Author: Huang, X.
Paper Title Page
MOPAB077 Anomaly Detection in Accelerator Facilities Using Machine Learning 304
 
  • A. Das
    Stanford University, Stanford, California, USA
  • M. Borland, L. Emery, X. Huang, H. Shang, G. Shen
    ANL, Lemont, Illinois, USA
  • D.F. Ratner
    SLAC, Menlo Park, California, USA
  • R.M. Smith, G.M. Wang
    BNL, Upton, New York, USA
 
  Syn­chro­tron light sources are user fa­cil­i­ties and usu­ally run about 5000 hours per year to sup­port many beam­lines op­er­a­tions in par­al­lel. Re­li­a­bil­ity is a key pa­ra­me­ter to eval­u­ate ma­chine per­for­mance. Even many fa­cil­i­ties have achieved >95% beam re­li­a­bil­ity, there are still many hours of un­sched­uled down­time and every hour lost is a waste of op­er­a­tion costs along with a big im­pact on in­di­vid­ual sched­uled user ex­per­i­ments. Pre­ven­tive main­te­nance on sub­sys­tems and quick re­cov­ery from ma­chine trips are the basic strate­gies to achieve high re­li­a­bil­ity, which heav­ily de­pends on ex­perts’ ded­i­ca­tion. Re­cently, SLAC, APS, and NSLS-II col­lab­o­rated to de­velop ma­chine-learn­ing-based ap­proaches aim­ing to solve both sit­u­a­tions, hard­ware fail­ure pre­dic­tion and ma­chine fail­ure di­ag­no­sis to find the root sources. In this paper, we re­port our fa­cil­ity op­er­a­tion sta­tus, de­vel­op­ment progress, and plans.  
poster icon Poster MOPAB077 [1.240 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-MOPAB077  
About • paper received ※ 16 May 2021       paper accepted ※ 14 June 2021       issue date ※ 01 September 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
MOPAB213 Characterization of Linear Optics and Beam Parameters for the APS Booster with Turn-by-Turn BPM Data 703
 
  • X. Huang, H. Shang, C. Yao
    ANL, Lemont, Illinois, USA
 
  We take turn-by-turn (TBT) BPM data on the en­ergy ramp of the APS Booster, and an­a­lyze the data with the in­de­pen­dent com­po­nent analy­sis. The ex­trac­tion kicker was used to ex­cite the be­ta­tron mo­tion. The lin­ear op­tics of the ma­chine is char­ac­ter­ized with the TBT BPM data. We also an­a­lyze the de­co­her­ence pat­tern of the kicked beam, from which we are able to de­rive beam dis­tri­b­u­tion pa­ra­me­ters, such as the mo­men­tum spread.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-MOPAB213  
About • paper received ※ 13 May 2021       paper accepted ※ 11 June 2021       issue date ※ 19 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
MOPAB214 Linear Optics Measurement for the APS Ring with Turn-by-Turn BPM Data 707
 
  • X. Huang, V. Sajaev, Y.P. Sun, A. Xiao
    ANL, Lemont, Illinois, USA
 
  We mea­sure the lin­ear op­tics of the APS stor­age ring from turn-by-turn BPM data taken when the beam is ex­cited with an in­jec­tion kicker. De­co­her­ence due to chro­matic­ity and am­pli­tude-de­pen­dent de­tun­ing is ob­served and com­pared to the­o­retic pre­dic­tions. In­de­pen­dent com­po­nent analy­sis is used to an­a­lyze the data, which sep­a­rates the be­ta­tron nor­mal modes and syn­chro­tron mo­tion, de­spite con­t­a­m­i­na­tion of bad BPMs. The beta func­tions and phase ad­vances are sub­se­quently ob­tained. The method is used to study the lin­ear op­tics per­tur­ba­tion of an in­ser­tion de­vice.  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-MOPAB214  
About • paper received ※ 12 May 2021       paper accepted ※ 09 June 2021       issue date ※ 01 September 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB286 Experience with On-line Optimizers for APS Linac Front End Optimization 2151
 
  • H. Shang, M. Borland, X. Huang, Y. Sun
    ANL, Lemont, Illinois, USA
  • M. Song, Z. Zhang
    SLAC, Menlo Park, California, USA
 
  Funding: * Work supported by U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357 and BES R&D project FWP 2020-ANL-34573
While the APS linac lat­tice is set up using a model de­vel­oped with EL­E­GANT, the thermionic RF gun front end beam dy­nam­ics has been dif­fi­cult to model. One of the is­sues is that beam prop­er­ties from the thermionic gun can vary from time to time. As a re­sult, linac front end beam tun­ing is re­quired to es­tab­lish good match­ing and max­i­mize the charge trans­ported through the linac. We have been using a tra­di­tional sim­plex op­ti­mizer to find the best set­tings for the gun front end mag­nets and steer­ing mag­nets. How­ever, it takes a long time and re­quires some fair ini­tial con­di­tions. There­fore, we im­ported other on-line op­ti­miz­ers, such as ro­bust con­ju­gate di­rec­tion search (RCDS) which is a clas­sic op­ti­mizer as sim­plex, multi-ob­jec­tive par­ti­cle swarm (MOPSO), and multi-gen­er­a­tion gauss­ian process op­ti­mizer (MG-GPO) which is based on ma­chine learn­ing tech­nique. In this paper we re­port our ex­pe­ri­ence with these on-line op­ti­miz­ers for max­i­mum bunch charge trans­porta­tion ef­fi­ciency through the linac.
 
poster icon Poster TUPAB286 [2.964 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB286  
About • paper received ※ 12 May 2021       paper accepted ※ 08 July 2021       issue date ※ 29 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
THPAB082 Recent Operational Experience with Thermionic RF Guns at the APS 3959
 
  • Y. Sun, M. Borland, G.I. Fystro, X. Huang, H. Shang
    ANL, Lemont, Illinois, USA
 
  Funding: Work supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357
The elec­tron beam at the Ar­gonne Ad­vanced Pho­ton Source (APS) is gen­er­ated from an S-band thermionic RF gun. There are two lo­ca­tions at the fron­tend of the linac where thermionic RF guns are in­stalled – RG1 and RG2. Three so-called gen­er­a­tion-III guns are avail­able, two are in­stalled at RG1 and RG2, one is a spare. In re­cent years, these guns are show­ing signs of aging after over a cou­ple of decades of op­er­a­tions. RF trips started to occur, and we had to re­duce the nom­i­nal op­er­at­ing rf power to al­le­vi­ate the prob­lem. In ad­di­tion, beam gen­er­ated by RG1 suf­fers from low trans­porta­tion ef­fi­ciency from the gun to the linac, and beam tra­jec­tory is un­sta­ble which re­sults in charge in­sta­bil­i­ties. Re­cently, APS ob­tained a new type of pro­to­type gun and it was beam com­mis­sioned in the linac. In this paper, we re­port our op­er­a­tional ex­pe­ri­ence with these thermionic rf guns in­clud­ing thermionic-cath­ode beam ex­trac­tion, gun front-end op­ti­miza­tion for max­i­mum charge trans­mis­sion through the linac, linac lat­tice setup to match beam for in­jec­tion into the Par­ti­cle Ac­cu­mu­la­tor Ring (PAR) and op­ti­miza­tion for max­i­mum PAR in­jec­tion ef­fi­ciency.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB082  
About • paper received ※ 19 May 2021       paper accepted ※ 28 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 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)  
 
WEPAB305 Teeport: Break the Wall Between the Optimization Algorithms and Problems 3387
 
  • Z. Zhang, X. Huang, M. Song
    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.
Op­ti­miza­tion al­go­rithms/tech­niques such as ge­netic al­go­rithm (GA), par­ti­cle swarm op­ti­miza­tion (PSO) and Gauss­ian process (GP) have been widely used in the ac­cel­er­a­tor field to tackle com­plex de­sign/on­line op­ti­miza­tion prob­lems. How­ever, con­nect­ing the al­go­rithm with the op­ti­miza­tion prob­lem can be dif­fi­cult, some­times even un­re­al­is­tic, since the al­go­rithms and prob­lems could be im­ple­mented in dif­fer­ent lan­guages, might re­quire spe­cific re­sources, or have phys­i­cal con­straints. We in­tro­duce an op­ti­miza­tion plat­form named Teeport that is de­vel­oped to ad­dress the above issue. This real-time com­mu­ni­ca­tion (RTC) based plat­form is par­tic­u­larly de­signed to min­i­mize the ef­fort of in­te­grat­ing the al­go­rithms and prob­lems. Once in­te­grated, the users are granted a rich fea­ture set, such as mon­i­tor­ing, con­trol­ling, and bench­mark­ing. Some real-life ap­pli­ca­tions of the plat­form are also dis­cussed.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-WEPAB305  
About • paper received ※ 20 May 2021       paper accepted ※ 02 July 2021       issue date ※ 27 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)