Keyword: network
Paper Title Other Keywords Page
MOPAB182 Automated Synchrotron Lattice Design and Optimisation Using a Multi-Objective Genetic Algorithm lattice, synchrotron, dipole, superconducting-magnet 616
 
  • X. Zhang, S.L. Sheehy
    The University of Melbourne, Melbourne, Victoria, Australia
  • E. Benedetto
    TERA, Novara, Italy
  • E. Benedetto
    CERN, Meyrin, Switzerland
 
  Funding: This work is partially supported by the Australian Government Research Training Program Scholarship.
As part of the Next Ion Med­ical Ma­chine Study (NIMMS), we pre­sent a new method for de­sign­ing syn­chro­tron lat­tices. A step-wise ap­proach was used to gen­er­ate ran­dom lat­tice struc­tures from a set of feed­for­ward neural net­works. These lat­tice de­signs are op­ti­mised by evolv­ing the net­works over many it­er­a­tions with a multi-ob­jec­tive ge­netic al­go­rithm (MOGA). The final set of so­lu­tions rep­re­sent the most effi- cient and fea­si­ble lat­tices which sat­isfy the de­sign con­straints. It is up to the lat­tice de­signer to choose a de­sign that best suits the in­tended ap­pli­ca­tion. The au­to­mated al­go­rithm pre­sented here ran­domly sam­ples from all pos­si­ble lat­tice lay­outs and reaches the global op­ti­mum over many it­er­a­tions. The re­quire­ments of an ef­fi­cient ex­trac­tion scheme in hadron ther­apy syn­chro­trons im­pose strin­gent con­straints on the lat- tice op­ti­cal func­tions. Using this al­go­rithm al­lows us to find the global op­ti­mum that is tai­lored to these con­straints and to fully utilise the flex­i­bil­i­ties pro­vided by new tech­nol­ogy.
 
poster icon Poster MOPAB182 [6.006 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-MOPAB182  
About • paper received ※ 15 May 2021       paper accepted ※ 23 June 2021       issue date ※ 14 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
MOPAB281 Research on Resolution Evaluation of Stripline BPM at SXFEL-UF FEL, electron, linac, experiment 892
 
  • B. Gao, J. Chen, Y.B. Leng
    SSRF, Shanghai, People’s Republic of China
 
  48 stripline BPMs are in­stalled in the in­jec­tion sec­tion and lin­ear ac­cel­er­a­tion sec­tion of Shang­hai X-ray Free Elec­tron Laser (SXFEL) for elec­tron beam po­si­tion mea­sure­ment. These two sec­tions re­quire res­o­lu­tion of 20 µm@​100pC.​ Res­o­lu­tion eval­u­a­tion is an im­por­tant step in BPM in­stal­la­tion and com­mis­sion­ing. This paper pre­sents BPM res­o­lu­tion eval­u­a­tion meth­ods based on cor­re­la­tion analy­sis. Ex­per­i­men­tal meth­ods, data pro­cess­ing and re­sult analy­sis will be dis­cussed  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-MOPAB281  
About • paper received ※ 19 May 2021       paper accepted ※ 27 May 2021       issue date ※ 02 September 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
MOPAB286 Towards a Data Science Enabled MeV Ultrafast Electron Diffraction System electron, experiment, real-time, laser 906
 
  • M.A. Fazio, S. Biedron, M. Martínez-Ramón, D.J. Monk, S.I. Sosa Guitron
    UNM-ECE, Albuquerque, USA
  • M. Babzien, K.A. Brown, M.G. Fedurin, J.J. Li, M.A. Palmer, J. Tao
    BNL, Upton, New York, USA
  • S. Biedron, T. Talbott
    UNM-ME, Albuquerque, New Mexico, USA
  • J. Chen, A.J. Hurd, N.A. Moody, R. Prasankumar, C. Sweeney
    LANL, Los Alamos, New Mexico, USA
  • D. Martin, M.E. Papka
    ANL, Lemont, Illinois, USA
 
  Funding: US DOE, SC, BES, MSE, award DE-SC0021365 and DOE NNSA award 89233218CNA000001 through DOE’s EPSCoR program in Office of BES with resources of DOE SC User Facilities BNL’s ATF and ALCF.
A MeV ul­tra­fast elec­tron dif­frac­tion (MUED) in­stru­ment is a unique char­ac­ter­i­za­tion tech­nique to study ul­tra­fast processes in ma­te­ri­als by a pump-probe tech­nique. This rel­a­tively young tech­nol­ogy can be ad­vanced fur­ther into a turn-key in­stru­ment by using data sci­ence and ar­ti­fi­cial in­tel­li­gence (AI) mech­a­nisms in con­junc­tions with high-per­for­mance com­put­ing. This can fa­cil­i­tate au­to­mated op­er­a­tion, data ac­qui­si­tion and real time or near- real time pro­cess­ing. AI based sys­tem con­trols can pro­vide real time feed­back on the elec­tron beam which is cur­rently not pos­si­ble due to the use of de­struc­tive di­ag­nos­tics. Deep learn­ing can be ap­plied to the MUED dif­frac­tion pat­terns to re­cover valu­able in­for­ma­tion on sub­tle lat­tice vari­a­tions that can lead to a greater un­der­stand­ing of a wide range of ma­te­r­ial sys­tems. A data sci­ence en­abled MUED fa­cil­ity will also fa­cil­i­tate the ap­pli­ca­tion of this tech­nique, ex­pand its user base, and pro­vide a fully au­to­mated state-of-the-art in­stru­ment. We will dis­cuss the progress made on the MUED in­stru­ment in the Ac­cel­er­a­tor Test Fa­cil­ity of Brookhaven Na­tional Lab­o­ra­tory.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-MOPAB286  
About • paper received ※ 20 May 2021       paper accepted ※ 09 June 2021       issue date ※ 25 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
MOPAB288 Real-Time Edge AI for Distributed Systems (READS): Progress on Beam Loss De-Blending for the Fermilab Main Injector and Recycler real-time, operation, distributed, FPGA 912
 
  • K.J. Hazelwood, M.R. Austin, M.A. Ibrahim, V.P. Nagaslaev, A. Narayanan, D.J. Nicklaus, A.L. Saewert, B.A. Schupbach, K. Seiya, R.M. Thurman-Keup, N.V. Tran
    Fermilab, Batavia, Illinois, USA
  • H. Liu, S. Memik, R. Shi, M. Thieme
    Northwestern University, Evanston, Illinois, USA
  • A. Narayanan
    Northern Illinois University, DeKalb, Illinois, USA
 
  The Fer­mi­lab Main In­jec­tor en­clo­sure houses two ac­cel­er­a­tors, the Main In­jec­tor and Re­cy­cler. Dur­ing nor­mal op­er­a­tion, high in­ten­sity pro­ton beams exist si­mul­ta­ne­ously in both. The two ac­cel­er­a­tors share the same beam loss mon­i­tors (BLM) and mon­i­tor­ing sys­tem. Beam losses in the Main In­jec­tor en­clo­sure are mon­i­tored for tun­ing the ac­cel­er­a­tors and ma­chine pro­tec­tion. Losses are cur­rently at­trib­uted to a spe­cific ma­chine based on tim­ing. How­ever, this method alone is in­suf­fi­cient and often in­ac­cu­rate, re­sult­ing in more dif­fi­cult ma­chine tun­ing and un­nec­es­sary ma­chine down­time. Ma­chine ex­perts can often dis­tin­guish the cor­rect source of beam loss. This sug­gests a ma­chine learn­ing (ML) model may be pro­ducible to help de-blend losses be­tween ma­chines. Work is un­der­way as part of the Fer­mi­lab Real-time Edge AI for Dis­trib­uted Sys­tems Pro­ject (READS) to de­velop a ML em­pow­ered sys­tem that col­lects streamed BLM data and ad­di­tional ma­chine read­ings to infer in real-time, which ma­chine gen­er­ated beam loss.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-MOPAB288  
About • paper received ※ 19 May 2021       paper accepted ※ 29 July 2021       issue date ※ 13 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
MOPAB314 Surrogate Modeling for MUED with Neural Networks electron, experiment, gun, operation 970
 
  • D.J. Monk, S. Biedron, M.A. Fazio, M. Martínez-Ramón, S.I. Sosa Guitron
    UNM-ECE, Albuquerque, USA
  • M. Babzien, K.A. Brown, M.A. Palmer, J. Tao
    BNL, Upton, New York, USA
  • D. Martin, M.E. Papka
    ANL, Lemont, Illinois, USA
  • T. Talbott
    UNM-ME, Albuquerque, New Mexico, USA
 
  Elec­tron dif­frac­tion is among the most com­plex and in­flu­en­tial in­ven­tions of the last cen­tury and con­tributes to re­search in many areas of physics and en­gi­neer­ing. Not only does it aid in prob­lems like ma­te­ri­als and plasma re­search, elec­tron dif­frac­tion sys­tems like the MeV ul­tra-fast elec­tron dif­frac­tion(MUED) in­stru­ment at the Brookhaven Na­tional Lab(BNL) also pre­sent op­por­tu­ni­ties to ex­plore/im­ple­ment sur­ro­gate mod­el­ing meth­ods using ar­ti­fi­cial in­tel­li­gence/ma­chine learn­ing/deep learn­ing al­go­rithms. Run­ning the MUED sys­tem re­quires ex­tended pe­ri­ods of un­in­ter­rupted run­time, skilled op­er­a­tors, and many vary­ing pa­ra­me­ters that de­pend on the de­sired out­put. These prob­lems lend them­selves to tech­niques based on neural net­works(NNs), which are suited to mod­el­ing, sys­tem con­trols, and analy­sis of time-vary­ing/multi-pa­ra­me­ter sys­tems. NNs can be de­ployed in model-based con­trol areas and can be used sim­u­late con­trol de­signs, planned ex­per­i­ments, and to sim­u­late em­ploy­ment of new com­po­nents. Sur­ro­gate mod­els based on NNs pro­vide fast and ac­cu­rate re­sults, ideal for real-time con­trol sys­tems dur­ing con­tin­u­ous op­er­a­tion and may be used to iden­tify ir­reg­u­lar beam be­hav­ior as they de­velop.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-MOPAB314  
About • paper received ※ 20 May 2021       paper accepted ※ 07 June 2021       issue date ※ 15 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
MOPAB344 Machine Learning Models for Breakdown Prediction in RF Cavities for Accelerators cavity, operation, vacuum, linac 1068
 
  • C. Obermair, A. Apollonio, T. Cartier-Michaud, N. Catalán Lasheras, L. Felsberger, W.L. Millar, W. Wuensch
    CERN, Geneva, Switzerland
  • C. Obermair, F. Pernkopf
    TUG, Graz, Austria
 
  Radio Fre­quency (RF) break­downs are one of the most preva­lent lim­its in RF cav­i­ties for par­ti­cle ac­cel­er­a­tors. Dur­ing a break­down, field en­hance­ment as­so­ci­ated with small de­for­ma­tions on the cav­ity sur­face re­sults in elec­tri­cal arcs. Such arcs de­grade a pass­ing beam and if they occur fre­quently, they can cause ir­repara­ble dam­age to the RF cav­ity sur­face. In this paper, we pro­pose a ma­chine learn­ing ap­proach to pre­dict the oc­cur­rence of break­downs in CERN’s Com­pact LIn­ear Col­lider (CLIC) ac­cel­er­at­ing struc­tures. We dis­cuss state-of-the-art al­go­rithms for data ex­plo­ration with un­su­per­vised ma­chine learn­ing, break­down pre­dic­tion with su­per­vised ma­chine learn­ing, and re­sult val­i­da­tion with Ex­plain­able-Ar­ti­fi­cial In­tel­li­gence (Ex­plain­able AI). By in­ter­pret­ing the model pa­ra­me­ters of var­i­ous ap­proaches, we go fur­ther in ad­dress­ing op­por­tu­ni­ties to elu­ci­date the physics of a break­down and im­prove ac­cel­er­a­tor re­li­a­bil­ity and op­er­a­tion.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-MOPAB344  
About • paper received ※ 20 May 2021       paper accepted ※ 16 July 2021       issue date ※ 11 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
MOPAB411 Quantifying DNA Damage in Comet Assay Images Using Neural Networks proton, software, experiment, radiation 1233
 
  • S.J.K. Dhinsey, T. Greenshaw, C.P. Welsch
    The University of Liverpool, Liverpool, United Kingdom
  • J.L. Parsons
    Cancer Research Centre, University of Liverpool, Liverpool, United Kingdom
  • C.P. Welsch
    Cockcroft Institute, Warrington, Cheshire, United Kingdom
 
  Funding: This work was supported by the STFC Liverpool Centre for Doctoral Training on Data Intensive Science (LIV. DAT) under grant agreement ST/P006752/1.
Pro­ton ther­apy for can­cer treat­ment is a rapidly grow­ing field and in­creas­ing ev­i­dence sug­gests it in­duces more com­plex DNA dam­age than pho­ton ther­apy. Ac­cu­rate com­par­i­son be­tween the two treat­ments re­quires quan­tifi­ca­tion of the DNA dam­age the cause, which can be as­sessed using the Comet Assay. The pro­gram out­lined here is based on neural net­work ar­chi­tec­ture and aims to speed up analy­sis of Comet Assay im­ages and pro­vide ac­cu­rate, quan­tifi­able as­sess­ment of the DNA dam­age lev­els ap­par­ent in in­di­vid­ual cells. The Comet Assay is an es­tab­lished tech­nique in which DNA frag­ments are spread out under the in­flu­ence of an elec­tric field, pro­duc­ing a comet-like ob­ject. The elon­ga­tion and in­ten­sity of the comet tail (con­sist­ing of DNA frag­ments) in­di­cate the level of dam­age in­curred. Many meth­ods to mea­sure this dam­age exist, using a va­ri­ety of al­go­rithms. How­ever, these can be time con­sum­ing, so often only a small frac­tion of the comets avail­able in an image are analysed. The au­to­matic analy­sis pre­sented in this con­tri­bu­tion aims to im­prove this. To sup­ple­ment the train­ing and test­ing of the net­work, a Monte Carlo model will also be pre­sented to cre­ate sim­u­lated comet assay im­ages.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-MOPAB411  
About • paper received ※ 19 May 2021       paper accepted ※ 09 June 2021       issue date ※ 16 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB052 Current Study of Applying Machine Learning to Accelerator Physics at IHEP electron, lattice, target, photon 1477
 
  • J. Wan, Y. Jiao
    IHEP, Beijing, People’s Republic of China
 
  Funding: National Natural Science Foundation of China(No.11922512), Youth Innovation Promotion Association of Chinese Academy of Sciences(No.Y201904) and National Key R&D Program of China(No.2016YFA0401900).
In re­cent years, ma­chine learn­ing (ML) has at­tracted in­creas­ing in­ter­est among the ac­cel­er­a­tor field. As a com­plex col­lec­tion of mul­ti­ple phys­i­cal sub­sys­tems, the de­sign and op­er­a­tion of an ac­cel­er­a­tor can be very non­lin­ear and com­pli­cated, while ML is taken as a pow­er­ful tool to solve such non­lin­ear and com­pli­cated prob­lems. In this study, we re­port on sev­eral suc­cess­ful ap­pli­ca­tions of ML to ac­cel­er­a­tor physics at IHEP. The non­lin­ear dy­nam­ics op­ti­miza­tion of the High En­ergy Pho­ton Source (HEPS) that is a 4th-gen­er­a­tion light source is a chal­leng­ing topic. In this op­ti­miza­tion, we use a ML sur­ro­gate model to fast se­lect the po­ten­tially com­pet­i­tive so­lu­tions for a mul­ti­ob­jec­tive ge­netic al­go­rithm that can sig­nif­i­cantly im­prove the con­ver­gence rate and the di­ver­sity among ob­tained so­lu­tions. Be­sides, we also tried to apply a gen­er­a­tive ad­ver­sar­ial net to solve one-to-many prob­lems of lon­gi­tu­di­nal beam cur­rent pro­file shap­ing. Un­like most su­per­vised ma­chine learn­ing meth­ods than can­not learn one-to-many maps, the gen­er­a­tive ad­ver­sar­ial net-based method is able to pre­dict mul­ti­ple so­lu­tions in­stead of one for a 4-di­pole chi­cane to re­al­ize sev­eral de­sired cus­tom cur­rent pro­files.
 
poster icon Poster TUPAB052 [0.913 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB052  
About • paper received ※ 11 May 2021       paper accepted ※ 21 June 2021       issue date ※ 27 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB060 Machine Learning on Beam Lifetime and Top-Up Efficiency operation, storage-ring, emittance, photon 1499
 
  • Y.P. Sun
    ANL, Lemont, Illinois, USA
 
  Funding: The work is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357.
Both un­su­per­vised and su­per­vised ma­chine learn­ing tech­niques are em­ployed for au­to­matic clus­ter­ing, mod­el­ing and pre­dic­tion of Ad­vanced Pho­ton Source (APS) stor­age ring beam life­time and top-up ef­fi­ciency archived in op­er­a­tions. The naive Bayes clas­si­fier al­go­rithm is de­vel­oped and com­bined with k-means clus­ter­ing to im­prove ac­cu­racy, where the un­su­per­vised clus­ter­ing of APS beam life­time and top-up ef­fi­ciency is con­sis­tent with ei­ther true label from data archive or Gauss­ian ker­nel den­sity es­ti­ma­tion. Ar­ti­fi­cial neural net­work al­go­rithms have been de­vel­oped, and em­ployed for train­ing and mod­el­ling the ar­bi­trary re­la­tions of beam life­time and top-up ef­fi­ciency on many ob­serv­able pa­ra­me­ters. The pre­dic­tions from ar­ti­fi­cial neural net­work rea­son­ably agree with the APS op­er­a­tion data.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB060  
About • paper received ※ 22 May 2021       paper accepted ※ 21 June 2021       issue date ※ 22 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB061 Anomaly Detection by Principal Component Analysis and Autoencoder Approach operation, power-supply, storage-ring, photon 1502
 
  • Y.P. Sun
    ANL, Lemont, Illinois, USA
 
  Funding: The work is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357.
Sev­eral dif­fer­ent ap­proach are em­ployed to iden­tify the ab­nor­mal events in some Ad­vanced Pho­ton Source (APS) op­er­a­tion archived dataset, where di­men­sion­al­ity re­duc­tion are per­formed by ei­ther prin­ci­pal com­po­nent analy­sis or au­toen­coder ar­ti­fi­cial neural net­work. It is ob­served that the APS stored beam dump event, which is trig­gered by mag­net power sup­ply fault, may be pre­dicted by an­a­lyz­ing the mag­nets ca­pac­i­tor tem­per­a­tures dataset. There is rea­son­able agree­ment among two prin­ci­pal com­po­nent analy­sis based ap­proaches and the au­toen­coder ar­ti­fi­cial neural net­work ap­proach, on pre­dict­ing fu­ture over­all sys­tem fault which may re­sult in a stored beam dump in the APS stor­age ring.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB061  
About • paper received ※ 22 May 2021       paper accepted ※ 18 June 2021       issue date ※ 19 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB198 ESS DTL Tuning Using Machine Learning Methods cavity, DTL, linac, proton 1872
 
  • J.S. Lundquist, N. Milas, E. Nilsson
    ESS, Lund, Sweden
  • S. Werin
    Lund University, Lund, Sweden
 
  The Eu­ro­pean Spal­la­tion Source, cur­rently under con­struc­tion in Lund, Swe­den, will be the world’s most pow­er­ful neu­tron source. It is dri­ven by a pro­ton linac with a cur­rent of 62.5 mA, 2.86 ms long pulses at 14 Hz. The final sec­tion of its nor­mal-con­duct­ing front-end con­sists of a 39 m long drift tube linac (DTL) di­vided into five tanks, de­signed to ac­cel­er­ate the pro­ton beam from 3.6 MeV to 90 MeV. The high beam cur­rent and power im­pose chal­lenges to the de­sign and tun­ing of the ma­chine and the RF am­pli­tude and phase have to be set within 1% and 1 de­gree of the de­sign val­ues. The usual method used to de­fine the RF set-point is sig­na­ture match­ing, which can be a time con­sum­ing and chal­leng­ing process, and new tech­niques to meet the grow­ing com­plex­ity of ac­cel­er­a­tor fa­cil­i­ties are highly de­sir­able. In this paper we study the usage of Ma­chine Learn­ing to de­ter­mine the RF op­ti­mum am­pli­tude and phase. The data from a sim­u­lated phase scan is fed into an ar­ti­fi­cial neural net­work in order to iden­tify the needed changes to achieve the best tun­ing. Our test for the ESS DTL1 shows promis­ing re­sults, and fur­ther de­vel­op­ment of the method will be out­lined.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB198  
About • paper received ※ 17 May 2021       paper accepted ※ 21 June 2021       issue date ※ 13 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB215 Novel Non-Linear Particle Tracking Approach Employing Lie Algebraic Theory in the TensorFlow Environment quadrupole, lattice, focusing, operation 1920
 
  • J. Frank, M. Arlandoo, P. Goslawski, J. Li, T. Mertens, M. Ries, L. Vera Ramirez
    HZB, Berlin, Germany
 
  With this paper we pre­sent first re­sults for en­cod­ing Lie trans­for­ma­tions as com­pu­ta­tional graphs in Ten­sor­flow that are used as lay­ers in a neural net­work. By im­ple­ment­ing a re­cur­sive dif­fer­en­ti­a­tion scheme and em­ploy­ing Lie al­ge­braic ar­gu­ments we were able to re­pro­duce the di­a­grams for well known lat­tice con­fig­u­ra­tions. We track through sim­ple op­ti­cal lat­tices that are en­coun­tered as the main con­stituents of ac­cel­er­a­tors and demon­strate the flex­i­bil­ity and mod­u­lar­ity our ap­proach of­fers. The neural net­work can rep­re­sent the op­ti­cal lat­tice with pre­de­fined co­ef­fi­cients al­low­ing for par­ti­cle track­ing for beam dy­nam­ics or can learn from ex­per­i­men­tal data to fine-tune beam op­tics.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB215  
About • paper received ※ 12 May 2021       paper accepted ※ 31 August 2021       issue date ※ 21 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB216 Modeling Particle Stability Plots for Accelerator Optimization Using Adaptive Sampling simulation, collider, resonance, dynamic-aperture 1923
 
  • M. Schenk, L. Coyle, T. Pieloni
    EPFL, Lausanne, Switzerland
  • M. Giovannozzi, A. Mereghetti
    CERN, Meyrin, Switzerland
  • E. Krymova, G. Obozinski
    SDSC, Lausanne, Switzerland
 
  Funding: This work is partially funded by the Swiss Data Science Center (SDSC), project C18-07.
One key as­pect of ac­cel­er­a­tor op­ti­miza­tion is to max­i­mize the dy­namic aper­ture (DA) of a ring. Given the num­ber of ad­justable pa­ra­me­ters and the com­pute-in­ten­sity of DA sim­u­la­tions, this task can ben­e­fit sig­nif­i­cantly from ef­fi­cient search al­go­rithms of the avail­able pa­ra­me­ter space. We pro­pose to grad­u­ally train and im­prove a sur­ro­gate model of the DA from Six­Track sim­u­la­tions while ex­plor­ing the pa­ra­me­ter space with adap­tive sam­pling meth­ods. Here we re­port on a first model of the par­ti­cle sta­bil­ity plots using con­vo­lu­tional gen­er­a­tive ad­ver­sar­ial net­works (GAN) trained on a sub­set of Six­Track nu­mer­i­cal sim­u­la­tions for dif­fer­ent ring con­fig­u­ra­tions of the Large Hadron Col­lider at CERN.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB216  
About • paper received ※ 19 May 2021       paper accepted ※ 17 June 2021       issue date ※ 22 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB285 Broadband Imaging of Coherent Radiation as a Single-Shot Bunch Length Monitor with Femtosecond Resolution radiation, electron, simulation, detector 2147
 
  • J. Wolfenden, R.B. Fiorito, E. Kukstas, C.P. Welsch
    The University of Liverpool, Liverpool, United Kingdom
  • M. Brandin, B.S. Kyle, E. Mansten, S. Thorin
    MAX IV Laboratory, Lund University, Lund, Sweden
  • R.B. Fiorito, C.P. Welsch, J. Wolfenden
    Cockcroft Institute, Warrington, Cheshire, United Kingdom
  • E. Mansten
    Lund University, Division of Atomic Physics, Lund, Sweden
  • T.H. Pacey
    STFC/DL/ASTeC, Daresbury, Warrington, Cheshire, United Kingdom
 
  Funding: This work is supported by the AWAKE-UK project funded by STFC and the STFC Cockcroft core grant No. ST/G008248/1
Bunch length mea­sure­ments with fem­tosec­ond res­o­lu­tion are a key com­po­nent in the op­ti­mi­sa­tion of beam qual­ity in FELs, stor­age rings, and plasma-based ac­cel­er­a­tors. This con­tri­bu­tion pre­sents the de­vel­op­ment of a novel sin­gle-shot bunch length mon­i­tor with fem­tosec­ond res­o­lu­tion, based on broad­band imag­ing of the spa­tial dis­tri­b­u­tion of emit­ted co­her­ent ra­di­a­tion. The tech­nique can be ap­plied to many ra­di­a­tion sources; in this study the focus is co­her­ent tran­si­tion ra­di­a­tion (CTR) at the MAX IV Short Pulse Fa­cil­ity. Bunch lengths of in­ter­est at this fa­cil­ity are <100 fs FWHM; there­fore the CTR is in the THz to Far-IR range. To this end, a THz imag­ing sys­tem has been de­vel­oped, util­is­ing high re­sis­tiv­ity float zone sil­i­con lenses and a py­ro­elec­tric cam­era; build­ing upon pre­vi­ous re­sults where sin­gle-shot com­pres­sion mon­i­tor­ing was achieved. This con­tri­bu­tion pre­sents sim­u­la­tions of this new CTR imag­ing sys­tem to demon­strate the syn­chro­tron ra­di­a­tion mit­i­ga­tion and imag­ing ca­pa­bil­ity pro­vided, along­side ini­tial mea­sure­ments and a bunch length fit­ting al­go­rithm, ca­pa­ble of shot-to-shot op­er­a­tion. A new ma­chine learn­ing analy­sis method is also dis­cussed.
 
poster icon Poster TUPAB285 [2.008 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB285  
About • paper received ※ 17 May 2021       paper accepted ※ 24 June 2021       issue date ※ 23 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB297 Data Archive System for Superconducting RIKEN Linear Accelerator at RIBF controls, EPICS, experiment, cyclotron 2178
 
  • A. Uchiyama, N. Fukunishi, M. Kidera, M. Komiyama
    RIKEN Nishina Center, Wako, Japan
 
  At RIKEN Nishina Cen­ter, su­per­con­duct­ing RIKEN Lin­ear Ac­cel­er­a­tor (SRI­LAC) was newly in­stalled at down­stream of ex­ist­ing ac­cel­er­a­tor and up­graded for the search ex­per­i­ments of su­per-heavy-el­e­ments with atomic num­bers of 119 and higher. For the data archiv­ing and the data vi­su­al­iza­tion in RI Beam Fac­tory (RIBF) pro­ject, we have uti­lized RIBF­CAS (RIBF con­trol archive sys­tem) since 2009. For the num­ber of archived data point was ex­pected to in­crease dra­mat­i­cally for SRI­LAC, we in­tro­duced the Archiver Ap­pli­ance for im­prove­ment of the data archiv­ing per­for­mance. On the other hand, to re­al­ize a user-friendly sys­tem about the data vi­su­al­iza­tion, the data of RIBF­CAS and the Archiver Ap­pli­ance should be vi­su­al­ized on the same sys­tem. In this sys­tem, by im­ple­ment­ing a Web ap­pli­ca­tion to con­vert the RIBF­CAS data to JSON for­mat, it be­came pos­si­ble to unify the data for­mat with the Archiver Ap­pli­ance and dis­play the data with the same viewer soft­ware. In the SRI­LAC beam com­mis­sion­ing, it be­came to use­ful sys­tem for find­ing anom­alies and un­der­stand­ing the be­hav­ior of su­per­con­duct­ing cav­ity. In this con­fer­ence, we re­port the sys­tem im­ple­men­ta­tion, de­vel­oped tool, and the fu­ture plan in de­tail.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB297  
About • paper received ※ 19 May 2021       paper accepted ※ 10 June 2021       issue date ※ 17 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB309 Alignment Verification and Monitoring Strategies for the Sirius Light Source alignment, monitoring, radiation, survey 2210
 
  • R. Oliveira Neto, R. Junqueira Leão, L.R. Leão
    CNPEM, Campinas, SP, Brazil
 
  The ap­proach for the align­ment of Sir­ius is the use of portable co­or­di­nate metrol­ogy in­stru­ments in a com­mon ref­er­ence, via a net­work of sta­ble points pre­vi­ously sur­veyed. This type of net­work is com­posed of a dense dis­tri­b­u­tion of points ma­te­ri­al­ized in the form of em­bed­ded tar­get hold­ers on the spe­cial slab and ra­di­a­tion shield­ing. Phe­nom­ena such as ground move­ments, tem­per­a­ture gra­di­ents and vi­bra­tions could lead to mis­align­ment of the com­po­nents, pos­si­bly caus­ing a degra­da­tion in ma­chine per­for­mance. There­fore, the rel­a­tive po­si­tions of the ac­cel­er­a­tor mag­nets need to be pe­ri­od­i­cally ver­i­fied along with the struc­tures sur­round­ing it to en­sure a good ref­er­ence to fu­ture align­ment op­er­a­tions. This paper will pre­sent the sta­tus of Sir­ius mon­i­tor­ing sys­tems, in­clud­ing data from the first months of op­er­a­tion of the hy­dro­sta­tic lev­el­ling sen­sors. Also, pos­si­bil­i­ties with sim­pli­fied net­work mea­sure­ments for de­tect­ing struc­tural de­for­ma­tions and as­sess­ing its sta­bil­ity will be pre­sented, along with a pro­posal of a pho­togram­met­ric re­con­struc­tion of the align­ment pro­file of the stor­age ring. Fi­nally, it will be shown a com­pi­la­tion of analy­sis on the de­for­ma­tion of the Sir­ius fa­cil­i­ties.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB309  
About • paper received ※ 20 May 2021       paper accepted ※ 01 July 2021       issue date ※ 27 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB310 Establishing a Metrological Reference Network for the Alignment of Sirius alignment, survey, controls, laser 2214
 
  • H. Geraissate, G.R. Rovigatti de Oliveira
    LNLS, Campinas, Brazil
  • R. Junqueira Leão
    CNPEM, Campinas, SP, Brazil
 
  Sir­ius is the Brazil­ian 4th gen­er­a­tion syn­chro­tron light source. It con­sists of three elec­tron ac­cel­er­a­tors and it has room for up to 38 beam­lines. To make the align­ment of Sir­ius com­po­nents pos­si­ble, there is a need for a net­work of points com­pris­ing the in­stal­la­tion vol­ume, al­low­ing the lo­ca­tion of portable co­or­di­nate in­stru­ments on a com­mon ref­er­ence frame. This work de­scribes the de­vel­op­ment of such net­works for the whole Sir­ius fa­cil­ity. The lay­out of the net­works is pre­sented to­gether with the sur­vey strate­gies. De­tails are given on how the cal­cu­la­tions com­bined laser track­ers and op­ti­cal level mea­sure­ments data and how the Earth cur­va­ture com­pen­sa­tion was per­formed. A novel laser tracker ori­en­ta­tion tech­nique ap­plied for link­ing net­works on dif­fer­ent en­vi­ron­ments is also pre­sented. Fi­nally, the un­cer­tainty es­ti­ma­tion for the re­sult­ing net­work and its de­for­ma­tion his­tory is shown.  
poster icon Poster TUPAB310 [4.084 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB310  
About • paper received ※ 20 May 2021       paper accepted ※ 07 June 2021       issue date ※ 21 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB315 Development of Disaster Prevention System for Accelerator Tunnel radiation, operation, neutron, real-time 2228
 
  • K. Ishii, K. Bessho, M. Yoshioka
    KEK, Ibaraki, Japan
  • Y. Kawabata, H. Matsuda, K. Matsumoto
    Tobishima Corp., Tokyo, Japan
  • S. Tagashira
    Kansai University, Osaka, Japan
  • N. Yamamoto
    J-PARC, KEK & JAEA, Ibaraki-ken, Japan
 
  Funding: This work is supported by Health Labor Sciences Research Grant of Japan
In an en­closed space such as a par­ti­cle ac­cel­er­a­tor tun­nel, en­sur­ing worker safety dur­ing a dis­as­ter is an issue of crit­i­cal im­por­tance. It is nec­es­sary to have a sys­tem in which the man­ager can know from out­side the tun­nel whether there is any worker left be­hind and whether the worker is es­cap­ing in the right di­rec­tion. Be­cause a global po­si­tion­ing sys­tem (GPS) is not avail­able in the tun­nel, we are de­vel­op­ing a dis­as­ter pre­ven­tion sys­tem that uses Wi-Fi to trans­mit the po­si­tion­ing of work­ers and two-way com­mu­ni­ca­tion. The Wi-Fi ac­cess point (AP) in­stalled in the tun­nel should be ra­di­a­tion re­sis­tant. Ad­di­tion­ally, the equip­ment car­ried by the worker is con­ve­nient and easy to carry. We tested the ra­di­a­tion hard­ness of com­mer­cial AP de­vices and de­vel­oped a smart­phone ap­pli­ca­tion to per­form lo­ca­tion in­for­ma­tion trans­mis­sion and si­mul­ta­ne­ous char­ac­ter trans­mis­sion. In 2019, we in­stalled the sys­tem on the J-PARC Main Ring and started its op­er­a­tion. In this paper, the func­tions of the de­vel­oped sys­tem and its prospects are de­scribed.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB315  
About • paper received ※ 19 May 2021       paper accepted ※ 10 June 2021       issue date ※ 25 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB327 Developing Robust Digital Twins and Reinforcement Learning for Accelerator Control Systems at the Fermilab Booster controls, booster, power-supply, FPGA 2268
 
  • D.L. Kafkes
    Fermilab, Batavia, Illinois, USA
  • M. Schram
    JLab, Newport News, Virginia, USA
 
  Funding: This research was sponsored by the Fermilab Laboratory Directed Research and Development Program under Project ID FNAL-LDRD-2019-027: Accelerator Control with Artificial Intelligence.
We de­scribe the of­fline ma­chine learn­ing (ML) de­vel­op­ment for an ef­fort to pre­cisely reg­u­late the Gra­di­ent Mag­net Power Sup­ply (GMPS) at the Fer­mi­lab Booster ac­cel­er­a­tor com­plex via a Field-Pro­gram­ma­ble Gate Array (FPGA). As part of this ef­fort, we cre­ated a dig­i­tal twin of the Booster-GMPS con­trol sys­tem by train­ing a Long Short-Term Mem­ory (LSTM) to cap­ture its full dy­nam­ics. We out­line the path we took to care­fully val­i­date our dig­i­tal twin be­fore de­ploy­ing it as a re­in­force­ment learn­ing (RL) en­vi­ron­ment. Ad­di­tion­ally, we demon­strate the use of a Deep Q-Net­work (DQN) pol­icy model with the ca­pa­bil­ity to reg­u­late the GMPS against re­al­is­tic time-vary­ing per­tur­ba­tions.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB327  
About • paper received ※ 18 May 2021       paper accepted ※ 22 June 2021       issue date ※ 20 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB413 Rapid Browser-Based Visualization of Large Neutron Scattering Datasets neutron, scattering, experiment, detector 2494
 
  • D.L. Bruhwiler, K. Bruhwiler, P. Moeller, R. Nagler
    RadiaSoft LLC, Boulder, Colorado, USA
  • C.M. Hoffmann, Z.J. Morgan, A.T. Savici, M.G. Tucker
    ORNL, Oak Ridge, Tennessee, USA
  • A. Kuhn, J. Mensmann, P. Messmer, M. Nienhaus, S. Roemer, D. Tatulea
    NVIDIA, Santa Clara, USA
 
  Funding: This work is supported by the US Department of Energy, Office of Science, Office of Basic Energy Sciences under Award No. DE-SC0021551.
Neu­tron scat­ter­ing makes in­valu­able con­tri­bu­tions to the phys­i­cal, chem­i­cal, and nanos­truc­tured ma­te­ri­als sci­ences. Sin­gle crys­tal dif­frac­tion ex­per­i­ments col­lect vol­u­met­ric scat­ter­ing data sets rep­re­sent­ing the in­ter­nal struc­ture re­la­tions by com­bin­ing datasets of many in­di­vid­ual set­tings at dif­fer­ent ori­en­ta­tions, times and sam­ple en­vi­ron­ment con­di­tions. In par­tic­u­lar, we con­sider data from the sin­gle-crys­tal dif­frac­tion ex­per­i­ments at ORNL.* A new tech­ni­cal ap­proach for rapid, in­ter­ac­tive vi­su­al­iza­tion of re­mote neu­tron data is being ex­plored. The NVIDIA IndeX 3D vol­u­met­ric vi­su­al­iza­tion frame­work** is being used via the HTML5 client viewer from NVIDIA, the Par­aView plu­gin***, and new Jupyter note­books, which will be re­leased to the com­mu­nity with an open source li­cense.
* L. Coates et al., Rev. Sci. Instrum. 89, 092802 (2018).
** https://developer.nvidia.com/nvidia-index
*** https://blog.kitware.com/nvidia-index-plugin-in-paraview-5-5
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB413  
About • paper received ※ 18 May 2021       paper accepted ※ 21 July 2021       issue date ※ 26 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB073 An Overview of the Radio-Frequency System for an Inverse Compton X-Ray Source Based on CLIC Technology klystron, LLRF, controls, laser 2759
 
  • T.G. Lucas, O.J. Luiten, P.H.A. Mutsaers, X.F.D. Stragier, H.A. Van Doorn, F.M. van Setten, H.J.M. van den Heuvel, M.L.M.C. van der Sluis
    TUE, Eindhoven, The Netherlands
 
  Funding: This project is financed by the "Interreg V programme Flanders-Netherlands" with financial support of the European Fund for Regional Development.
Com­pact in­verse Comp­ton scat­ter­ing X-ray sources are gain­ing in pop­u­lar­ity as the fu­ture of lab-based x-ray sources. Smart*Light is one such fa­cil­ity, under com­mis­sion­ing at Eind­hoven Uni­ver­sity of Tech­nol­ogy (TU/e), which is based on high gra­di­ent X-band tech­nol­ogy orig­i­nally de­signed for the Com­pact Lin­ear Col­lider (CLIC) and its test stands lo­cated at CERN. Crit­i­cal to the beam qual­ity is the RF sys­tem which aims to de­liver 10-24 MW RF pulses at rep­e­ti­tion rates up to 1 kHz with a high am­pli­tude and phase sta­bil­ity of <0.5\% and <0.65~° al­low­ing it to ad­here to strict syn­chronic­ity con­di­tions at the in­ter­ac­tion point. This work overviews the de­sign of the high power and low level RF sys­tems for Smart*Light.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB073  
About • paper received ※ 19 May 2021       paper accepted ※ 23 June 2021       issue date ※ 29 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB184 Optimization of Medical Accelerators proton, medical-accelerators, FEL, detector 3042
 
  • C.P. Welsch
    The University of Liverpool, Liverpool, United Kingdom
  • C.P. Welsch
    Cockcroft Institute, Warrington, Cheshire, United Kingdom
 
  Funding: This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sk’odowska-Curie grant agreement No 675265.
Be­tween 2016 and 2020, 15 Fel­lows have car­ried out col­lab­o­ra­tive re­search within the 4 MEUR Op­ti­miza­tion of Med­ical Ac­cel­er­a­tors (OMA) EU-funded in­no­v­a­tive train­ing net­work. Based at uni­ver­si­ties, re­search and clin­i­cal fa­cil­i­ties, as well as in­dus­try part­ners in sev­eral Eu­ro­pean coun­tries, the Fel­lows have suc­cess­fully de­vel­oped a range of beam and pa­tient imag­ing tech­niques, im­proved bi­o­log­i­cal and phys­i­cal mod­els in Monte Carlo codes, and also help im­prove the de­sign of ex­ist­ing and fu­ture clin­i­cal fa­cil­i­ties. This paper gives an overview of the re­search out­comes of this net­work. It pre­sents re­sults from track­ing and LET mea­sure­ments with the MiniPIX-TimePIX de­tec­tor for 60 MeV clin­i­cal pro­tons, a new treat­ment plan­ning ap­proach ac­count­ing for prompt gamma range ver­i­fi­ca­tion and in­ter­frac­tional anatom­i­cal changes, and sum­ma­rizes find­ings from high-gra­di­ent test­ing of an S-band, nor­mal-con­duct­ing low phase ve­loc­ity ac­cel­er­at­ing struc­ture. Fi­nally, it gives a brief over-view of the sci­en­tific and train­ing events or­ga­nized by the OMA con­sor­tium.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB184  
About • paper received ※ 16 May 2021       paper accepted ※ 14 July 2021       issue date ※ 21 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB203 RFQ Beam Dynamics Optimization Using Machine Learning rfq, simulation, focusing, quadrupole 3100
 
  • D. Koser, J.M. Conrad, L.H. Waites, D. Winklehner
    MIT, Cambridge, Massachusetts, USA
  • A. Adelmann, M. Frey, S. Mayani
    PSI, Villigen PSI, Switzerland
 
  To ef­fi­ciently in­ject a high-cur­rent H2+ beam into the 60 MeV dri­ver cy­clotron for the pro­posed Iso­DAR pro­ject in neu­trino physics, a novel di­rect-in­jec­tion scheme is planned to be im­ple­mented using a com­pact ra­dio-fre­quency quadru­pole (RFQ) as a pre-buncher, being par­tially in­serted into the cy­clotron yoke. To op­ti­mize the RFQ beam dy­nam­ics de­sign, ma­chine learn­ing ap­proaches were in­ves­ti­gated for cre­at­ing a sur­ro­gate model of the RFQ. The re­quired sam­ple datasets are gen­er­ated by stan­dard beam dy­nam­ics sim­u­la­tion tools like PARMTEQM and RFQ­Gen or more so­phis­ti­cated PIC sim­u­la­tions. By re­duc­ing the com­pu­ta­tional com­plex­ity of multi-ob­jec­tive op­ti­miza­tion prob­lems, sur­ro­gate mod­els allow to per­form sen­si­tiv­ity stud­ies and an op­ti­miza­tion of the cru­cial RFQ beam out­put pa­ra­me­ters like trans­mis­sion and emit­tances. The time to so­lu­tion might be re­duced by up to sev­eral or­ders of mag­ni­tude. Here we dis­cuss dif­fer­ent meth­ods of sur­ro­gate model cre­ation (poly­no­mial chaos ex­pan­sion and neural net­works) and iden­tify pre­sent lim­i­ta­tions of sur­ro­gate model ac­cu­racy.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB203  
About • paper received ※ 20 May 2021       paper accepted ※ 01 July 2021       issue date ※ 30 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB289 Machine Learning Based Spatial Light Modulator Control for the Photoinjector Laser at FLUTE laser, electron, target, experiment 3332
 
  • C. Xu, E. Bründermann, A.-S. Müller, M.J. Nasse, A. Santamaria Garcia, C. Sax, C. Widmann
    KIT, Karlsruhe, Germany
  • A. Eichler
    DESY, Hamburg, Germany
 
  Funding: C. Xu acknowledges the support by the DFG-funded Doctoral School "Karlsruhe School of Elementary and Astroparticle Physics: Science and Technology".
FLUTE (Fer­n­in­frarot Linac- und Test-Ex­per­i­ment) at KIT is a com­pact linac-based test fa­cil­ity for novel ac­cel­er­a­tor tech­nol­ogy and a source of in­tense THz ra­di­a­tion. FLUTE is de­signed to pro­vide a wide range of elec­tron bunch charges from the pC- to nC-range, high elec­tric fields up to 1.2 GV/m, and ul­tra-short THz pulses down to the fs-timescale. The elec­trons are gen­er­ated at the RF pho­toin­jec­tor, where the elec­tron gun is dri­ven by a com­mer­cial ti­ta­nium sap­phire laser. In this kind of setup the elec­tron beam prop­er­ties are de­ter­mined by the pho­toin­jec­tor, but more im­por­tantly by the char­ac­ter­is­tics of the laser pulses. Spa­tial light mod­u­la­tors can be used to trans­versely and lon­gi­tu­di­nally shape the laser pulse, of­fer­ing a flex­i­ble way to shape the laser beam and sub­se­quently the elec­tron beam, in­flu­enc­ing the pro­duced THz pulses. How­ever, non­lin­ear ef­fects in­her­ent to the laser ma­nip­u­la­tion (trans­porta­tion, com­pres­sion, third har­monic gen­er­a­tion) can dis­tort the orig­i­nal pulse. In this paper we pro­pose to use ma­chine learn­ing meth­ods to ma­nip­u­late the laser and elec­tron bunch, aim­ing to gen­er­ate tai­lor-made THz pulses. The method is demon­strated ex­per­i­men­tally in a test setup.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB289  
About • paper received ※ 19 May 2021       paper accepted ※ 06 July 2021       issue date ※ 26 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB304 Multi-Objective Multi-Generation Gaussian Process Optimizer operation, framework, simulation, storage-ring 3383
 
  • X. Huang, M. Song, Z. Zhang
    SLAC, Menlo Park, California, USA
 
  Funding: DOE, Office of Science, Office of Basic Energy Sciences, DE-AC02-76SF00515 and FWP 2018-SLAC-100469 Computing Science, Office of Advanced Scientific Computing Research, FWP 2018-SLAC-100469ASCR.
We pre­sent a multi-ob­jec­tive evo­lu­tion­ary op­ti­miza­tion al­go­rithm that uses Gauss­ian process (GP) re­gres­sion-based mod­els to se­lect trial so­lu­tions in a multi-gen­er­a­tion it­er­a­tive pro­ce­dure. In each gen­er­a­tion, a sur­ro­gate model is con­structed for each ob­jec­tive func­tion with the sam­ple data. The mod­els are used to eval­u­ate so­lu­tions and to se­lect the ones with a high po­ten­tial be­fore they are eval­u­ated on the ac­tual sys­tem. Since the trial so­lu­tions se­lected by the GP mod­els tend to have bet­ter per­for­mance than other meth­ods that only rely on ran­dom op­er­a­tions, the new al­go­rithm has much higher ef­fi­ciency in ex­plor­ing the pa­ra­me­ter space. Sim­u­la­tions with mul­ti­ple test cases show that the new al­go­rithm has a sub­stan­tially higher con­ver­gence speed and sta­bil­ity than NSGA-II, MOPSO, and some other re­cent pre­s­e­lec­tion-as­sisted al­go­rithms.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB304  
About • paper received ※ 17 May 2021       paper accepted ※ 12 July 2021       issue date ※ 28 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB306 Applying Machine Learning to Optimization of Cooling Rate at Low Energy RHIC Electron Cooler electron, simulation, experiment, emittance 3391
 
  • Y. Gao, K.A. Brown, P.S. Dyer, S. Seletskiy, H. Zhao
    BNL, Upton, New York, USA
 
  Funding: Work supported by Brookhaven Science Associates, LLC under Contract No. DE-SC0012704 with the U.S. Department of Energy.
The Low En­ergy RHIC elec­tron Cooler (LEReC) is a novel, state-of-the-art, elec­tron ac­cel­er­a­tor for cool­ing RHIC ion beams, which was re­cently built and com­mis­sioned. Op­ti­miza­tion of cool­ing with LEReC re­quires fine-tun­ing of nu­mer­ous LEReC pa­ra­me­ters. In this work, ini­tial op­ti­miza­tion re­sults of using Ma­chine Learn­ing (ML) meth­ods - Bayesian Op­ti­miza­tion (BO) and Q-learn­ing are pre­sented. Spe­cially, we focus on ex­plor­ing the in­flu­ence of the elec­tron tra­jec­tory on the cool­ing rate. In the first part, sim­u­la­tions are con­ducted by uti­liz­ing a LEReC sim­u­la­tor. The re­sults show that both meth­ods have the ca­pa­bil­ity of de­riv­ing elec­tron po­si­tions that can op­ti­mize the cool­ing rate. More­over, BO takes fewer sam­ples to con­verge than the Q-learn­ing method. In the sec­ond part, Bayesian op­ti­miza­tion is fur­ther trained on the his­tor­i­cal cool­ing data. In the new sam­ples gen­er­ated by the BO, the per­cent­age of larger cool­ing rates data is greatly en­hanced com­pared with the orig­i­nal his­tor­i­cal data.
 
poster icon Poster WEPAB306 [1.083 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB306  
About • paper received ※ 12 May 2021       paper accepted ※ 01 July 2021       issue date ※ 24 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB308 Measurement-Based Surrogate Model of the SLAC LCLS-II Injector laser, simulation, controls, cathode 3395
 
  • L. Gupta, Y.K. Kim
    University of Chicago, Chicago, Illinois, USA
  • A.L. Edelen, C.E. Mayes, A.A. Mishra, N.R. Neveu
    SLAC, Menlo Park, California, USA
 
  Funding: This project was funded by the DOE SCGSR Program.
There is sig­nif­i­cant ef­fort within par­ti­cle ac­cel­er­a­tor physics to use ma­chine learn­ing meth­ods to im­prove mod­el­ing of ac­cel­er­a­tor com­po­nents. Such mod­els can be made re­al­is­tic and rep­re­sen­ta­tive of ma­chine com­po­nents by train­ing them with mea­sured data. These mod­els could be used as vir­tual di­ag­nos­tics or for model-based con­trol when fast feed­back is needed for tun­ing to dif­fer­ent user set­tings. To pro­to­type such a model, we demon­strate how a ma­chine learn­ing based sur­ro­gate model of the SLAC LCLS-II pho­to­cath­ode in­jec­tor was de­vel­oped. To cre­ate ma­chine-based data, laser mea­sure­ments were taken at the LCLS using the vir­tual cath­ode cam­era. These mea­sure­ments were used to sam­ple par­ti­cles, re­sult­ing in re­al­is­tic elec­tron bunches, which were then prop­a­gated through the in­jec­tor via the Astra space charge sim­u­la­tion. By doing this, the model is not only able to pre­dict many bulk elec­tron beam pa­ra­me­ters and dis­tri­b­u­tions which are often hard to mea­sure or not usu­ally avail­able to mea­sure, but the pre­dic­tions are more re­al­is­tic rel­a­tive to tra­di­tion­ally sim­u­lated train­ing data. The meth­ods for train­ing such mod­els, as well as model ca­pa­bil­i­ties and fu­ture work are pre­sented here.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB308  
About • paper received ※ 26 May 2021       paper accepted ※ 27 July 2021       issue date ※ 24 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB309 Study and Design of the Appropriate High-Performance Computing System for Beamline Data Analysis Application at Iranian Light Source Facility (ILSF) software, hardware, experiment, data-analysis 3399
 
  • A. Khaleghi, M. Akbari
    ILSF, Tehran, Iran
  • H. Haedar, K. Mahmoudi, M. Takhttavani
    IKIU, Qazvin, Iran
  • S. Mahmoudi
    Sharif University of Technology (SUT), Tehran, Iran
 
  Data analy­sis is a very im­por­tant step in doing ex­per­i­ments at light sources, where mul­ti­ple ap­pli­ca­tion and soft­ware pack­ages are used for this pur­pose. In this paper we have re­viewed some soft­ware pack­ages that are used for data analy­sis and de­sign at Iran­ian Light Source Fa­cil­ity then ac­cord­ing to their pro­cess­ing needs, after tak­ing in mind dif­fer­ent HPC sce­nar­ios a suit­able ar­chi­tec­ture for de­ploy­ment of the ILSF HPC is pre­sented. The pro­posed ar­chi­tec­ture is a clus­ter of 64 com­put­ing nodes con­nected through Eth­er­net and In­fini­Band net­work run­ning a Linux op­er­at­ing sys­tem with sup­port of MPI par­al­lel en­vi­ron­ment.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB309  
About • paper received ※ 19 May 2021       paper accepted ※ 23 July 2021       issue date ※ 01 September 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB310 Study and Design of a High-Performance Computing Infrastructure for Iranian Light Source Facility Based on the Accelerator Physicists and Engineers’ Applications Requirements software, hardware, Ethernet, simulation 3402
 
  • K. Mahmoudi, H. Haedar, A. Khaleghi
    IKIU, Qazvin, Iran
  • M. Akbari, A. Khaleghi
    ILSF, Tehran, Iran
  • S. Mahmoudi
    IUST, Narmac, Tehran, Iran
 
  Syn­chro­tron de­sign and op­er­a­tion are one of the com­plex tasks which re­quires a lot of pre­cise com­pu­ta­tion. As an ex­am­ple, we could men­tion the sim­u­la­tions done for cal­cu­lat­ing the im­ped­ance bud­get of the ma­chine which re­quires a no­table amount of com­pu­ta­tional power. In this paper we are going to re­view dif­fer­ent HPC sce­nar­ios suit­able for this mat­ter then we will pre­sent our de­sign of a suit­able HPC based on the ac­cel­er­a­tor physi­cists and en­gi­neers’ needs. Going through dif­fer­ent HPC sce­nar­ios such as shared mem­ory ar­chi­tec­tures, dis­trib­uted mem­ory ar­chi­tec­tures, clus­ter, grid and cloud com­put­ing we con­clude im­ple­men­ta­tion of a ded­i­cated com­put­ing clus­ter can be de­sired for ILSF. Clus­ter com­put­ing pro­vides the op­por­tu­nity for easy and saleable sci­en­tific com­pu­ta­tion for ILSF also an­other ad­van­tage is that its re­sources can be used for run­ning cloud or grid com­put­ing plat­forms as well.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB310  
About • paper received ※ 19 May 2021       paper accepted ※ 19 July 2021       issue date ※ 12 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB318 Prediction and Clustering of Longitudinal Phase Space Images and Machine Parameters Using Neural Networks and K-Means Algorithm FEL, simulation, electron, ECR 3417
 
  • M. Maheshwari
    STFC/DL, Daresbury, Warrington, Cheshire, United Kingdom
  • D.J. Dunning, J.K. Jones, M.P. King, H.R. Kockelbergh, A.E. Pollard
    STFC/DL/ASTeC, Daresbury, Warrington, Cheshire, United Kingdom
 
  Ma­chine learn­ing al­go­rithms were used for image and pa­ra­me­ter recog­ni­tion and gen­er­a­tion with the aim to op­ti­mise the CLARA fa­cil­ity at Dares­bury, using start-to-end sim­u­la­tion data. Con­vo­lu­tional and fully con­nected neural net­works were trained using Ten­sor­Flow-Keras for dif­fer­ent in­stances, with ex­am­ples in­clud­ing pre­dict­ing Lon­gi­tu­di­nal Phase Space (LPS) im­ages with ma­chine pa­ra­me­ters as input and FEL pa­ra­me­ter pre­dic­tion (e.g. pulse en­ergy) from LPS im­ages. The K-means clus­ter­ing al­go­rithm was used to clus­ter the LPS im­ages to high­light pat­terns within the data. Ma­chine learn­ing tech­niques can en­hance the way large amounts of data are processed and analysed and so have great po­ten­tial for ap­pli­ca­tion in ac­cel­er­a­tor sci­ence R&D.  
poster icon Poster WEPAB318 [1.062 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB318  
About • paper received ※ 17 May 2021       paper accepted ※ 05 July 2021       issue date ※ 21 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB329 LCLS-II Average Current Monitor cavity, vacuum, coupling, simulation 3443
 
  • P. Borchard, J.S. Hoh
    Dymenso LLC, San Francisco, USA
 
  The LCLS-II pro­ject at SLAC is a high power up­grade to the ex­ist­ing free-elec­tron laser fa­cil­ity. The LCLS-II Ac­cel­er­a­tor Sys­tem will in­clude a new 4 GeV con­tin­u­ous-wave su­per­con­duct­ing lin­ear ac­cel­er­a­tor in the first kilo­me­ter of the SLAC lin­ear ac­cel­er­a­tor tun­nel and sup­ple­ments the ex­ist­ing low power pulsed linac. Av­er­age Cur­rent Mon­i­tors (ACMs) are needed to pro­tect against ex­ces­sive beam power which might oth­er­wise cause dam­age to the beam dumps. The ACM cav­i­ties are pill­box-shaped stain­less steel RF cav­ity with two ra­dial probe ports with cou­plers, one ra­dial test port with a cou­pler, and a mech­a­nism for me­chan­i­cally fine-tun­ing the cav­ity res­o­nant fre­quency. The ACM RF cav­i­ties will be lo­cated at points of known or con­strained beam en­ergy and will mon­i­tor the beam cur­rent, a safety sys­tem will trip off the beam if the beam power ex­ceeds the al­lowed value.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB329  
About • paper received ※ 19 May 2021       paper accepted ※ 16 June 2021       issue date ※ 22 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB356 Proposal of an Alignment System for HALF: The Reference Network of Alignment alignment, monitoring, real-time, simulation 3533
 
  • X. Li, J.X. Chen, X.Y. He, W. Wang, Z.Y. Wang
    USTC/NSRL, Hefei, Anhui, People’s Republic of China
  • J.X. Chen, T. Luo
    IHEP CSNS, Guangdong Province, People’s Republic of China
 
  As a fourth-gen­er­a­tion light source based on the dif­frac­tion-lim­ited stor­age ring, Hefei Ad­vanced Light Fa­cil­ity (HALF) has higher re­quire­ments for mag­nets align­ment in ac­cu­racy, ef­fi­ciency, and re­li­a­bil­ity. In this paper, the Ref­er­ence Net­work of Align­ment (RNA) sys­tem is pro­posed to im­prove the mag­netic axis align­ment ac­cu­racy on the ra­dial di­rec­tion of the beam­line. Herein, we mainly in­tro­duce the con­cept de­sign and the the­o­ret­i­cal analy­sis of the RNA sys­tem, which cen­ter on the novel fu­sion method of sen­sors. A sim­u­la­tion re­sult shows that it is cred­i­ble to as­sume the RNA sys­tem can achieve an align­ment in­stal­la­tion ac­cu­racy of 20 µm and a dis­place­ment mon­i­tor­ing ac­cu­racy of 10 µm.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB356  
About • paper received ※ 16 May 2021       paper accepted ※ 21 June 2021       issue date ※ 31 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB373 The Energy Management System in NSRRC operation, controls, MMI, radiation 3585
 
  • C.S. Chen, W.S. Chan, Y.Y. Cheng, Y.F. Chiu, Y.-C. Chung, K.C. Kuo, M.T. Lee, Y.C. Lin, C.Y. Liu, Z.-D. Tsai
    NSRRC, Hsinchu, Taiwan
 
  Tai­wan has been suf­fer­ing from a short­age of nat­ural re­sources for more than two decades. As stated by the En­ergy Sta­tis­tics Hand­book 2019 of Tai­wan, up to 97.90% of en­ergy sup­ply was im­ported from abroad. This kind of en­ergy con­sump­tion struc­ture is frag­ile rel­a­tively. Not men­tion to the total do­mes­tic en­ergy con­sump­tion an­nual growth rate is 1.97% in twenty years. Ei­ther the semi­con­duc­tor or the in­te­grated cir­cuit-re­lated in­dus­try is de­vel­oped vig­or­ously in Tai­wan. All the facts cause us to face the en­ergy prob­lems squarely. There­fore, an en­ergy man­age­ment sys­tem (EnMS) was in­stalled in NSRRC in 2019 to pur­sue more ef­fi­cient en­ergy use. With the ad­van­tages of the Archive Viewer - a util­ity su­per­vi­sory con­trol and data ac­qui­si­tion sys­tem in NSRRC, the data of en­ergy use could be traced con­ve­niently and widely. The model of en­ergy use has been built to re­view pe­ri­od­i­cally, fur­ther­more, it pro­vides us the ac­cor­dance to re­place the de­graded equip­ment and alerts us if the fail­ure oc­curs.  
poster icon Poster WEPAB373 [0.497 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB373  
About • paper received ※ 21 May 2021       paper accepted ※ 22 July 2021       issue date ※ 11 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB374 The Southern Hemisphere’s First X-Band Radio-Frequency Test Facility at the University of Melbourne electron, klystron, gun, GUI 3588
 
  • M. Volpi, R.P. Rassool, S.L. Sheehy, G. Taylor, S.D. Williams
    The University of Melbourne, Melbourne, Victoria, Australia
  • M.J. Boland
    CLS, Saskatoon, Saskatchewan, Canada
  • M.J. Boland
    University of Saskatchewan, Saskatoon, Canada
  • N. Catalán Lasheras, S. Gonzalez Anton, G. McMonagle, S. Stapnes, W. Wuensch
    CERN, Meyrin, Switzerland
  • R.T. Dowd, K. Zingre
    AS - ANSTO, Clayton, Australia
 
  The first South­ern Hemi­sphere X-band Lab­o­ra­tory for Ac­cel­er­a­tors and Beams (X-LAB) is under con­struc­tion at the Uni­ver­sity of Mel­bourne, and it will op­er­ate CERN X-band test stand con­tain­ing two 12GHz 6MW kly­stron am­pli­fiers. By power com­bi­na­tion through hy­brid cou­plers and the use of pulse com­pres­sors, up to 50 MW of peak power can be sent to any of 2 test slots at pulse rep­e­ti­tion rates up to 400 Hz. The test stand is ded­i­cated to RF con­di­tion­ing and test­ing CLIC’s high gra­di­ent ac­cel­er­at­ing struc­tures be­yond 100 MV/m. It will also form the basis for de­vel­op­ing a com­pact ac­cel­er­a­tor for med­ical ap­pli­ca­tions, such as ra­dio­ther­apy and com­pact light sources. Aus­tralian re­searchers work­ing as part of a col­lab­o­ra­tion be­tween the Uni­ver­sity of Mel­bourne, in­ter­na­tional uni­ver­si­ties, na­tional in­dus­tries, the Aus­tralian Syn­chro­tron -ANSTO, Cana­dian Light Source and the CERN be­lieve that cre­at­ing a lab­o­ra­tory for novel ac­cel­er­a­tor re­search in Aus­tralia could drive tech­no­log­i­cal and med­ical in­no­va­tion.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB374  
About • paper received ※ 18 May 2021       paper accepted ※ 06 July 2021       issue date ※ 30 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB398 A C-Band RF Mode Launcher with Quadrupole Field Components Cancellation for High Brightness Applications quadrupole, GUI, brightness, linac 3638
 
  • G. Pedrocchi
    SBAI, Roma, Italy
  • D. Alesini, F. Cardelli, A. Gallo, A. Giribono, B. Spataro
    INFN/LNF, Frascati, Italy
  • G. Castorina
    AVO-ADAM, Meyrin, Switzerland
  • L. Ficcadenti
    INFN-Roma, Roma, Italy
  • M. Migliorati, A. Mostacci, L. Palumbo
    Sapienza University of Rome, Rome, Italy
 
  The R&D of high gra­di­ent ra­diofre­quency de­vices is aimed to de­velop in­no­v­a­tive and com­pact ac­cel­er­at­ing stuc­tures based on new man­u­fac­tor­ing tech­niques and ma­te­ri­als in order to pro­duce de­vices op­er­at­ing with the high­est ac­cel­er­at­ing gra­di­ent. Re­cent stud­ies have shown a large in­crease in the max­i­mum sus­tained RF sur­face elec­tric fields in cop­per struc­ture op­er­at­ing at cryo­genic tem­per­a­ture. These novel ap­proaches allow sig­nif­i­cant per­for­mance im­prove­ments of RF pho­toin­jec­tors. In­deed the op­er­a­tion at high sur­face fields re­sults in con­sid­er­able in­crease of elec­tron bril­liance. This re­quires high field qual­ity in the RF pho­toin­jec­tor and specif­i­cally in its poweer cou­pler. In this work we pre­sent a novel power cou­pler for the RF pho­toin­jec­tor. The cou­pler is a com­pact C-band TM01 mode launcher with a four­fold sym­me­try which min­i­mized both the di­pole and the quadru­pole RF field com­po­nents.  
poster icon Poster WEPAB398 [1.799 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB398  
About • paper received ※ 13 May 2021       paper accepted ※ 06 July 2021       issue date ※ 23 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
WEPAB412 Use of a Noise IoT Detection System to Measure the Environmental Noise in Taiwan Light Source monitoring, real-time, site, experiment 3671
 
  • P.J. Wen, S.P. Kao, S.Y. Lin, Y.C. Lin
    NSRRC, Hsinchu, Taiwan
 
  In the past, the method of gen­eral noise mon­i­tor­ing al­tered lit­tle; noise was still mea­sured with a human hand-held mo­bile de­vice, or the mea­sure­ment at fixed sites was made using tra­di­tional ana­logue data-stor­age equip­ment. In re­cent years, with the rapidly im­proved net­work trans­mis­sion ca­pa­bil­i­ties, the de­vel­op­ment of a small noise-de­tec­tion IoT sys­tem al­lows the de­tec­tion data to be trans­mit­ted wire­lessly with­out need for human strength mea­sure­ments, and records noise in­for­ma­tion. The sta­tis­tics of sub­se­quent noise data be­come a basis for analy­sis and im­prove­ment. Tai­wan Light Source (TLS) beam­lines have many vac­uum pumps, cool­ing pumps, liq­uid-ni­tro­gen pres­sure-re­lief sys­tems, com­puter servers etc. that gen­er­ate much noise. This study is ex­pected to pre­pare for in­stal­la­tion of noise de­tec­tion. The sys­tem uses a noise-de­tec­tion box to de­tect, to dis­close louder lo­ca­tions, to col­lect noise data, to de­ter­mine the source and type of noise source, and to pro­vide in­for­ma­tion to re­duce the noise of the work­ing en­vi­ron­ment. The TLS noise-de­tec­tion re­sults find that the in­ner-ring area has less noise and are more sta­ble than the outer ring area.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB412  
About • paper received ※ 14 May 2021       paper accepted ※ 24 June 2021       issue date ※ 27 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THXB06 Results of the First Alignment Run for Sirius alignment, laser, operation, survey 3728
 
  • R. Junqueira Leão, R. Oliveira Neto
    CNPEM, Campinas, SP, Brazil
  • H. Geraissate, F. Rodrigues, G.R. Rovigatti de Oliveira
    LNLS, Campinas, Brazil
 
  It is widely known that the po­si­tion of par­ti­cle ac­cel­er­a­tor com­po­nents is crit­i­cal for its per­for­mance. For the lat­est gen­er­a­tion light sources, whose mag­netic lat­tice is op­ti­mized for achiev­ing very low emit­tance, the tol­er­a­ble mis­align­ments are in the order of a few dozen mi­crom­e­ters. Due to the perime­ter of these ma­chines, the re­quire­ments push the lim­its of large-vol­ume di­men­sional metrol­ogy and as­so­ci­ated in­stru­ments and tech­niques. Re­cently a fine align­ment cam­paign of the Sir­ius ac­cel­er­a­tors was con­ducted fol­low­ing the pre-align­ment per­formed dur­ing the in­stal­la­tion phase. To con­form with the strict rel­a­tive po­si­tion­ing de­mands, mea­sure­ment good prac­tices were fol­lowed, and sev­eral 3D metrol­ogy pro­ce­dures were de­vel­oped. Also, to im­prove po­si­tion­ing res­o­lu­tion, high rigid­ity trans­la­tion de­vices were pro­duced. Fi­nally, the spe­cial tar­get hold­ers de­signed as re­mov­able fidu­cials for the mag­nets were re­vis­ited to as­sure max­i­mum re­li­a­bil­ity. Data pro­cess­ing al­go­rithms were im­ple­mented to eval­u­ate the align­ment re­sults in a ro­bust and agile man­ner. This paper will pre­sent the final po­si­tion­ing er­rors for Sir­ius mag­nets with an ex­pres­sion of the es­ti­mated un­cer­tainty.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THXB06  
About • paper received ※ 20 May 2021       paper accepted ※ 02 July 2021       issue date ※ 01 September 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THPAB068 Denoising of Optics Measurements Using Autoencoder Neural Networks optics, simulation, MMI, controls 3915
 
  • E. Fol, R. Tomás García
    CERN, Meyrin, Switzerland
 
  Noise arte­facts can ap­pear in op­tics mea­sure­ments data due to in­stru­men­ta­tion im­per­fec­tions or un­cer­tain­ties in the ap­plied analy­sis meth­ods. A spe­cial type of semi-su­per­vised neural net­works, au­toen­coders, are widely ap­plied to de­nois­ing tasks in image and sig­nal pro­cess­ing as well as to gen­er­a­tive mod­el­ing. Re­cently, an au­toen­coder-based ap­proach for de­nois­ing and re­con­struc­tion of miss­ing data has been de­vel­oped to im­prove the qual­ity of phase mea­sure­ments ob­tained from har­monic analy­sis of LHC turn-by-turn data. We pre­sent the re­sults achieved on sim­u­la­tions demon­strat­ing the po­ten­tial of the new method and dis­cuss the ef­fect of the noise in light of op­tics cor­rec­tions com­puted from the cleaned data.  
poster icon Poster THPAB068 [0.881 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB068  
About • paper received ※ 19 May 2021       paper accepted ※ 13 July 2021       issue date ※ 02 September 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THPAB191 Physics-Enhanced Reinforcement Learning for Optimal Control lattice, controls, simulation, alignment 4150
 
  • A.N. Ivanov, I.V. Agapov, A. Eichler, S. Tomin
    DESY, Hamburg, Germany
 
  We pro­pose an ap­proach for in­cor­po­rat­ing ac­cel­er­a­tor physics mod­els into re­in­force­ment learn­ing agents. The pro­posed ap­proach is based on the Tay­lor map­ping tech­nique for sim­u­la­tion of the par­ti­cle dy­nam­ics. The re­sult­ing com­pu­ta­tional graph is rep­re­sented as a poly­no­mial neural net­work and em­bed­ded into the tra­di­tional re­in­force­ment learn­ing agents. The ap­pli­ca­tion of the model is demon­strated in a non­lin­ear sim­u­la­tion model of beam trans­mis­sion. The com­par­i­son of the ap­proach with the tra­di­tional nu­mer­i­cal op­ti­miza­tion as well as neural net­works based agents demon­strates bet­ter con­ver­gence of the pro­posed tech­nique.  
poster icon Poster THPAB191 [0.846 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB191  
About • paper received ※ 11 May 2021       paper accepted ※ 29 July 2021       issue date ※ 24 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THPAB197 Enhancing Efficiency of Multi-Objective Neural-Network-Assisted Nonlinear Dynamics Lattice Optimization via 1-D Aperture Objectives & Objective Focusing lattice, focusing, simulation, storage-ring 4156
 
  • Y. Hidaka, D.A. Hidas, F. Plassard, T.V. Shaftan, G.M. Wang
    BNL, Upton, New York, USA
 
  Funding: This work is supported by U.S. DOE under Contract No. DE-SC0012704.
Mutli-ob­jec­tive op­ti­miz­ers such as multi-ob­jec­tive ge­netic al­go­rithm (MOGA) have been quite pop­u­lar in dis­cov­er­ing de­sir­able lat­tice so­lu­tions for ac­cel­er­a­tors. How­ever, even these suc­cess­ful al­go­rithms can be­come in­ef­fec­tive as the di­men­sion and range of the search space in­crease due to ex­po­nen­tial growth in the amount of ex­plo­ration re­quired to find global op­tima. This dif­fi­culty is even more ex­ac­er­bated by the re­source-in­ten­sive and time-con­sum­ing ten­dency for the eval­u­a­tions of non­lin­ear beam dy­nam­ics. Lately the use of sur­ro­gate mod­els based on neural net­work has been draw­ing at­ten­tion to al­le­vi­ate this prob­lem. Fol­low­ing this trend, to fur­ther en­hance the ef­fi­ciency of non­lin­ear lat­tice op­ti­miza­tion for stor­age rings, we pro­pose to re­place typ­i­cally used ob­jec­tives with those that are less time-con­sum­ing and to focus on a sin­gle ob­jec­tive con­structed from mul­ti­ple ob­jec­tives, which can max­i­mize uti­liza­tion of the trained mod­els through local op­ti­miza­tion and ob­jec­tive gra­di­ent ex­trac­tion. We demon­strate these en­hance­ments using a NSLS-II up­grade lat­tice can­di­date as an ex­am­ple.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB197  
About • paper received ※ 20 May 2021       paper accepted ※ 23 June 2021       issue date ※ 10 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THPAB201 A Machine Learning Technique for Dynamic Aperture Computation dynamic-aperture, simulation, hadron, distributed 4172
 
  • B. Dalena, M. Ben Ghali
    CEA-IRFU, Gif-sur-Yvette, France
 
  Cur­rently, dy­namic aper­ture cal­cu­la­tions of high-en­ergy hadron col­lid­ers are per­formed through com­puter sim­u­la­tions, which are both a re­source-heavy and time-costly processes. The aim of this study is to use a reser­voir com­put­ing ma­chine learn­ing model in order to achieve a faster ex­trap­o­la­tion of dy­namic aper­ture val­ues. A re­cur­rent echo-state net­work (ESN) ar­chi­tec­ture is used as a basis for this work. Re­cur­rent net­works are bet­ter fit­ted to ex­trap­o­la­tion tasks while the reser­voir echo-state struc­ture is com­pu­ta­tion­ally ef­fec­tive. Model train­ing and val­i­da­tion is con­ducted on a set of "seeds" cor­re­spond­ing to the sim­u­la­tion re­sults of dif­fer­ent ma­chine con­fig­u­ra­tions. Ad­just­ments in the model ar­chi­tec­ture, man­ual met­ric and data se­lec­tion, hy­per-pa­ra­me­ters tun­ing and the in­tro­duc­tion of new pa­ra­me­ters en­abled the model to re­li­ably achieve good per­for­mance on ex­am­in­ing test­ing sets.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB201  
About • paper received ※ 14 May 2021       paper accepted ※ 22 July 2021       issue date ※ 02 September 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THPAB243 Optimizing Mu2e Spill Regulation System Algorithms extraction, controls, simulation, resonance 4281
 
  • A. Narayanan
    Northern Illinois University, DeKalb, Illinois, USA
  • K.J. Hazelwood, M.A. Ibrahim, V.P. Nagaslaev, D.J. Nicklaus, P.S. Prieto, B.A. Schupbach, K. Seiya, R.M. Thurman-Keup, N.V. Tran
    Fermilab, Batavia, Illinois, USA
  • H. Liu, S. Memik, R. Shi, M. Thieme
    Northwestern University, Evanston, Illinois, USA
 
  Funding: The work has been performed at Fermilab. Fermilab is managed by Fermi Research Alliance, LLC (FRA), acting under Contract No. DE-AC02-07CH11359.
A slow ex­trac­tion sys­tem is being de­vel­oped for the Fer­mi­lab’s De­liv­ery Ring to de­liver pro­tons to the Mu2e ex­per­i­ment. Dur­ing the ex­trac­tion, the beam on tar­get ex­pe­ri­ences small in­ten­sity vari­a­tions owing to many fac­tors. Var­i­ous adap­tive learn­ing al­go­rithms will be em­ployed for beam reg­u­la­tion to achieve the re­quired spill qual­ity. We dis­cuss here pre­lim­i­nary re­sults of the slow and fast reg­u­la­tion al­go­rithms val­i­da­tion through the com­puter sim­u­la­tions be­fore their im­ple­men­ta­tion in the FPGA. Par­ti­cle track­ing with sex­tu­pole res­o­nance was used to de­ter­mine the fine shape of the spill pro­file. Fast semi-an­a­lyt­i­cal sim­u­la­tion schemes and Ma­chine Learn­ing mod­els were used to op­ti­mize the fast reg­u­la­tion loop.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB243  
About • paper received ※ 20 May 2021       paper accepted ※ 28 July 2021       issue date ※ 20 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THPAB260 Detection and Classification of Collective Beam Behaviour in the LHC extraction, operation, controls, injection 4318
 
  • L. Coyle, F. Blanc, T. Pieloni, M. Schenk
    EPFL, Lausanne, Switzerland
  • X. Buffat, M. Solfaroli Camillocci, J. Wenninger
    CERN, Meyrin, Switzerland
  • E. Krymova, G. Obozinski
    SDSC, Lausanne, Switzerland
 
  Col­lec­tive in­sta­bil­i­ties can lead to a se­vere de­te­ri­o­ra­tion of beam qual­ity, in terms of re­duced beam in­ten­sity and in­creased beam emit­tance, and con­se­quently a re­duc­tion of the col­lider’s lu­mi­nos­ity. It is there­fore cru­cial for the op­er­a­tion of the CERN’s Large Hadron Col­lider to un­der­stand the con­di­tions in which they ap­pear in order to find ap­pro­pri­ate mit­i­ga­tion mea­sures. Using bunch-by-bunch and turn-by-turn beam am­pli­tude data, cour­tesy of the trans­verse damper’s ob­ser­va­tion box (Ob­s­Box), a novel ma­chine learn­ing based ap­proach is de­vel­oped to both de­tect and clas­sify these in­sta­bil­i­ties. By train­ing an au­toen­coder neural net­work on the Ob­s­Box am­pli­tude data and using the model’s re­con­struc­tion error, in­sta­bil­i­ties and other phe­nom­ena are sep­a­rated from nom­i­nal beam be­hav­iour. Ad­di­tion­ally, the la­tent space en­cod­ing of this au­toen­coder of­fers a unique image like rep­re­sen­ta­tion of the beam am­pli­tude sig­nal. Lever­ag­ing this la­tent space rep­re­sen­ta­tion al­lows us to clus­ter the var­i­ous types of anom­alous sig­nals.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB260  
About • paper received ※ 19 May 2021       paper accepted ※ 19 July 2021       issue date ※ 27 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THPAB268 Hierarchical Intelligent Real-Time Optimal Control for LLRF Using Time Series Machine Learning Methods and Transfer Learning controls, LLRF, cavity, simulation 4329
 
  • R. Pirayesh, S. Biedron
    UNM-ME, Albuquerque, New Mexico, USA
  • S. Biedron, J.A. Diaz Cruz, M. Martínez-Ramón
    UNM-ECE, Albuquerque, USA
  • J.A. Diaz Cruz
    SLAC, Menlo Park, California, USA
 
  Funding: supported by DOE, Office of Science, Office of High Energy Physics, under award number DE-SC0019468, Contract No. DE-AC02-76SF00515, also supported Office of Basic Energy Sciences. ALCF, Element Aero
Ma­chine learn­ing (ML) has re­cently been ap­plied to Low-level RF (LLRF) con­trol sys­tems to keep the volt­age and phase of Su­per­con­duct­ing Ra­diofre­quency (SRF) cav­i­ties sta­ble within 0.01 de­gree in phase and 0.01% am­pli­tude as con­straints. Model pre­dic­tive con­trol (MPC) uses an op­ti­miza­tion al­go­rithm of­fline to min­i­mize a cost func­tion with con­straints on the states and con­trol input. The sur­ro­gate model op­ti­mally con­trols the cav­i­ties on­line. Time se­ries deep ML struc­tures in­clud­ing re­cur­rent neural net­work (RNN) and long short-term mem­ory (LSTM) can model the con­trol input of MPC and dy­nam­ics of LLRF as a sur­ro­gate model. When the pre­dicted states di­verge from the mea­sured states more than a thresh­old at each time step, the states’ mea­sure­ments from the cav­ity fine-tune the sur­ro­gate model with trans­fer learn­ing. MPC does the op­ti­miza­tion of­fline again with the up­dated sur­ro­gate model, and, next, trans­fer learn­ing fine-tunes the sur­ro­gate model with the new data from the op­ti­mal con­trol in­puts. The sur­ro­gate model pro­vides us with a com­pu­ta­tion­ally faster and ac­cu­rate mod­el­ing of MPC and LLRF, which in turn re­sults in a more sta­ble con­trol sys­tem.
Machine learning, Surrogate model, control, LLRF, MPC, Transfer learning
 
poster icon Poster THPAB268 [0.377 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB268  
About • paper received ※ 16 May 2021       paper accepted ※ 13 July 2021       issue date ※ 18 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THPAB273 Spectral Reconstruction for FACET-II Compton Spectrometer photon, electron, site, positron 4346
 
  • Y. Zhuang, B. Naranjo, J.B. Rosenzweig, M. Yadav
    UCLA, Los Angeles, USA
 
  Funding: This work was supported by DOE Contract DE-SC0009914, NSF Grant No. PHY-1549132, and DARPA GRIT Contract 20204571.
The Comp­ton spec­trom­e­ter under de­vel­op­ment at UCLA for FACET-II is a ver­sa­tile tool to an­a­lyze gamma-ray spec­tra in a sin­gle shot, in which the en­ergy and an­gu­lar po­si­tion of the in­com­ing pho­tons are recorded by ob­serv­ing the mo­menta and po­si­tion of Comp­ton scat­tered elec­trons. We pre­sent meth­ods to re­con­struct the pri­mary spec­trum from these data via ma­chine learn­ing and the EM Al­go­rithm. A multi-layer fully con­nected neural net­work is used to per­form the re­gres­sion task of re­con­struct­ing both the dou­ble-dif­fer­en­tial spec­trum and the pho­ton en­ergy spec­trum in­ci­dent with zero an­gu­lar off­set. We pre­sent the ex­pected per­for­mance of these tech­niques, con­cen­trat­ing on the achiev­able en­ergy res­o­lu­tion.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB273  
About • paper received ※ 20 May 2021       paper accepted ※ 28 July 2021       issue date ※ 16 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THPAB349 Feed-Forward Neural Network Based Modelling of an Ultrafast Laser for Enhanced Control laser, controls, electron, cathode 4478
 
  • A. Aslam, M. Martínez-Ramón, S.D. Scott
    UNM-ECE, Albuquerque, USA
  • S. Biedron
    Argonne National Laboratory, Office of Naval Research Project, Argonne, Illinois, USA
  • S. Biedron
    Element Aero, Chicago, USA
  • S. Biedron
    UNM-ME, Albuquerque, New Mexico, USA
  • M. Burger, J. Murphy
    NERS-UM, Ann Arbor, Michigan, USA
  • K.M. Krushelnick, J. Nees, A.G.R. Thomas
    University of Michigan, Ann Arbor, Michigan, USA
  • Y. Ma
    IHEP, Beijing, People’s Republic of China
  • Y. Ma
    Michigan University, Ann Arbor, Michigan, USA
 
  Funding: Acknowledgements: This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of High Energy Physics, under award number DE-SC0019468.
The ap­pli­ca­tions of ma­chine learn­ing in today’s world en­com­pass all fields of life and phys­i­cal sci­ences. In this paper, we im­ple­ment a ma­chine learn­ing based al­go­rithm in the con­text of laser physics and par­ti­cle ac­cel­er­a­tors. Specif­i­cally, a neural net­work-based op­ti­mi­sa­tion al­go­rithm has been de­vel­oped that of­fers en­hanced con­trol over an ul­tra­fast fem­tosec­ond laser in com­par­i­son to the tra­di­tional Pro­por­tional In­te­gral and de­riv­a­tive (PID) con­trols. This re­search opens a new po­ten­tial of util­is­ing ma­chine learn­ing and even deep learn­ing tech­niques to im­prove the per­for­mance of sev­eral dif­fer­ent lasers and ac­cel­er­a­tors sys­tems.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB349  
About • paper received ※ 20 May 2021       paper accepted ※ 02 July 2021       issue date ※ 17 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
FRXC01 Superconducting Radio-Frequency Cavity Fault Classification Using Machine Learning at Jefferson Laboratory cavity, cryomodule, SRF, radio-frequency 4535
 
  • C. Tennant, A. Carpenter, T. Powers, L.S. Vidyaratne
    JLab, Newport News, Virginia, USA
  • K.M. Iftekharuddin, M. Rahman
    ODU, Norfolk, Virginia, USA
  • A.D. Shabalina
    STFC/DL, Daresbury, Warrington, Cheshire, United Kingdom
 
  Funding: This work supported by the U.S. Department of Energy, Office of Science, Office of Nuclear Physics under Contract No. DE-AC05-06OR23177.
We re­port on the de­vel­op­ment of ma­chine learn­ing mod­els for clas­si­fy­ing C100 su­per­con­duct­ing ra­diofre­quency (SRF) cav­ity faults in the Con­tin­u­ous Elec­tron Beam Ac­cel­er­a­tor Fa­cil­ity (CEBAF) at Jef­fer­son Lab. Of the 418 SRF cav­i­ties in CEBAF, 96 are de­signed with a dig­i­tal low-level RF sys­tem con­fig­ured such that a cav­ity fault trig­gers record­ings of RF sig­nals for each of eight cav­i­ties in the cry­omod­ule. Sub­ject mat­ter ex­perts an­a­lyze the col­lected time-se­ries data and iden­tify which of the eight cav­i­ties faulted first and clas­sify the type of fault. This in­for­ma­tion is used to find trends and strate­gi­cally de­ploy mit­i­ga­tions to prob­lem­atic cry­omod­ules. How­ever, man­u­ally la­bel­ing the data is la­bo­ri­ous and time-con­sum­ing. By lever­ag­ing ma­chine learn­ing, near real-time - rather than post­mortem - iden­ti­fi­ca­tion of the of­fend­ing cav­ity and clas­si­fi­ca­tion of the fault type has been im­ple­mented. We dis­cuss the per­for­mance of the ma­chine learn­ing mod­els dur­ing a re­cent physics run. We also dis­cuss ef­forts for fur­ther in­sights into fault types through un­su­per­vised learn­ing tech­niques and pre­sent pre­lim­i­nary work on cav­ity and fault pre­dic­tion using data col­lected prior to a fail­ure event.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-FRXC01  
About • paper received ※ 16 May 2021       paper accepted ※ 01 July 2021       issue date ※ 13 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)