulence is sub-critical, giving rise to spot-like flow structures that,\nfurther away from the transition, eventually spread throughout the domain. We\nprovide evidence that elastic turbulence is organised around unstable coherent\nstates that are localised close to the channel midplane.", "category": "physics_flu-dyn" }, { "text": "Diagnosing tracer transport in convective penetration of a stably\n stratified layer: We use large-eddy simulations to study the penetration of a buoyant plume\ncarrying a passive tracer into a stably stratified layer with constant buoyancy\nfrequency. Using a buoyancy-tracer volume distribution, we develop a method for\nobjectively partitioning buoyancy-tracer space into three regions, each of\nwhich corresponds to a coherent region in physical space. Specifically, we\nidentify a source region where undiluted plume fluid enters the stratified\nlayer, a transport region where much of the transition from undiluted to mixed\nfluid occurs in the plume cap, and an accumulation region corresponding to the\nradially spreading intrusion. This method enables quantification of different\nmeasures of turbulence and mixing within each of the three regions, including\npotential energy and turbulent kinetic energy dissipation rates, an activity\nparameter, and the instantaneous mixing efficiency. We find that the most\nintense buoyancy gradients lie in a thin layer at the cap of the penetrating\nplume. This forms the primary stage of mixing between plume and environment and\nexhibits a mixing efficiency around 50%. Newly generated mixtures of\nenvironmental and plume fluid joining the intrusion are subjected to relatively\nweak turbulence and weaker buoyancy gradients as mixtures are homogenised. As\nthe intrusion spreads radially, environmental fluid surrounding the intrusion\nis mixed into the intrusion with moderate mixing efficiency. This dominates the\ntotal entrainment of environmental fluid into the plume as a whole. However,\nthe 'strongest' entrainment, as measured by the specific entrainment rate, is\nlargest in the plume cap where the most buoyant environmental fluid is\nentrained.", "category": "physics_flu-dyn" }, { "text": "A note on forces exerted by a Stokeslet on confining boundaries: We consider a Stokeslet applied to a viscous fluid next to an infinite, flat\nwall, or in-between two parallel walls. We calculate the forces exerted by the\nresulting flow on the confining boundaries, and use the results obtained to\nestimate the hydrodynamic contribution to the pressure exerted on boundaries by\nforce-free self-propelled particles.", "category": "physics_flu-dyn" }, { "text": "The Transition from Wall Modes to Multimodality in Liquid Gallium\n Magnetoconvection: Coupled laboratory-numerical experiments of Rayleigh-B\\'enard convection\n(RBC) in liquid gallium subject to a vertical magnetic field are presented. The\nexperiments are carried out in two cylindrical containers with\ndiameter-to-height aspect ratio $\\Gamma = 1.0$ and $2.0$ at varying thermal\nforcing (Rayleigh numbers $10^5 \\lesssim Ra \\lesssim 10^8$) and magnetic field\nstrength (Chandrasekhar numbers $0\\lesssim Ch \\lesssim 3\\times 10^5$).\nLaboratory measurements and numerical simulations confirm that\nmagnetoconvection in our finite cylindrical tanks onsets via non-drifting\nwall-attached modes, in good agreement with asymptotic predictions for a\nsemi-infinite domain. With increasing supercriticality, the experimental and\nnumerical thermal measurements and the numerical velocity data reveal\ntransitions between wall mode states with different azimuthal mode numbers and\nbetween wall-dominated convection to wall and interior multimodality. These\ntransitions are also reflected in the heat transfer data, which combined with\nprevious studies, connect onset to supercritical turbulent behaviors in liquid\nmetal magnetoconvection over a large parameter space. The gross heat transfer\nbehaviors between magnetoconvection and rotating convection in liquid metals\nare compared and discussed.", "category": "physics_flu-dyn" }, { "text": "An analytical study of the MHD clamshell instability on a sphere: This paper studies the instability of two-dimensional magnetohydrodynamic\n(MHD) systems on a sphere using analytical methods. The underlying flow\nconsists of a zonal differential rotation and a toroidal magnetic field is\npresent. Semicircle rules that prescribe the possible domain of the wave\nvelocity in the complex plane for general flow and field profiles are derived.\nThe paper then sets out an analytical study of the `clamshell instability',\nwhich features field lines on the two hemispheres tilting in opposite\ndirections (Cally 2001, Sol. Phys. vol. 199, pp. 231--249). An asymptotic\nsolution for the instability problem is derived for the limit of weak shear of\nthe zonal flow, via the method of matched asymptotic expansions. It is shown\nthat when the zonal flow is solid body rotation, there exists a neutral mode\nthat tilts the magnetic field lines, referred to as the `tilting mode'. A weak\nshear of the zonal flow excites the critical layer of the tilting mode, which\nreverses the tilting direction to form the clamshell pattern and induces the\ninstability. The asymptotic solution provides insights into properties of the\ninstability for a range of flow and field profiles. A remarkable feature is\nthat the magnetic field affects the instability only through its local\nbehaviour in the critical layer.", "category": "physics_flu-dyn" }, { "text": "Free surface water-waves generated by instability of an exponential\n shear flow: The stability of an exponential current in water to infinitesimal\nperturbations in the presence of gravity and capillarity is investigated. Some\nnew results on the generation of gravity-capillary waves are presented which\nsupplement the previous works of Morland, Saffman \\& Yuen (1991) and Young \\&\nWolfe (2014), namely in finite depth. To consider perturbations of much larger\nscales, a specific attention is paid to the stability of the exponential\ncurrent only in the presence of gravity.", "category": "physics_flu-dyn" }, { "text": "Physical invariance in neural networks for subgrid-scale scalar flux\n modeling: In this paper we present a new strategy to model the subgrid-scale scalar\nflux in a three-dimensional turbulent incompressible flow using\nphysics-informed neural networks (NNs). When trained from direct numerical\nsimulation (DNS) data, state-of-the-art neural networks, such as convolutional\nneural networks, may not preserve well known physical priors, which may in turn\nquestion their application to real case-studies. To address this issue, we\ninvestigate hard and soft constraints into the model based on classical\ntransformation invariances and symmetries derived from physical laws. From\nsimulation-based experiments, we show that the proposed\ntransformation-invariant NN model outperforms both purely data-driven ones as\nwell as parametric state-of-the-art subgrid-scale models. The considered\ninvariances are regarded as regularizers on physical metrics during the a\npriori evaluation and constrain the distribution tails of the predicted\nsubgrid-scale term to be closer to the DNS. They also increase the stability\nand performance of the model when used as a surrogate during a large-eddy\nsimulation. Moreover, the transformation-invariant NN is shown to generalize to\nregimes that have not been seen during the training phase.", "category": "physics_flu-dyn" }, { "text": "Comparison of Wide and Compact Fourth Order Formulations of the\n Navier-Stokes Equations: In this study the numerical performances of wide and compact fourth order\nformulation of the steady 2-D incompressible Navier-Stokes equations will be\ninvestigated and compared with each other. The benchmark driven cavity flow\nproblem will be solved using both wide and compact fourth order formulations\nand the numerical performances of both formulations will be presented and also\nthe advantages and disadvantages of both formulations will be discussed.", "category": "physics_flu-dyn" }, { "text": "On physically redundant and irrelevant features when applying Lie-group\n symmetry analysis to hydrodynamic stability analysis: Every linear system of partial differential equations (PDEs) admits a scaling\nsymmetry in its dependent variables. In conjunction with other admitted\nsymmetries of linear type, the associated invariant solution condition poses a\nlinear eigenvalue problem. If this problem is structured such that the spectral\ntheorem applies, then the general solution of the considered linear PDE system\nis obtained by summing or integrating the invariant eigenfunctions (modes) over\nall eigenvalues, depending on whether the spectrum of the operator is discrete\nor continuous. By first studying the 1-D diffusion equation as a demonstrating\nexample, this method is then applied to a relevant 2-D problem from\nhydrodynamic stability analysis. The aim of this study is to draw attention to\nthe following two independent facts that need to be addressed in future studies\nwhen constructing solutions for linear PDEs with the method of Lie-symmetries:\n(i) Although each new symmetry leads to a mathematically different spectral\ndecomposition, they may all be physically redundant to standard ones and do not\nreveal a new physical mechanism behind the overall considered dynamical\nprocess, as incorrectly asserted, for example, in the recent studies by the\ngroup of Oberlack et al. Hence, with regard to linear stability analysis, no\nphysically \"new\" or more \"general\" modes are generated by this method than the\nones already established. (ii) Next to the eigenvalue parameters, each single\nmode can also acquire non-system parameters, depending on the choice of its\nunderlying symmetry. These symmetry-induced parameters, however, are all\nphysically irrelevant, since their effect on a single mode will cancel when\nconsidering all modes collectively. In particular, the collective action of all\nsingle modes is identical for all symmetry-based decompositions and thus\nindistinguishable when considering the full physical fields.", "category": "physics_flu-dyn" }, { "text": "A model of laminated wave turbulence: A model of laminated wave turbulence is presented. This model consists of two\nco-existing layers - one with continuous waves' spectra, covered by KAM theory\nand Kolmogorov-like power spectra, and one with discrete waves' spectra,\ncovered by discrete classes of waves and Clipping method. Some known laboratory\nexperiments and numerical simulations are explained in the frame of this model.", "category": "physics_flu-dyn" }, { "text": "Low-frequency resolvent analysis of the laminar oblique shock wave /\n boundary layer interaction: Resolvent analysis is used to study the low-frequency behaviour of the\nlaminar oblique shock wave / boundary layer interaction (SWBLI). It is shown\nthat the computed optimal gain, which can be seen as a transfer function of the\nsystem, follows a first-order low-pass filter equation, recovering the results\nof Touber and Sandham (JFM, 2011). This behaviour is understood as proceeding\nfrom the excitation of a single stable, steady global mode whose damping rate\nsets the time scale of the filter. Different Mach and Reynolds numbers are\nstudied, covering different recirculation lengths $L$. This damping rate is\nfound to scale as $1/L$, leading to a constant Strouhal number $St_L$ as\nobserved in the literature. It is associated with a breathing motion of the\nrecirculation bubble. This analysis furthermore supports the idea that the\nlow-frequency dynamics of the SWBLI is a forced dynamics, in which background\nperturbations continuously excite the flow. The investigation is then carried\nout for 3D perturbations for which two regimes are identified. At low wave\nnumbers of the order of $L$, a modal mechanism similar to that of 2D\nperturbations is found and exhibits larger values of the optimal gain. At\nlarger wave numbers of the order of the boundary layer thickness, the growth of\nstreaks, which results from a non-modal mechanism, is detected. No interaction\nwith the recirculation region is observed. Based on these results, the\npotential prevalence of 3D effects in the low-frequency dynamics of the SWBLI\nis discussed.", "category": "physics_flu-dyn" }, { "text": "Influence of Magnetic Force on the Flow Stability in a Rectangular duct: The stability of the flow under the magnetic force is one of the classical\nproblems in fluid mechanics. In this paper, the flow in a rectangular duct with\ndifferent Hartmann (Ha) number is simulated. The finite volume method and the\nSIMPLE algorithm are used to solve a system of equations and the energy\ngradient theory is then used to study the (associated) stability of\nmagnetohydrodynamics (MHD). The flow stability of MHD flow for different\nHartmann (Ha) number, from Ha=1 to 40, at the fixed Reynolds number, Re=190 are\ninvestigated. The simulation is validated firstly against the simulation in\nliterature. The results show that, with the increasing Ha number, the\ncenterline velocity of the rectangular duct with MHD flow decreases and the\nabsolute value of the gradient of total mechanical energy along the streamwise\ndirection increases. The maximum of K appears near the wall in both coordinate\naxis of the duct. According to the energy gradient theory, this position of the\nmaximum of K would initiate flow instability (if any) than the other positions.\nThe higher the Hartmann number is, the smaller the K value becomes, which means\nthat the fluid becomes more stable in the presence of higher magnetic force. As\nthe Hartmann number increases, the K value in the parallel layer decreases more\nsignificantly than in the Hartmann layer. The most dangerous position of\ninstability tends to migrate towards wall of the duct as the Hartmann number\nincreases. Thus, with the energy gradient theory, the stability or instability\nin the rectangular duct can be controlled by modulating the magnetic force.", "category": "physics_flu-dyn" }, { "text": "Axial flow fan performance in a forced draught air-cooled heat exchanger\n for a sCO2 Brayton cycle: An axial flow cooling fan has been designed for use in a concentrated solar\npower plant. The plant is based on a supercritical carbon dioxide (sCO2)\nBrayton cycle, and uses a forced draft air-cooled heat exchanger (ACHE) for\ncooling. The fan performance has been investigated using both computational\nfluid dynamics (CFD) and scaled fan tests. This paper presents a CFD model that\nintegrates the fan with the heat exchanger. The objective is to establish a\nfoundation for similar models and to contribute to the development of efficient\nACHE units designed for sCO2 power cycles. The finned-tube bundle is\nsimplified, with a Porous Media Model representing the pressure drop through\nthe bundle. Pressure inlet and -outlet boundary conditions are used, meaning\nthe air flow rate is solved based on the fan and tube bundle interaction. The\nflow rate predicted by the CFD model is 0.5% higher than the analytical\nprediction, and 3.6% lower than the design value, demonstrating that the\nassumptions used in the design procedure are reasonable. The plenum height is\nalso found to affect the flow rate, with shorter plenums resulting in higher\nflow rates and fan efficiencies, and longer plenums resulting in more uniform\ncooling air flow.", "category": "physics_flu-dyn" }, { "text": "Thermocapillary-driven fluid flow within microchannels: Surface tension gradients induce Marangoni flow, which may be exploited for\nfluid transport. At the micrometer scale, these surface-driven flows can be\nmore significant than those driven by pressure. By introducing fluid-fluid\ninterfaces on the walls of microfluidic channels, we use surface tension\ngradients to drive bulk fluid flows. The gradients are specifically induced\nthrough thermal energy, exploiting the temperature dependence of a fluid-fluid\ninterface to generate thermocapillary flow. In this report, we provide the\ndesign concept for a biocompatible, thermocapillary microchannel capable of\nbeing powered by solar irradiation. Using temperature gradients on the order of\ndegrees Celsius per centimeter, we achieve fluid velocities on the order of\nmillimeters per second. Following experimental observations, fluid dynamic\nmodels, and numerical simulation, we find that the fluid velocity is linearly\nproportional to the provided temperature gradient, enabling full control of the\nfluid flow within the microchannels.", "category": "physics_flu-dyn" }, { "text": "Standing shock prevents propagation of sparks in supersonic explosive\n flows: Volcanic jet flows in explosive eruptions emit radio frequency signatures,\nindicative of their fluid dynamic and electrostatic conditions. The emissions\noriginate from sparks supported by an electric field built up by the ejected\ncharged volcanic particles. When shock-defined, low-pressure regions confine\nthe sparks, the signatures may be limited to high-frequency content\ncorresponding to the early components of the avalanche-streamer-leader\nhierarchy. Here, we image sparks and a standing shock together in a transient\nsupersonic jet of micro-diamonds entrained in argon. Fluid dynamic and kinetic\nsimulations of the experiment demonstrate that the observed sparks originate\nupstream of the standing shock. The sparks are initiated in the rarefaction\nregion, and cut off at the shock, which would limit their radio frequency\nemissions to a tell-tale high-frequency regime. We show that sparks transmit an\nimpression of the explosive flow, and open the way for novel instrumentation to\ndiagnose currently inaccessible explosive phenomena.", "category": "physics_flu-dyn" }, { "text": "An approach to the Riemann problem in the light of a reformulation of\n the state equation for SPH inviscid ideal flows: a highlight on spiral\n hydrodynamics in accretion discs: In physically inviscid fluid dynamics, \"shock capturing\" methods adopt either\nan artificial viscosity contribution or an appropriate Riemann solver\nalgorithm. These techniques are necessary to solve the strictly hyperbolic\nEuler equations if flow discontinuities (the Riemann problem) are to be solved.\nA necessary dissipation is normally used in such cases. An explicit artificial\nviscosity contribution is normally adopted to smooth out spurious heating and\nto treat transport phenomena. Such a treatment of inviscid flows is also widely\nadopted in the Smooth Particle Hydrodynamics (SPH) finite volume free\nLagrangian scheme. In other cases, the intrinsic dissipation of Godunov-type\nmethods is implicitly useful. Instead \"shock tracking\" methods normally use the\nRankine-Hugoniot jump conditions to solve such problems. A simple, effective\nsolution of the Riemann problem in inviscid ideal gases is here proposed, based\non an empirical reformulation of the equation of state (EoS) in the Euler\nequations in fluid dynamics, whose limit for a motionless gas coincides with\nthe classical EoS of ideal gases. The application of such an effective solution\nto the Riemann problem excludes any dependence, in the transport phenomena, on\nparticle smoothing resolution length $h$ in non viscous SPH flows. Results on\n1D shock tube tests, as well as examples of application for 2D turbulence and\n2D shear flows are here shown. As an astrophysical application, a much better\nidentification of spiral structures in accretion discs in a close binary (CB),\nas a result of this reformulation is also shown here.", "category": "physics_flu-dyn" }, { "text": "Supercriticality to subcriticality in dynamo transitions: Evidence from numerical simulations suggest that the nature of dynamo\ntransition changes from supercritical to subcritical as the magnetic Prandtl\nnumber is decreased. To explore this interesting crossover we first use direct\nnumerical simulations to investigate the hysteresis zone of a subcritical\nTaylor-Green dynamo. We establish that a well defined boundary exists in this\nhysteresis region which separates dynamo states from the purely hydrodynamic\nsolution. We then propose simple dynamo models which show similar crossover\nfrom supercritical to subcritical dynamo transition as a function of the\nmagnetic Prandtl number. Our models show that the change in the nature of\ndynamo transition is connected to the stabilizing or de-stabilizing influence\nof governing non-linearities.", "category": "physics_flu-dyn" }, { "text": "Fluid dynamics in the spirit of Cartan: A coordinate-free formulation of\n fluid dynamics for an inviscid fluid in inertial and non-inertial frames: Using Cartan's exterior calculus, we derive a coordinate-free formulation of\nthe Euler equations. These equations are invariant under Galileian\ntransformations, which constitute a global symmetry. With the introduction of\nan appropriate generalized Coriolis force, these equations become symmetric\nunder general coordinate transformations.\n We show how exterior calculus simplifies dramatically the derivation of\nconservation laws. We also discuss the advantage of an exterior calculus\nformulation with respect to symmetry-preserving discretizations of the\nequations.", "category": "physics_flu-dyn" }, { "text": "Effect of internal friction on the coil-stretch transition in turbulent\n flows: A polymer in a turbulent flow undergoes the coil-stretch transition when the\nWeissenberg number, i.e. the product of the Lyapunov exponent of the flow and\nthe relaxation time of the polymer, surpasses a critical value. The effect of\ninternal friction on the transition is studied by means of Brownian dynamics\nsimulations of the elastic dumbbell model in a homogeneous and isotropic,\nincompressible, turbulent flow and analytical calculations for a stochastic\nvelocity gradient. The results are explained by adapting the large deviations\ntheory of Balkovsky et al. [Phys. Rev. Lett., 2000, 84, 4765] to an elastic\ndumbbell with internal viscosity. In turbulent flows, a distinctive feature of\nthe probability distribution of polymer extensions is its power-law behaviour\nfor extensions greater than the equilibrium length and smaller than the contour\nlength. It is shown that although internal friction does not modify the\ncritical Weissenberg number for the coil-stretch transition, it makes the slope\nof the probability distribution steeper, thus rendering the transition sharper.\nInternal friction therefore provides a possible explanation for the steepness\nof the distribution of polymer extensions observed in experiments at large\nWeissenberg numbers.", "category": "physics_flu-dyn" }, { "text": "Machine Learning model for gas-liquid interface reconstruction in CFD\n numerical simulations: The volume of fluid (VoF) method is widely used in multi-phase flow\nsimulations to track and locate the interface between two immiscible fluids. A\nmajor bottleneck of the VoF method is the interface reconstruction step due to\nits high computational cost and low accuracy on unstructured grids. We propose\na machine learning enhanced VoF method based on Graph Neural Networks (GNN) to\naccelerate the interface reconstruction on general unstructured meshes. We\nfirst develop a methodology to generate a synthetic dataset based on paraboloid\nsurfaces discretized on unstructured meshes. We then train a GNN based model\nand perform generalization tests. Our results demonstrate the efficiency of a\nGNN based approach for interface reconstruction in multi-phase flow simulations\nin the industrial context.", "category": "physics_flu-dyn" }, { "text": "Wavelet analysis of Wave motion: In this paper high resolution wave probe records are examined using wavelet\ntechniques with a view to determining the sources and relative contributions of\ncapillary wave energy along representative wind wave forms. Wavelets enable\ncomputations of conditional spectra and turn out to be powerful tools for the\nstudy of the development and propagation of capillary waves. They also enable\nthe detailed analyses of the relative contributions to the spectrum of the wave\npeaks and troughs.", "category": "physics_flu-dyn" }, { "text": "A POD-Galerkin reduced order model of a turbulent convective buoyant\n flow of sodium over a backward-facing step: A Finite-Volume based POD-Galerkin reduced order modeling strategy for\nsteady-state Reynolds averaged Navier--Stokes (RANS) simulation is extended for\nlow-Prandtl number flow. The reduced order model is based on a full order model\nfor which the effects of buoyancy on the flow and heat transfer are\ncharacterized by varying the Richardson number. The Reynolds stresses are\ncomputed with a linear eddy viscosity model. A single gradient diffusion\nhypothesis, together with a local correlation for the evaluation of the\nturbulent Prandtl number, is used to model the turbulent heat fluxes. The\ncontribution of the eddy viscosity and turbulent thermal diffusivity fields are\nconsidered in the reduced order model with an interpolation based data-driven\nmethod. The reduced order model is tested for buoyancy-aided turbulent liquid\nsodium flow over a vertical backward-facing step with a uniform heat flux\napplied on the wall downstream of the step. The wall heat flux is incorporated\nwith a Neumann boundary condition in both the full order model and the reduced\norder model. The velocity and temperature profiles predicted with the reduced\norder model for the same and new Richardson numbers inside the range of\nparameter values are in good agreement with the RANS simulations. Also, the\nlocal Stanton number and skin friction distribution at the heated wall are\nqualitatively well captured. Finally, the reduced order simulations, performed\non a single core, are about $10^5$ times faster than the RANS simulations that\nare performed on eight cores.", "category": "physics_flu-dyn" }, { "text": "Hidden scale invariance of intermittent turbulence in a shell model: It is known that scale invariance is broken in the developed hydrodynamic\nturbulence due to intermittency, substantiating complexity of turbulent flows.\nHere we challenge the concept of broken scale invariance by establishing a\nhidden self-similarity in intermittent turbulence. Using a simplified (shell)\nmodel, we derive a nonlinear spatiotemporal scaling symmetry of inviscid\nequations, which are reformulated in terms of intrinsic times introduced at\ndifferent scales of motion. Numerical analysis persuasively confirms that this\nsymmetry is restored in a statistical sense within the inertial interval. At\nthe end, we discuss implications of this result for the Navier-Stokes system.", "category": "physics_flu-dyn" }, { "text": "Investigation of the transfer and dissipation of energy in isotropic\n turbulence: A parallel pseudospectral code for the direct numerical simulation (DNS) of\nisotropic turbulence has been developed. The code has been extensively\nbenchmarked using established results from literature.\n The code has been used to conduct a series of runs for freely-decaying\nturbulence. We explore the use of power-law decay of the total energy to\ndetermine an evolved time and compare with the use of dynamic quantities such\nas the peak dissipation rate, maximum transport power and velocity derivative\nskewness.\n Stationary turbulence has also been investigated, where we ensure that the\nenergy input rate remains constant for all runs. We present results for\nReynolds numbers up to R{\\lambda} = 335 on a 1024^3 lattice. An exploitation of\nthe pseudospectral technique is used to calculate second and third-order\nstructure functions from the energy and transfer spectra, with a comparison\npresented to the real-space calculation. An alternative to ESS is discussed,\nwith the second-order exponent found to approach 2/3.\n The dissipation anomaly is considered for forced and free-decay. The\nK\\'arm\\'an-Howarth equation (KHE) is studied and a derivation of a new work\nterm presented. The balance of energy represented by the KHE is then\ninvestigated. Based on the KHE, we develop a model for the behaviour of the\ndimensionless dissipation coefficient that predicts C{\\epsilon} =\nC{\\epsilon}(\\infty) + C_L/R_L, with C{\\epsilon}(\\infty) = 0.47 and C_L = 19.1\nobtained from DNS data.\n Theoretical methods based on RG and statistical closures are still being\ndeveloped to study turbulence. The dynamic RG procedure used by Forster, Nelson\nand Stephen (FNS) is considered in some detail and a disagreement in the\nliterature is resolved here. The application of statistical closure and\nrenormalized perturbation theory is discussed and a new two-time model\nprobability density functional presented.", "category": "physics_flu-dyn" }, { "text": "On the Relaxation of Turbulence at High Reynolds Numbers: Turbulent motions in a fluid relax at a certain rate once stirring has\nstopped. The role of the most basic parameter in fluid mechanics, the Reynolds\nnumber, in setting the relaxation rate is not generally known. This paper\nconcerns the high-Reynolds-number limit of the process. In a classical\ngrid-turbulence wind-tunnel experiment that both reached higher Reynolds\nnumbers than ever before and covered a wide range of them ($10^4 < Re = UM/\\nu\n< 5\\times10^6$), we measured the relaxation rate with the unprecedented\nprecision of about 2\\%. Here $U$ is the mean speed of the flow, $M$ the forcing\nscale, and $\\nu$ the kinematic viscosity of the fluid. We observed that the\nrelaxation rate was Reynolds-number independent, which contradicts some models\nand supports others.", "category": "physics_flu-dyn" }, { "text": "Streamwise oscillation of spanwise velocity at the wall of a channel for\n turbulent drag reduction: Steady forcing at the wall of a channel flow is studied via DNS to assess its\nability of yielding reductions of turbulent friction drag. The wall forcing\nconsists of a stationary distribution of spanwise velocity that alternates in\nthe streamwise direction. The idea behind the forcing builds upon the existing\ntechnique of the spanwise wall oscillation, and exploits the convective nature\nof the flow to achieve an unsteady interaction with turbulence.\n The analysis takes advantage of the equivalent laminar flow, that is solved\nanalytically to show that the energetic cost of the forcing is unaffected by\nturbulence. In a turbulent flow, the alternate forcing is found to behave\nsimilarly to the oscillating wall; in particular an optimal wavelength is found\nthat yields a maximal reduction of turbulent drag. The energetic performance is\nsignificantly improved, with more than 50% of maximum friction saving at large\nintensities of the forcing, and a net energetic saving of 23% for smaller\nintensities.\n Such a steady, wall-based forcing may pave the way to passively interacting\nwith the turbulent flow to achieve drag reduction through a suitable\ndistribution of roughness, designed to excite a selected streamwise wavelength.", "category": "physics_flu-dyn" }, { "text": "Optimizing wave-generation and wave-damping in 3D-flow simulations with\n implicit relaxation-zones: In finite-volume-based flow-simulations with free-surface waves, wave\nreflections at the domain boundaries can cause substantial errors in the\nresults and must therefore be minimized. This can be achieved via `implicit\nrelaxation zones', but only if the relaxation zone's case-dependent parameters\nare optimized. This work proposes an analytical approach for optimizing these\nparameters. The analytical predictions are compared against results from\n2D-flow simulations for different water depths, flow solvers, and relaxation\nfunctions, and against results from 3D-flow simulations with strongly\nwave-reflecting bodies subjected to nonlinear free-surface waves. The present\nresults demonstrate that the proposed approach satisfactorily predicts both the\noptimum parameter settings and the upper-limit for the corresponding reflection\ncoefficients $C_{\\mathrm{R}}$. Simulation results for $C_{\\mathrm{R}}$ were\nmostly below or equal to the analytical predictions, but never more than\n$3.4\\%$ larger. Therefore, the proposed approach can be recommended for\nengineering practice. Furthermore, it is shown that implicit relaxation zones\ncan be considered as a special-case of forcing zones, a family of approaches\nwhich includes among others absorbing layers, damping zones and sponge layers.\nThe commonalities and differences between these approaches are discussed,\nincluding to what extend the present findings are applicable to these other\napproaches and vice versa.", "category": "physics_flu-dyn" }, { "text": "Spatiotemporal Superresolution Measurement based on POD and Sparse\n Regression applied to a Supersonic Jet measured by PIV and Near-field\n Microphone: The present study proposed the framework of the spatiotemporal\nsuperresolution measurement based on the sparse regression with dimensionality\nreduction using the proper orthogonal decomposition (POD). The\nnon-time-resolved particle image velocimetry (PIV) and the time-resolved\nnear-field acoustic measurements using microphones were simultaneously\nperformed for a Mach 1.35 supersonic jet. POD is applied to PIV and microphone\ndata matrices and the sparse linear regression model of the reduced-order data\nis calculated using the least absolute shrinkage and selection operator\nregression. The effects of the hyperparameters of the superresolution\nmeasurement were quantitatively evaluated through randomized cross-validation.\nThe superresolved velocity field indicated the smooth convection of the\nvelocity fluctuations associated with the screech tone, while the convection of\nthe large-scale structures at the downstream side was not observed. The\nproposed framework can reconstruct the unsteady fluctuation with multiple\nfrequency phenomena, although the reconstruction is limited to the phenomena\nthat are associated with the microphone output.", "category": "physics_flu-dyn" }, { "text": "Machine learning based surrogate models for microchannel heat sink\n optimization: Microchannel heat sinks are an efficient cooling method for semiconductor\npackages. However, to properly cool increasingly complex and thermally dense\ncircuits, microchannel designs should be improved and expanded on. In this\npaper, microchannel designs with secondary channels and with ribs are\ninvestigated using computational fluid dynamics and are coupled with a\nmulti-objective optimization algorithm to determine and propose optimal\nsolutions based on observed thermal resistance and pumping power. A workflow\nthat combines Latin hypercube sampling, machine learning-based surrogate\nmodeling and multi-objective optimization is proposed. Random forests, gradient\nboosting algorithms and neural networks were considered during the search for\nthe best surrogate. We demonstrated that tuned neural networks can make\naccurate predictions and be used to create an acceptable surrogate model.\nOptimized solutions show a negligible difference in overall performance when\ncompared to the conventional optimization approach. Additionally, solutions are\ncalculated in one-fifth of the original time. Generated designs attain\ntemperatures that are lower by more than 10% under the same pressure limits as\na convectional microchannel design. When limited by temperature, pressure drops\nare reduced by more than 25%. Finally, the influence of each design variable on\nthe thermal resistance and pumping power was investigated by employing the\nSHapley Additive exPlanations technique. Overall, we have demonstrated that the\nproposed framework has merit and can be used as a viable methodology in\nmicrochannel heat sink design optimization.", "category": "physics_flu-dyn" }, { "text": "Microfluidic methods to form artificial cells and to study basic\n functions of membranes: Petra S. Dittrich is associate professor for Bioanalytics at the Department\nof Biosystems Science and Engineering at ETH Z\\\"urich. Here she describes the\nmicrofluidic devices that her lab develops to facilitate comprehensive studies\non membranes with high resolution imaging techniques.", "category": "physics_flu-dyn" }, { "text": "Clustering of passive impurities in MHD turbulence: The transport of heavy, neutral or charged, point-like particles by\nincompressible, resistive magnetohydrodynamic (MHD) turbulence is investigated\nby means of high-resolution numerical simulations. The spatial distribution of\nsuch impurities is observed to display strong deviations from homogeneity, both\nat dissipative and inertial range scales. Neutral particles tend to cluster in\nthe vicinity of coherent vortex sheets due to their viscous drag with the flow,\nleading to the simultaneous presence of very concentrated and almost empty\nregions. The signature of clustering is different for charged particles. These\nexhibit in addition to the drag the Lorentz-force. The regions of spatial\ninhomogeneities change due to attractive and repulsive vortex sheets. While\nsmall charges increase clustering, larger charges have a reverse effect.", "category": "physics_flu-dyn" } ]