Author: Chang, Y.H.
Paper Title Page
THPAB311 Using Linear Regression to Model the Parameters of the Flat Wires in TLS-EPU56 4399
 
  • S.J. Huang, Y.H. Chang, T.Y. Chung
    NSRRC, Hsinchu, Taiwan
  • Y.W. Chen
    Academia Sinica, Taipei, Taiwan
  
  Al­though a the­o­ret­i­cal cal­cu­la­tion might pre­dict the set cur­rents of the flat wires, which are used to com­pen­sate the de­vi­a­tion in the Be­ta­tron tune caused by the el­lip­ti­cally po­lar­ized un­du­la­tor (EPU), those set cur­rents must still be tuned in re­al­ity. To ap­proach this re­al­ity, a strat­egy of Ma­chine Learn­ing was adopted, which in­cluded col­lect­ing real-con­di­tion data and using a lin­ear-re­gres­sion model to ad­just the pa­ra­me­ters of the flat wires. After train­ing the model, the pre­dic­tions in vari­ables tune x, tune y and beam size x were com­pared with the re­quired amount of cor­rec­tion of the EPU at var­i­ous gaps and phases. To prove the fea­si­bil­ity of this method, a test was per­formed under the real con­di­tions of ac­cel­er­a­tor Tai­wan Light Source (TLS).  
poster icon Poster THPAB311 [1.226 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-THPAB311  
About • paper received ※ 13 May 2021       paper accepted ※ 28 June 2021       issue date ※ 30 August 2021  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)