Author: Cartier-Michaud, T.
Paper Title Page
MOPAB344 Machine Learning Models for Breakdown Prediction in RF Cavities for Accelerators 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)  
 
TUPAB325 Data-Driven Risk Matrices for CERN’s Accelerators 2260
 
  • T. Cartier-Michaud, A. Apollonio, G.B. Blarasin, B. Todd, J.A. Uythoven
    CERN, Geneva, Switzerland
 
  Funding: Research supported by the HL-LHC project.
A risk ma­trix is a com­mon tool used in risk as­sess­ment, defin­ing risk lev­els with re­spect to the sever­ity and prob­a­bil­ity of the oc­cur­rence of an un­de­sired event. Risk lev­els can then be used for dif­fer­ent pur­poses, e.g. defin­ing sub­sys­tem re­li­a­bil­ity or per­son­nel safety re­quire­ments. Over the his­tory of the Large Hadron Col­lider (LHC), sev­eral risk ma­tri­ces have been de­fined to guide sys­tem de­sign. Ini­tially, these were fo­cused on ma­chine pro­tec­tion sys­tems, more re­cently these have also been used to pri­or­i­tize con­sol­i­da­tion ac­tiv­i­ties. A new data-dri­ven de­vel­op­ment of risk ma­tri­ces for CERN’s ac­cel­er­a­tors is pre­sented in this paper, based on data col­lected in the CERN Ac­cel­er­a­tor Fault Tracker (AFT). The data-dri­ven ap­proach im­proves the gran­u­lar­ity of the as­sess­ment, and lim­its un­cer­tainty in the risk es­ti­ma­tion, as it is based on op­er­a­tional ex­pe­ri­ence. In this paper the au­thors in­tro­duce the math­e­mat­i­cal frame­work, based on op­er­a­tional fail­ure data, and pre­sent the re­sult­ing risk ma­trix for LHC.
 
poster icon Poster TUPAB325 [0.499 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB325  
About • paper received ※ 19 May 2021       paper accepted ※ 24 June 2021       issue date ※ 17 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)  
 
TUPAB345 Availability Modeling of the Solid-State Power Amplifiers for the CERN SPS RF Upgrade 2308
 
  • L. Felsberger, A. Apollonio, T. Cartier-Michaud, E. Montesinos, J.C. Oliveira, J.A. Uythoven
    CERN, Geneva, Switzerland
 
  Funding: This project has received funding from the Euratom research and training programme 2019-2020 under grant agreement No 945077.
As part of the LHC In­jec­tor Up­grade pro­gram a com­plete over­haul of the Super Pro­ton Syn­chro­tron Ra­dio-Fre­quency (RF) sys­tem took place. New cav­i­ties have been in­stalled for which the solid-state tech­nol­ogy was cho­sen to de­liver a com­bined RF power of 2 MW from 2560 RF am­pli­fiers. This strat­egy promises high avail­abil­ity as the sys­tem con­tin­ues op­er­a­tion when some of the am­pli­fiers fail. This study quan­ti­fies the op­er­a­tional avail­abil­ity that can be achieved with this new in­stal­la­tion. The eval­u­a­tion is based on a Monte Carlo sim­u­la­tion of the sys­tem using the novel Avail­Sim4 sim­u­la­tion soft­ware. A model based on life­time es­ti­ma­tions of the RF mod­ules is com­pared against data from early op­er­a­tional ex­pe­ri­ence. Sen­si­tiv­ity analy­ses have been made, that give in­sight to the cho­sen op­er­a­tional sce­nario. With the in­creas­ing use of solid-state RF power am­pli­fiers, the find­ings of this study pro­vide a use­ful ref­er­ence for fu­ture ap­pli­ca­tion of this tech­nol­ogy in par­ti­cle ac­cel­er­a­tors.
 
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB345  
About • paper received ※ 19 May 2021       paper accepted ※ 01 July 2021       issue date ※ 19 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)