id stringlengths 36 36 | text stringlengths 1 1k | source stringlengths 42 191 | page int64 0 1.03k |
|---|---|---|---|
eab74cac-ab2f-44f3-9dd3-04cd66787f16 | [10]S.Lopaciuk,R.D.Maria,G.Iadarola,andF.V.derVeken,
“Empowering a broad and diverse community in beam dy-
namicssimulationswithXsuite”,in Proc.IPAC’25 ,Taipei,
Taiwan, pp. 1702–1705, Nov. 2025.
doi:1/zero.alt4.18429/JACoW-IPAC2/zero.alt425-WEBN117th International Particle Accelerator Conference, Deauville, France JACo... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5145.pdf | 3 |
43136518-54b6-46bb-b2e1-12923f92db05 | PHASE SPACE TOMOGRAPHY CONSTRAINED BY THE
VLASOV–FOKKER–PLANCK EQUATION
USING EOSD DIAGNOSTICS AT KARA
F. Donoso∗, M. Brosi, E. Bründermann, M. Frank, S. Funkner, J. Gethmann, A.-S. Müller,
G. Niehues, M. Reissig, P. Schreiber
Karlsruhe Institute of Technology, Karlsruhe, Germany
Abstract
This study presents the first ... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 0 |
3efe380a-0136-4bf9-a08b-6e963fa124d1 | Usingthisframework,wesuccessfullyreconstructthelon-
gitudinal phase space of the electron bunch across different
dynamicalregimes,rangingfromthestablestatetotheonset
ofmicro-bunching. Fromthereconstructedphasespace, we
derive macroscopic beam observables including the longitu-
dinalbunchlength,energy-spreadrelatedhoriz... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 0 |
ca72ad76-a45a-496c-9de6-725f12689af4 | spectral decoding (EOSD) measurements acquired at the
KarlsruheResearchAccelerator(KARA).Thispartialdif-
ferential equation-constrained framework enforces physical
consistency between the reconstructed initial density and its
turn-by-turnevolution. AstheEOSDmodulationencodes
the electric field induced by the bunch with... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 0 |
2906dfe6-b07b-49ed-9e2d-b14062fdb8e3 | retardations,Γ(𝑡𝑛), obtained from this trajectory at time, 𝑡𝑛,
and the measured signals, ¯Γ𝑛for𝑛=1,...,𝑚 . The regular-
ization term, 𝜆𝑅(𝜓), is incorporated to mitigate challenges
∗felipe.donoso@kit.edusuch as noise in the measurement data and the ill-posed
nature of the inverse problem. The temporal evolutio... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 0 |
46aa7fcf-3cbb-4017-af05-bd41f58c5b2e | sponse [3,4].
min
𝜓∗≥0𝑚∑︁
𝑛=1∥Γ(𝑡𝑛)−¯Γ𝑛∥2+𝜆𝑅(𝜓∗)
s.t.𝜕𝜓
𝜕𝑡+𝜕𝐻
𝜕𝑝𝜕𝜓
𝜕𝑞−𝜕𝐻
𝜕𝑞𝜕𝜓
𝜕𝑝=𝛽𝑑𝜕
𝜕𝑝(𝑝𝜓)+𝐷𝜕2𝜓
𝜕𝑝2,
𝜌(𝑞,𝑡)=∫
𝜓(𝑞,𝑝,𝑡)d𝑝,
Γ(𝑡𝑛)=𝐺(𝜌(𝑞,𝑡𝑛)),
𝜓∗=𝜓(𝑞,𝑝,𝑡 1).
(1)
To solve Equation (1), the continuous variables are dis-
cretized. This is achieved by solving the ... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 0 |
6d7f95e4-930d-40c1-b3ae-a521d328964a | convolutionofthebunchprofilewiththesystem’simpulse
response. By combining these discrete dynamic and mea-
surement operators, the continuous problem in Equation (1)
is recast into the matrix-based optimization problem [2]
shown in Eq. (2).
min
𝜓∗∥˜G𝜓−𝚪∥2+𝜆𝑅(𝜓)
s.t.M(𝜓(𝑠,𝑡 𝑛))=D·R K·RD·K(𝜓(𝑠,𝑡𝑛)),
˜G=©... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 0 |
224d8de6-a858-4028-a1d7-9cb03bcc4796 | In Eq.(2),˜Gdenotes the extended system matrix, built
iteratively turn-by-turn. It combines G, the Toeplitz con-
volution matrix for the EOSD impulse response, W, the
projection matrix from phase space to bunch profile, and M,
thetransfermatrixgoverningtheturn-by-turnevolutionof
the phase space.
The transfer matrix Min... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 1 |
2b18fb28-c1c7-4eb5-b7a5-b4a8476dc85a | Theprofilefromtheprecedingstepisthereforeusedasan
approximationtocomputethewakepotentialoperatorand
the transfer matrix Mfor the current turn. The resulting
least-squares optimization problem in Eq. (2)is then solved
usingtheprojected-restartednon-negativeconjugategradi-
ent method for the normal equations [6].
In this... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 1 |
b1d6af23-3ea2-4a75-96ba-c56b03d7ea9a | fects, such that 𝐻(𝑞,𝑝,𝑡)=𝐻 𝑒(𝑞,𝑝,𝑡)+𝐻 𝑐(𝑞,𝑡). These
components are defined as:
𝐻𝑒(𝑞,𝑝,𝑡)=1
2(𝑞2+𝑝2),
𝐻𝑐(𝑞,𝑝,𝑡)=𝑒𝑓rev
𝜎E𝑓s,0∫∞
𝑞∫∞
−∞˜𝜌(𝑓)𝑍(𝑓)𝑒𝑖2𝜋𝑓𝑞′𝑑𝑓𝑑𝑞′.
(3)
TheunperturbedHamiltonian, 𝐻𝑒,describesthestandard
harmonicsynchrotronmotionanddependsonthedimension-
less, general... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 1 |
17567710-1d17-4fda-8aa6-97c7f8bd8cd0 | frequency spectrum, ˜𝜌(𝑓), and the longitudinal impedance,
𝑍(𝑓),whichistheFouriertransformofthewakefieldand
describes the frequency-domain response of the accelerator
environment to the beam, including effects such as coherent
synchrotron radiation (CSR) self-interaction [5,9].Consequently, the solution of the VFPE... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 1 |
e562a484-6004-411d-8a45-ecc4d8afbc33 | of the measured data, the phase space reconstruction is con-
strainedbythismismatch,andtheinverseproblemmaylead
to biased or physically inconsistent phase spaces.
EXPERIMENTAL MEASUREMENTS
The experimental data for this study were acquired at the
Karlsruhe Research Accelerator, a storage ring with a cir-
cumference of ... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 1 |
083d4fb0-d9b5-40ee-aa37-b20b6d4c5ea6 | by a visible light monitor, used to determine the horizontal
bunch size in a dispersive section as a proxy for the energy
spread.
Thecoreofourreconstructionreliesontheelectro-optical
spectral decoding (EOSD) system. In this setup, an electro-
optical crystal is installed in-vacuum close to the beam
path [3,4,12]. As th... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 1 |
01370d54-78cb-4d8b-a619-ce7218413a6d | ablingturn-by-turnsingle-shotmeasurementsofthelongitu-
dinalbunchprofilewithsub-picosecondresolution[3,4,13].
The emitted CSR power is measured with fast THz detec-
torsbasedonroom-temperaturezero-biasSchottkybarrier
diodes,whicharesensitivetothecoherentradiationemitted
by the bunch in the THz range [16]. These detecto... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 1 |
eb9ab81c-5cb4-48ac-9f28-0f3a2e171159 | ISBN: 978-3-95450-252-3 ISSN: 2673-5350 doi: 10.18429/JACoW-IPAC2026-WEP5146
MC5.D11: Code Developments and Simulation Techniques 2585WEP: Poster session WEP5146
Content from this work may be used under the terms of the CC BY 4.0 licence (© 2026). Any distribution of this work must maintain attribution to the author(s)... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 1 |
6a8cef72-2963-4f97-a3fe-36bff219b5c0 | profileareusedhereasanindicatorforchangesintheenergy
spread.
Thesynchronizeddiagnosticsmeasurementsallowadi-
rect comparison of the reconstructed longitudinal phase
space with independent observables of the beam dynam-
ics [11,20]. For example, the synchronized data reveal
that the appearance of micro-bunching substruc... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 2 |
88f05917-31ce-4003-b066-5c35cb23f38b | inEq.(2). Eachrunoftheiterativealgorithm[2]isinitial-
izedwitha2DGaussianphasespacedistributionandutilizes
130consecutiveEOSDmeasurements,correspondingtohalf
ofasynchrotronperiod. Thecompletesequenceofexperi-
mental EOSD measurements used for the reconstructions is
displayedasasinograminFig.1(labeledas’longitudinal
pro... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 2 |
f5ea2205-3f23-45b5-8d19-90ef7fa6478a | size𝜎𝑥(white)andreconstructedenergyspread(cyan). Bot-
tom: THz power signal (black), and reconstructed CSR
power (cyan). All curves were smoothed with the same
moving-average filter.
Tovalidatethephysicalaccuracyofthereconstructions,
the macroscopic beam parameters extracted from the recon-
structedphasespacedensitie... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 2 |
77451b0b-5e87-4573-bfae-4c47ac711075 | function, asdetailed in [20];however,forthe dynamicsob-
servedhere,theycanbeassumedtobedirectlyproportional.
Furthermore,themacroscopicparameters 𝜎𝑧and𝜎𝐸from
ourreconstructionaredeterminedbyfittingaGaussianfunc-
tiontothe1Dprojectionsofthereconstructedphasespace
onto the position and energy deviation axes, respecti... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 2 |
5e8a72ca-8b82-47db-84ed-096f79dacb4b | (a)
(b)
Figure 2: Reconstructed phase space density at (a) turn
24000,showingasmoothdistributionwithoutvisiblemicro-
bunching,and(b)turn27000,showingpronouncedmicro-
bunching structures associated with stronger CSR emission;
thecolorgradientindicateslocaldensity,withlightercolors
corresponding to lower values.
CONCLUS... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 2 |
289488e9-57e6-498e-95c2-abc31c1945dc | experimentalvalidationofboththeinversereconstruction
method and the forward beam-dynamics solver. These re-
sultsestablishabasisforfuturephysics-informedprediction,
inverse modeling, and beam control in storage rings.17th International Particle Accelerator Conference, Deauville, France JACoW Publishing
ISBN: 978-3-9545... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 2 |
175657c1-28f3-45d0-b235-bc74d234796c | ACKNOWLEDGEMENTS
F.D. gratefully acknowledges support from the MathSEE
PhD Bridge Program and KCDS.
REFERENCES
[1]F. Donosoet al., “Longitudinal phase space density tomogra-
phyconstrainedbytheVlasov-Fokker-Planckequation”,in
Proc. IPAC’24, Nashville, TN, USA, May 2024, pp. 2350–
2353.doi:10.18429/JACoW-IPAC2024-WEPG55... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 3 |
ba9d6ad4-bdf6-4682-b4cf-f36ea1308711 | vacuumbunchprofilemonitorandmeasurementsatKARA
for use in the FCC-ee”, inProc. IPAC’24, Nashville, TN,
USA, May 2024, pp. 2354–2357.
doi:10.18429/JACoW-IPAC2024-WEPG56
[5]T. Boltzet al., On the Perturbation of Synchrotron Motion in
the Micro-Bunching Instability, 2022,
doi:10.48550/ARXIV.2211.15086
[6]S. Gazzola, P. C.... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 3 |
578d3e14-0c0c-4402-808e-0df610ffc6a2 | coherent THz radiation”, 2018.
doi:10.5445/IR/1000084466
[9]M. Brosiet al., “Studies of the Micro-Bunching Instability
inthePresenceofaDampingWiggler”,J. Phys. Conf. Ser.,
vol. 1067, p. 062017, 2018.
doi:10.1088/1742-6596/1067/6/062017
[10]M. Brosiet al., “Synchronous Measurements of Electron
Bunches Under the Influenc... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 3 |
03f9725c-27fd-4da0-8b87-3dffd0573e0a | doi:10.18429/JACoW-IPAC2023-THPL121
[13]S. Casalbuoni, H. Schlarb, B. Schmidt, P. Schmüser, B. Stef-
fen,andA.Winter,“Numericalstudiesontheelectro-optic
detection of femtosecond electron bunches”,Phys. Rev. Spec.
Top. Accel. Beams, vol. 11, no. 7, p. 072802, 2008.
doi:10.1103/PhysRevSTAB.11.072802
[14]L.Rotaet al.,“KAL... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 3 |
0fab3957-3847-4fb5-90b9-4642ad8bcebf | scopic investigation of the microbunching instability”,Phys.
Rev. Accel. Beams, vol. 21, no. 11, p. 110705, 2018.
doi:10.1103/PhysRevAccelBeams.21.110705
[17]M. Caselleet al., “KAPTURE-2. A picosecond sampling
systemforindividualTHzpulseswithhighrepetitionrate”,
J. Instrum., vol. 12, no. 01, p. C01040, 2017.
doi:10.108... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 3 |
866b90bf-ff0c-44ba-9be5-1ffee483c4d1 | transverse bunch signals at a storage ring”,Phys. Rev. Accel.
Beams, vol. 21, no. 10, p. 102803, 2018.
doi:10.1103/PhysRevAccelBeams.21.102803
[21]B.Kehreret al.,“Time-ResolvedEnergySpreadStudiesat
the ANKA Storage Ring”, inProc. IPAC’17, Copenhagen,
Denmark, May 2017, pp. 53–56.
doi:10.18429/JACoW-IPAC2017-MOOCB1
[22]... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 3 |
96370faf-799b-4e33-9c83-57f74d517903 | ISBN: 978-3-95450-252-3 ISSN: 2673-5350 doi: 10.18429/JACoW-IPAC2026-WEP5146
MC5.D11: Code Developments and Simulation Techniques 2587WEP: Poster session WEP5146
Content from this work may be used under the terms of the CC BY 4.0 licence (© 2026). Any distribution of this work must maintain attribution to the author(s)... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5146.pdf | 3 |
8d74842b-79b4-4306-94c7-aecd482f6fdd | DATA-DRIVEN RECONSTRUCTION OF ACCELERATOR LATTICE
ERRORS USING THREE BEAM POSITION MONITORS∗
Y. Hidaka†, G. Wang, Brookhaven National Laboratory, Upton, NY, United States
Abstract
There are numerous experimental techniques to quantify
errorsinaparticleacceleratorsothattheycanbecorrectedto
improve the performance of the... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 0 |
f0b0b65c-7787-4a9a-b891-1eded635263a | INTRODUCTION
Accurate identification and correction of lattice errors
areessentialforachievingoptimalperformanceinparticle
accelerators. Over the years, a wide range of beam-based
techniques have been developed to infer lattice imperfec-
tions using beam position monitor (BPM) data. These in-
clude methods based on orb... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 0 |
96eeb9a5-719f-469f-9376-a5af28bb3c45 | surements have been proposed. In particular, Maltseva et
al.[2]introducedamethodthatutilizesgeneratingfunctions
toreconstruct transversemomenta fromBPM triplets and
to compare adjacent lattice sections for error localization.
This approach demonstrated that inconsistencies between
reconstructed momenta can be used to i... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 0 |
32a914ce-b035-4946-b65e-a6d46b01227e | withinaboundedregion,RSDBOsuppressestheinfluence
of external disturbances and enables robust identification of
transversekicksoriginatinginsidethatregion. However,this
technique is fundamentally limited to perturbations that are
∗Worksupported byU.S.DOE under Contract No. DE-SC0012704.
†yhidaka@bnl.govindependentofthet... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 0 |
e6327d3b-b23b-492e-b180-d59ad84167dd | In this work, we extend this framework by introducing an
alternativediscrepancyevaluationmodelandamorecom-
putationally efficient phase space reconstruction method.
These modifications improve the scalability and practical
applicabilityofthemethod. Wecollectivelydenotethisgen-
eralizedframeworkasPIGLLET(ParameterIdenti... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 0 |
5985b9e2-8625-481e-b0ac-50ec8baa2913 | 𝑖)fromthemeasuredentranceandexit
positions (𝑥𝑖,𝑦𝑖,𝑥𝑓,𝑦𝑓). Once these angles are known, the
corresponding exit angles are obtained by a forward pass
throughthesamemodel. Iftheassumedelementparameters
are correct, these reconstructed phase-space coordinates are
self-consistent for all trajectories; otherwise, di... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 0 |
7e66fbf1-3b71-45a3-89c2-05b2645620db | 2,DSand̃ 𝑦′
2,US= ̃ 𝑦′
2,DS
for all trajectories. In P3 position matching (P3PM), used
below,IINAPisappliedonlytotheUSdata;theresulting
P1 phase-space coordinates are then tracked through both
sectionsandcomparedwiththeobservedP3positions,re-
quiring̃ 𝑥3,US= 𝑥3and̃ 𝑦3,US= 𝑦3.
P3PM avoids IINAP evaluation in the D... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 0 |
0ec0032f-7135-4586-8e3b-c85389376874 | the complexity of the fitted model. If the US section is
more complex, the same idea can be applied in reverse as
P1 position matching (P1PM).
Althoughasingletrajectorycansometimesbematched
with incorrect parameter values, simultaneous agreement
over many trajectories provides a stronger constraint. The
fittingproblemi... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 1 |
32e269ea-0a44-4ec7-9edc-91902d0c4723 | Step1: Linear seed First, a4×4linear transfer matrix
Misobtainedbytruncatingthefullnonlinearmap,andan
initial estimate is found by inverting Mwith the entrance
and exit BPM positions.
Step2: Newton–Raphsoncorrection Theestimateis
thenrefinedover 𝑁iterNewton–Raphsoniterationsbyprop-
agating the current estimate, evalua... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 1 |
038a9dd1-4193-4c98-9de8-fb2aaa088d7b | entiable PyTorch [ 6] implementation of the same integrator,
allowing either finite-difference or automatic-differentiation
Jacobians. S-IINAP(series) propagatesGTPSAobjects [ 7]
viathor_scsi [8,9]/gtpsa-cpp [10,11],sotheJacobian
isextractedfromtheseriescoefficients;itsaccuracydepends
on the chosen variable and paramet... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 1 |
eeb4279c-3a12-4c3b-a44b-bd4f5d5ffd58 | were tracked over ±3mm in𝑥1,±2mm in𝑦1,±300 𝜇rad
in𝑥′
1, and±200 𝜇rad in𝑦′
1. Incoming angles at P1 were
reconstructed from US position data and compared with
groundtruthusingthemedianabsoluteerror(MAE).The
0.0 0.5 1.0 1.5 2.0
s [m](a)USMIDDS
0 2 4 6 8
s [m]US-S1
US-Q1
US-S2
US-Q2
DS-S1
DS-S2(b)
Quadrupole
Sextupo... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 1 |
e91e4fb2-1503-4bea-9f72-65f520cb0a35 | difference Jacobians were adequate for these tests. For S-
IINAP,the accuracy floorwasset mainlybyvariable order,
whereasruntimeandmemoryscaledstronglywithtotalor-
der;increasingparameterorderdidnotimprovetheaccuracy
plateau at fixed variable order. In lattice (a), Table 1shows
that P- and T-IINAP reached roundoff-leve... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 1 |
150eb5d2-8b53-41d1-a1b0-531295b438b0 | mumresidentmemoryandtheshortestruntime. S-IINAP
memorywasnearlyindependentofiterationcountatfixed
order, but exceeded 1 GB and increased stronglywith both
variable and parameter orders. T-IINAP had the largest
memoryfootprint: itgrewfromabout5GBat 𝑁iter= 1to
about58GBat 𝑁iter= 13becausethedifferentiablegraph
accumula... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 1 |
1372a6f2-3021-40f8-a990-44067c0c07df | 3 1.08e-19 1.08e-19 1.00e-13 2.27e-17 5.15e-19
5 7.45e-20 6.78e-20 1.00e-13 2.27e-17 5.15e-1917th International Particle Accelerator Conference, Deauville, France JACoW Publishing
ISBN: 978-3-95450-252-3 ISSN: 2673-5350 doi: 10.18429/JACoW-IPAC2026-WEP5303
MC5.D01: Beam Optics Lattices, Correction Schemes, Transport 25... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 1 |
dd7b4bf8-2f31-4d78-84fd-2ccc4b189d64 | 2 4 6 8 10 12
Number of Iterations1019
1017
1015
1013
1011
109
107
Absolute Angle Error (rad)
Plain
T ensor
Series (3/2)
Series (4/2)
Series (5/2)
Series (6/2)
Series (7/2)
Series (8/2)
Series (9/2)Figure 2: Convergence comparison of different IINAP vari-
ants for lattice (b).
the tensor and series variants may still b... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 2 |
5a055fd4-2739-4c82-b66a-8f219caf73a8 | definesthelatticedesignaxis;hence,thesecenterscannotbe
includedinthefit. Thetotalnumberofparameterswas34.
A total of 100 Sobol-sampled trajectories were created in a
mannersimilartothatusedintheprevioussection. NoBPM
noise was added. Fitting was performed via least squares
with1000Sobol-sampledinitialguessesfortheparam... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 2 |
85579739-221a-40cd-9a9c-9f6b74bc8588 | in the fitting problem.
To reduce these degeneracies, another 100 trajectories
from a second state were added by reducing the nominal
DS-S1/H/0
DS-S1/tilt/0
DS-S1/xc/0
DS-S1/yc/0
DS-S2/H/0
DS-S2/tilt/0
DS-S2/xc/0
DS-S2/yc/0
P-2/C/0
P-2/gx/0
P-2/gy/0
P-2/tilt/0
P-2/xc/0
P-2/yc/0
P-3/C/0
P-3/gx/0
P-3/gy/0
P-3/tilt/0
US-Q... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 2 |
ea629409-f0d3-410c-8528-66b0b2ba8cf7 | P-2/yc/0,1
P-3/C/0,1
P-3/gx/0,1
P-3/gy/0,1
P-3/tilt/0,1
US-Q1/K/0
US-Q1/K/1
US-Q1/tilt/0,1
US-Q1/xc/0,1
US-Q1/yc/0,1
US-Q2/K/0
US-Q2/K/1
US-Q2/tilt/0,1
US-Q2/xc/0,1
US-Q2/yc/0,1
US-S1/H/0,1
US-S1/tilt/0,1
US-S1/xc/0,1
US-S1/yc/0,1
US-S2/H/0,1
US-S2/tilt/0,1
US-S2/xc/0,1
US-S2/yc/0,10.8
0.6
0.4
0.2
0.00.20.40.6Normalize... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 2 |
746b8af9-1b84-4ef9-bd22-dcf8d2f158a3 | Note that the labels 0 and 1 in the parameter names in the
figure denote the first and second states, respectively.
NoBPMnoisewasincludedintheresultsshownsofar.
The fitting is very sensitive to BPM noise, and the results
degrade substantially when BPM noise is included. Quanti-
fyingthissensitivity,andaddingregularizat... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 2 |
c1b77a52-a222-4120-96f0-ebff0a8501b1 | withP3PM,preliminarynoise-freesimulationsforanNSLS-
II lattice section showed that simultaneous fitting of magnet
strength,alignment, andBPM calibration errors is possible
with34–38fitted parameters. Thesingle-statefitexhibited
degeneracies for some parameters, especially transverse off-
sets, but adding a second state... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 2 |
fb0f2d90-158c-4ca7-8542-d14d9461c621 | NG GTPSA library.17th International Particle Accelerator Conference, Deauville, France JACoW Publishing
ISBN: 978-3-95450-252-3 ISSN: 2673-5350 doi: 10.18429/JACoW-IPAC2026-WEP5303
2590 MC5.D01: Beam Optics Lattices, Correction Schemes, TransportWEP5303 WEP: Poster session
Content from this work may be used under the t... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 2 |
dece55bb-53fc-46d3-88da-80211a326d67 | REFERENCES
[1]R. Tomás, M. Aiba, A. Franchi, and U. Iriso, “Review of
linearopticsmeasurementandcorrectionforchargedparticle
accelerators,”, Phys.Rev.Accel.Beams ,vol.20,p.054801,
2017. doi:1/zero.alt4.11/zero.alt43/PhysRevAccelBeams.2/zero.alt4./zero.alt4548/zero.alt41
[2]Yu.MaltsevaandI.A.Morozov,“Model-DependentAcce... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 3 |
1cebb2fc-9345-4e87-a300-d12e6cee7095 | doi:1/zero.alt4.18429/JACoW-IPAC2/zero.alt419-WEPGW1/zero.alt41
[4]Y. Hidaka et al., “Investigation on Mysterious Long-Term
Orbit Drift at NSLS-II”, in Proc. 10th Int. Particle Accel-
erator Conf. (IPAC’19) , Melbourne, Australia, May 2019,
pp. 2728–2731.
doi:1/zero.alt4.18429/JACoW-IPAC2/zero.alt419-WEPGW1/zero.alt42
... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 3 |
b031d6f8-eda8-4242-807f-7fb85aeb0c00 | Garnett,Eds.CurranAssociates,Inc.,2019,pp.8024–8035.
doi:1/zero.alt4.4855/zero.alt4/arXiv.1912./zero.alt417/zero.alt43
[7]L. Deniau and C. I. Tomoiaga, “Generalised Truncated
Power Series Algebra for Fast Particle Accelerator Trans-
port Maps”, in Proc. IPAC’15 , Richmond, VA, USA, May
2015, pp. 374–377.
doi:1/zero.alt... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 3 |
80efc7b7-e71a-43a6-8252-de89b3d5031f | Truncated Power Series Algebra, https://github.com/
hz-b/gtpsa-cpp ,
[11]P. Schnizer, J. Bengtsson, and W. Sulaiman Khail, “Small
TalkonAT”,in Proc.IPAC’22 ,Bangkok,Thailand,Jun.2022,
pp. 918–921.
doi:1/zero.alt4.18429/JACoW-IPAC2/zero.alt422-TUPOST/zero.alt429
[12]M. Berz, Modern Map Methods in Particle Beam Physics .... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5303.pdf | 3 |
f70cc5e3-3ddd-42b9-af3c-55fe43d60ded | ADJOINT SENSITIVITY ANALYSIS FOR DESIGN AND OPTIMIZATION∗
D.T. Abell†, V.B. Tembo, I. Haber, T.M. Antonsen, University of Maryland, College Park, USA
Abstract
The design of accelerator lattices involves evaluating
and optimizing Figures of Merit (FoMs) that characterize
abeam’sproperties.Theseproperties—hencetheFoMs—
d... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 0 |
474a4d5f-c946-4009-9784-1c49a91d5e1d | niquesallowonetoextractgradientinformationfromone
base-case simulation plus an additional one or two carefully
preparedsimulations [1].We demonstratethese techniques
usingtheacceleratorsimulationcodeWARP,andwepresent
our proof-of-concept results using several different FoMs
asthebasisforadjointanalysesofasimplebeamline... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 0 |
e61e2d4b-48c4-48ab-8c93-c5899b8384be | descent. (ii) It yields the sensitivity of a particular design
to errors in the lattice, and hence enables designers to set
construction tolerances.
Adjointmethodsprovideadramaticallymoreefficientway
to calculate parameter dependencies than straightforward
parameterdifferencing,whichrequiresaseparatesimulation
run for ... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 0 |
f1a44ea9-56f1-4cce-b266-f35c388d3ed7 | accelerators, we have recently deployed adjoint methods to
optimizeelectrostatic[8]andmagnetostatic[9]focusingsys-
tems, as well as to optimize the combination of quadrupole
and solenoid magnetic fields needed to transform—in the
∗ThisworksupportedinpartbytheUSDepartmentofEnergy,Officeof
Science, Office of High Energy ... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 0 |
52a400b8-d9b1-40cd-851e-52a4f43f50bc |
−𝛿𝜌(𝐴)𝛿𝜙(𝑇)+𝛿𝐽(𝐴)·𝛿𝐴(𝑇)
−
𝐴↔𝑇o
,(1)
where[𝐴↔𝑇]denotesthepreviousexpressioninsquare
brackets with the superscripts (𝑇)and(𝐴)interchanged.
On the left-hand side (LHS), we have a net change in the
symplecticareaseparatingthetrajectoriesoftwodifferent
perturbed solutions. On the right-hand side (RHS)... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 0 |
3fb5ce92-0f20-4397-b1bc-e1aabc4c2282 | instrumented simulations—constitutes the easy-to-compute
quantity that yields the change we seek todetermine in the
FoM of interest. Moreover, we canusethe linearity of inte-
grationtoaccumulateindividualcontributionstotheRHS,
andcomputethustheFoM gradientwithrespecttoasmany
parameters as we deem relevant.
We focus on ... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 0 |
e4824e48-07fe-4a65-b4d6-df41162f6573 | 𝑗=−𝜕𝑓
𝜕𝑥𝑗Δ,(3a)
where we define the perturbation scale parameter
Δ=Δ(𝜀)=𝑥typical𝑝typical
𝐹typical𝜀 (3b)17th International Particle Accelerator Conference, Deauville, France JACoW Publishing
ISBN: 978-3-95450-252-3 ISSN: 2673-5350 doi: 10.18429/JACoW-IPAC2026-WEP5304
2592 MC5.D11: Code Developments and Simula... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 0 |
ac9107f5-2b2a-41ee-a3ab-58c12c382411 | for small𝜀. Then, by the chain rule, the LHS becomes
∑︁
𝑗𝐼𝑗
𝐼h
𝛿𝑝(𝑇)
𝑗⊥·𝜕𝑓
𝜕𝑝𝑗⊥Δ
+𝜕𝑓
𝜕𝑥𝑗⊥Δ
·𝛿𝑥𝑗⊥(𝑇)i
𝐿=𝛿𝐹·Δ.
(4)
Toenforcetheperturbationvaluesattheexitplane,wemust
run the simulation forward, perturb the phase-space coördi-
nates as in (3a), and then run the simulation in reverse.
To s... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 1 |
767590c9-05e3-40ec-a010-64e035a46fff | 𝑙=1𝐾𝑙∑︁
𝑘=1𝜕𝐴𝑧
𝜕𝑔𝑙𝑘𝛿𝑔𝑙𝑘,(5)
where𝑔𝑙𝑘denotes a given parameter, with 𝑙labeling the𝑁𝑒
different elements, and 𝑘labeling each of the 𝐾𝑙variable
parameterswithinelement 𝑙.Weemphasizethefactthatitis
thedecompositionof 𝛿𝐴into(𝜕𝐴/𝜕𝑔𝑙𝑘)𝛿𝑔𝑙𝑘thatfacilitates
the simultaneous computation of mul... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 1 |
1b50c2ef-918b-421c-960d-3fd942dfd408 | scalevalueofparameter 𝑔𝑙𝑘;(iii)𝜎𝑙𝑘denotesa dimension-
lesssensitivity of𝐹withrespecttochangesinparameter 𝑔𝑙𝑘;
and (iv) the factor of 2 in the denominator of 𝜎𝑙𝑘appears
because we are taking a symmetric difference.
Ourgoalistocomputethesensitivities 𝜎𝑙𝑘;and(6b)tells
us that to do so, we need only (for th... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 1 |
2b8f7fbb-23d8-4fa8-9110-3a1117b17c11 | 𝜎𝑧[𝑄1](lower) as functions of the 𝜀defined in (3b). The
blue curves are quadratic fits to that data.
Thequadrupolesallhaveaneffectivelengthof 4.65cm,with
2cmfringeregionsateachendimplementedassmoothly-
varying profiles that increase as cos2𝜋𝑧/(2𝑙𝑓)at the en-
trance, and decrease in the same manner at the exit.... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 1 |
867d5823-b034-4d90-a7b5-6d0e1e630bd8 | sensitivityof 𝐹tothethreequadrupolegradients 𝑔𝑛andthe
three quadrupole positions 𝑧𝑛.
For values of 𝜀in the range 0.001 to 0.2 , we observe
a weak quadratic dependence of the computed sensitivity
on𝜀—see, for example, the graphics in figure 1—which
impliesourcomputationsliewellwithinthelinearregime
(i.e.theintegr... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 1 |
32f4978a-e707-4d7c-9dd4-37e3bc040ad8 | ISBN: 978-3-95450-252-3 ISSN: 2673-5350 doi: 10.18429/JACoW-IPAC2026-WEP5304
MC5.D11: Code Developments and Simulation Techniques 2593WEP: Poster session WEP5304
Content from this work may be used under the terms of the CC BY 4.0 licence (© 2026). Any distribution of this work must maintain attribution to the author(s)... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 1 |
9f26bda6-a7a2-47c4-a9b7-46d28cdd3f92 | Table 1: This table gives our adjoint and direct sensitivity
values for the six different sensitivities studied.
adjoint direct
𝜎[𝑔1] − 1.763605858906474 −1.763602604989369
𝜎[𝑔2] 0.897858308034822 0 .897856025105494
𝜎[𝑔3] − 0.366284605760255 −0.366284465747279
𝜎[𝑧1] 0.053614228190004 0 .053613157372817
𝜎[𝑧2] ... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 2 |
3d143c43-8b57-49ad-8cbb-aa1e0a2ac15b | pretthe outputvalues—thesensitivities 𝜎𝑙𝑘—interms ofa
given parameter’s impact on some relevant FoM. Moreover,
whereas prior work developed adjoint methods on the basis
ofmomentequations,herewestartedwithparticleequations
ofmotion,whichallowsustocapturemuchmoregranular
detail about the beam.
We emphasize that one ca... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 2 |
d7f7150e-9396-4d0a-80e5-b44b398c71ca | Andreas Adelmann of the Paul Scherrer Institute for early
discussions on this subject.
APPENDIX
Herewesketchaproofoftheadjointrelation(1)forthe
case of time-based codes.
Tobegin, weconsider perturbations tobothexternally ap-
pliedelectromagneticpotentialsandtheparticletrajectories.
Forperturbedpotentials,wewrite 𝜙+𝛿�... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 2 |
03066ba5-d3e4-4c61-86b9-5a1506d7a5c1 | 𝜕𝑥𝜕𝑝·𝛿𝑝−𝑞𝜕
𝜕𝑥(𝛿𝜙−𝑣·𝛿𝐴).
Subtract from this result the equations of motion for the
unperturbedtrajectoriestoobtainthefollowingequationsof
motion for the trajectory perturbations:
d𝛿𝑥𝑗
d𝑡=𝜕2𝐻
𝜕𝑝𝜕𝑥·𝛿𝑥𝑗+𝜕2𝐻
𝜕𝑝2·𝛿𝑝𝑗−𝑞𝑗𝜕
𝜕𝑝(𝑣𝑗·𝛿𝐴),(9a)
d𝛿𝑝𝑗
d𝑡=−𝜕2𝐻
𝜕𝑥2·𝛿𝑥𝑗−𝜕2𝐻
𝜕𝑥𝜕𝑝... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 2 |
ad203020-6cbe-454b-94cd-23de508c9cad | ∑︁
𝑗𝐼𝑗
𝐼
𝛿𝑥(𝐴)
𝑗·𝛿𝑝(𝑇)
𝑗−𝛿𝑥(𝑇)
𝑗·𝛿𝑝(𝐴)
𝑗 𝜏
0(11)
Replace the evaluation at 0and𝜏by∫𝜏
0d
d𝑡, carry out the
timederivative,andinsertequations(9).Keepinginmind
that the potentials do not depend on momenta, one finds,
aftersometediousbutroutineeffort,thatallsecond-order
derivatives of 𝐻cancel, ... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 2 |
051ac0ea-c80c-4515-872e-e86c03b7b220 | integralover 𝛿𝜌(𝐴)𝛿𝜙(𝑇).Similarly,onecanrewritetheterm
thatcontains 𝛿𝐴(𝑇)intermsofanintegralover 𝛿𝐽(𝐴)·𝛿𝐴(𝑇).
Theadjointrelation(1)resultsfromasimilarcomputation
using the Hamiltonian with 𝑧as the independent coördinate.17th International Particle Accelerator Conference, Deauville, France JACoW Publishi... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 2 |
cb89d9b6-36eb-48b4-ba3f-4113d6b52bbe | REFERENCES
[1]L. Dovlatyan, B.L. Beaudoin, S. Bernal, I. Haber, D.F. Sutter,
and T.M. Antonsen, “Optimization of flat to round
transformers with self-fields using adjoint techniques”, Phys.
Rev. Accel. Beams 25:044002 (2022).
doi:10.1103/physrevaccelbeams.25.044002
[2] S.W. Director and R.A. Rohrer, “The generalized ad... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 3 |
d5e9d81f-80e3-4624-932b-c5451ff4cc3a | generation by waves in toroidal geometry”, Phys. Fluids
25:1295–1296 (1982). doi:10.1063/1.863906[6] T.M. Antonsen, E.J. Paul, and M. Landreman, “Adjoint
approach to calculating shape gradients for three-dimensional
magnetic confinement equilibria”, J. Plasma Phys.
85:905850207 (2019).
doi:10.1017/S0022377819000254
[7]... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 3 |
a1d9d8f9-dc29-4bcc-89f1-dab79c7d9dc1 | doi:10.1063/1.5079629
[9] T.M. Antonsen, B.L. Beaudoin, L. Dovlatyan, and I. Haber,
“Adjoint approach to accelerator lattice design”, in Proc.
NAPAC’19 , Lansing, MI, USA, Sep. 2019, pp. 376–378.
doi:10.18429/JACoW-NAPAC2019-TUPLM0317th International Particle Accelerator Conference, Deauville, France JACoW Publishing
I... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5304.pdf | 3 |
5f09a10d-e5c6-498e-ae1f-050631f4c236 | STUDY OF BEAM COUPLING IMPEDANCE AND INSTABILITIES FOR
THE SUPER TAU-CHARM FACILITY∗
T. He†, W. Li, W. Liang, Z. Meng, Y. Zou, J. Tang, USTC, Hefei, China
D. Zhou, KEK, 1-1 Oho, Tsukuba 305-0801, Ibaraki, Japan
Abstract
The Super Tau-Charm Facility (STCF), a next-generation
electron-positroncolliderinChina,isdesignedto... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 0 |
06197e37-2d4c-467a-9395-891cd68035a5 | generation electron-positroncollider in China,operating at
center-of-mass energies from 2 to 7 GeV, with a target peak
luminosity exceeding 5×1034cm−2s−1at 4 GeV center-of-
mass energy [1]. To achieve this high luminosity, the STCF
requires high beam currents and low emittance, making
collective effects a critical chal... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 0 |
a1ab346d-0788-4318-9714-c2778df14955 | from dipoles, CWR from wigglers, RW with NEG coating,
andgeometriccontributionsfromcollimators,BPMs,bel-
lows, etc. Since detailed geometric models for the STCF
arecurrentlyunavailable,thegeometricimpedanceisesti-
matedbyscalingexistingimpedancebudgetsfromHALF,
FCC-ee[2],andSuperKEKB[3],basedonthebeampipe
aperture of S... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 0 |
1b698bf2-0b57-4e71-912e-6684cb26e2e6 | ForRW,theimpedanceisevaluatedusingtheIW2Dcode,
∗Work supported by the Fundamental Research Funds for the Central
Universities (No. WK2310000127) and the National Natural Science
Foundation of China (Grants No. 12375324 and No. 12105284).
†htlong@ustc.edu.cnconsidering NEG coating with conductivities of 105and
106S/m [6... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 0 |
05fae275-51fa-4a2e-8288-62b456c04560 | Synchrotron tune (𝜈 𝑠) 0.0147 0.0193 0.0228
Betatron tune (𝜈 𝑥/𝜈𝑦) 28.55/33.5828.55/31.5828.55/30.58
IMPEDANCE MODELING
Longitudinal Impedance
Based on the modeling methods described above, the lon-
gitudinal impedance contributions from CSR, CWR, RW,
and geometric sources are evaluated. Figure 2 shows the
longit... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 0 |
b5bad990-170e-4b41-b24a-ea2576b4450f | tials for a 0.5mmbunch. The corresponding kick factors
are listed in Table 2. Transverse impedance is dominated by
geometric components, especially in the vertical plane. The
geometrickickfactorsaresignificantlylargerthantheRW
contribution at both 1 GeV and 2 GeV.17th International Particle Accelerator Conference, Deau... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 0 |
19842eb5-ce6d-4a75-9f9a-5d6bf4112805 | Figure 2: Longitudinal short-range wake potential for
0.5mm bunch length.
Figure3: Transverseshort-rangewakepotentialsfor 0.5 mm
bunch length.
SINGLE-BUNCH INSTABILITIES
Longitudinal Microwave Instability
The microwave instability (MWI) is investigated using
macroparticletrackingwiththeSTABLEcode[7]and5mil-
lionmacropa... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 1 |
a72e64a9-db8d-4888-a2a4-84e96d4cf162 | threshold current.
Figure 4: Energy spread vs. single-bunch current from
microwave instability simulations.Table 2: Transverse Kick Factors at 1 GeV and 2 GeV
Component 1 GeV [V/pC] 2 GeV [V/pC]
RW (NEG:106S/m) 984 1974
Geometric 6600 6600
Transverse Mode Coupling Instability
The transverse mode coupling instability (T... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 1 |
86507af8-475c-44ce-ae30-7106d8df6188 | Figure5: TMCIthresholdcurrentsfromtrackingandana-
lytical methods.
COUPLED-BUNCH INSTABILITIES
Accelerating Mode
To suppress coupled-bunch instabilities driven by the ac-
celerating mode impedance, the STCF plans to use a room-
temperature 500 MHz cavityoperatingintheTM 020mode,
whichfeaturesstronghigher-ordermodedampi... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 1 |
c6a8775c-8d93-4f57-9eac-cccc94080aec | of half the revolution period effectively suppresses the most
dangerous mode 𝜇=−1. This may slightly increase the
growth rate of mode 𝜇=−2, but it remains below the radia-
tion damping rate [9]. Therefore, the PI feedback alone is
sufficient to stabilize the accelerating-mode-driven instabili-
ties.
Higher-Order Mode... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 1 |
70c88101-9ee3-4213-9544-831688e58f99 | ISBN: 978-3-95450-252-3 ISSN: 2673-5350 doi: 10.18429/JACoW-IPAC2026-WEP5601
MC5.D05: Coherent and Incoherent Instabilities Theory, Simulations, Code Development 2597WEP: Poster session WEP5601
Content from this work may be used under the terms of the CC BY 4.0 licence (© 2026). Any distribution of this work must maint... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 1 |
a54c4233-bd09-4408-9bcc-607557fc4635 | Table3: GrowthRates( 𝑠−1)ofLowCoupled-BunchModes
DrivenbytheAcceleratingModewithoutandwiththeRF
Feedback
𝜇2 GeV 3.5 GeV
Without FB With FB Without FB With FB
-1 412 0 562 0
-2 44 92 62 120
1/𝜏𝑧 112 155
PSOfitting algorithm. Theresults areshownin Fig.6. As-
sumingallcavitieshaveidenticalHOMfrequenciesandthat
these f... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 2 |
b0ba0b44-c7fb-4f9f-840b-c5d948b6b62a | revolutionfrequency, 𝛽𝑥/𝑦istheaveragebetafunctionatthe
cavitylocation,and Re𝑍⊥istherealpartofthetransverse
HOM impedance.
Ouranalysisindicatesthatifthelongitudinalbunch-by-
bunch feedback provides at least 2 msof damping and the
transverse feedbackprovides at least 1 ms, all HOM-driven
instabilities can be effectiv... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 2 |
49bdf5a6-3ff0-49ee-ab79-d4a1a2e323ac | modegrowthratereaches 1643s−1. Increasingchromaticity
to +3 reduces it only slightly to 1583s−1, which is still far
above the radiation damping rate of 56.5s−1. Therefore, a
fast transverse feedback system with a damping time below
0.6ms is required to suppress this instability.
Figure 7: RW-driven instability growth r... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 2 |
bd90c320-8921-41c0-8f40-fbda373f8459 | •HOM-driven instabilities require active feedback, with
dampingtimesofabout2mslongitudinallyand1ms
transversely.
•Transverse RW instability demand fast feedback sys-
tems (0.4–0.6 ms).
Theseresultsprovideabasisforfurtherimpedancereduction
and feedback design for the STCF.
REFERENCES
[1]X.-C. Aiet al., “Conceptual desig... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 2 |
036d1ca9-378f-4e08-91d4-700c17c04d7f | lished.
[4]R.L.Warnock,“Shieldedcoherentsynchrotronradiationand
itseffectonveryshortbunches”,StanfordLinearAccelerator
Center, Stanford, CA, USA, Rep. SLAC-PUB-5375, 1990.
[5]T.Heet al.,“Impactofcoherentwigglerradiationimpedance
in Tau-Charm factories”,Phys. Rev. Accel. Beams, vol. 28, p.
101002, 2025.
doi:10.1103/yqdj... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 2 |
77c570a6-c34d-484b-8438-d087db38d483 | [7]T.HeandZ.Bai,“Graphics-processing-unit-acceleratedsim-
ulation for longitudinal beam dynamics of arbitrary bunch
trainsinelectronstoragerings”,Phys. Rev. Accel. Beams,vol.
24, p. 104401, 2021.
doi:10.1103/PhysRevAccelBeams.24.104401
[8]Y. Weiet al., “Preliminary design of a 500 MHz normal-
conducting cavity for main... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 3 |
d957140d-a460-4fb2-8468-547ed58a96da | ISBN: 978-3-95450-252-3 ISSN: 2673-5350 doi: 10.18429/JACoW-IPAC2026-WEP5601
MC5.D05: Coherent and Incoherent Instabilities Theory, Simulations, Code Development 2599WEP: Poster session WEP5601
Content from this work may be used under the terms of the CC BY 4.0 licence (© 2026). Any distribution of this work must maint... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5601.pdf | 3 |
2f052eaf-534c-472a-b69f-cce150adfec7 | A REALISTIC PROPORTIONAL-INTEGRAL RF FEEDBACK MODEL FOR
LONGITUDINAL BEAM DYNAMICS SIMULATION
T. He∗, W. Liang, J. Xiao, X. Huang, University of Science and Technology of China, Hefei, China
Abstract
Modernfourth-generationstorageringlightsourcespre-
dominantly utilize digital I/Q-based proportional-integral
(PI)feedba... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5602.pdf | 0 |
f7eb21c2-0999-4c20-911b-dbed7adbce2d | viding a versatile tool for precise longitudinal beam dynam-
ics studies in single- and multi-RF configurations.
INTRODUCTION
Achievingpreciseregulationofradio-frequency(RF)cav-
itiesisfundamentaltotheperformanceofmodernstorage
ringlightsources,ataskpredominantlyaccomplishedbydig-
italI/Q-basedproportional-integral(PI)... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5602.pdf | 0 |
dfe06028-cb83-4c7d-b215-5fdea8e2e2f9 | simulation [2]. Other codes, such as MBTRACK2, directly
implement a PI controller framework to model the cavity
response[3,4]. Whilethesetoolscanmodelthefundamental
cavity dynamics accurately, systematic scans over PI gains,
loopdelays,andoperatingconditionsinmoderndigitalfeed-
back systems still demand substantial com... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5602.pdf | 0 |
60457b7d-25b5-4261-b0b7-b2d841affb80 | resentationoftransientbeam-cavity-feedbackcouplingin-
teractions. Second, and distinctively, STABLE leverages
NVIDIA CUDA for massive parallelization on GPUs. Com-
paredtoCPU-parallelizedframeworks,thisarchitecturepro-
vides a substantial increase in computational throughput
for large-scale parameter scans at a lower h... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5602.pdf | 0 |
b52560ea-e4ba-4f91-93b6-c5a91418007c | period. As illustrated in Fig. 1, we demonstrate that the
charge of each pulse equals1
2𝐼𝑔𝑇, where𝐼𝑔and𝑇are the
amplitudeandperiodofthegeneratorcurrent, respectively.
Figure1: Discretizationofacosinegeneratorcurrentinto
equivalent pulsed charges.
Analogous to representing a longitudinally distributed
bunch charge... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5602.pdf | 0 |
c2c99204-aa9a-4354-84cd-fc38537cef74 | and ˜𝑄𝑔, respectively. For a pulsed charge excitation at time
zero, the resulting cavity voltage phasor is:
˜𝑉𝑔=˜𝑄𝑔𝜔𝑟𝑅𝐿
𝑄𝐿. (2)
Here,𝜔𝑟,𝑅𝐿,and𝑄𝐿aretheangularresonantfrequency,
loaded shunt impedance, and loaded quality factor of the
cavity, respectively. Substituting Eq. (1)into Eq.(2)and17th Internat... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5602.pdf | 0 |
6e686060-b76a-4175-8685-b2bc5d9e41ca | Figure 2: Simplified block diagram of the RF system at SSRF [6].
using𝜔𝑟𝑇≈2𝜋yields:
˜𝑉𝑔=˜𝐼𝑔𝜋𝑅𝐿
𝑄𝐿. (3)
Using Eq. (3), the cavity voltage phasor induced by the
generator current can be computed iteratively:
˜𝑉𝑛
𝑔=˜𝑉𝑛−1
𝑔𝑒
𝑗−1
2𝑄𝐿
𝜔𝑟𝑇+˜𝐼𝑛
𝑔𝜋𝑅𝐿
𝑄𝐿, (4)
where the superscript 𝑛denotes th... | D:\Daily_ihep\Conferences\IPAC2026\WEP\WEP5602.pdf | 1 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.