Author: Singh, R.
Paper Title Page
SAPAF02 Optimization of Heavy-Ion Synchrotrons Using Nature-Inspired Algorithms and Machine Learning 15
 
  • S. Appel, W. Geithner, S. Reimann, M. Sapinski, R. Singh, D.M. Vilsmeier
    GSI, Darmstadt, Germany
  
  The ap­pli­ca­tion of ma­chine learn­ing and na­ture-in­spired op­ti­miza­tion meth­ods, like for ex­am­ple ge­netic al­go­rithms (GA) and par­ti­cle swarm op­ti­miza­tion (PSO) can be found in var­i­ous sci­en­tific/tech­ni­cal areas. In re­cent years, those ap­proaches are find­ing ap­pli­ca­tion in ac­cel­er­a­tor physics to a greater ex­tent. In this re­port, na­ture-in­spired op­ti­miza­tion as well as the ma­chine learn­ing will be shortly in­tro­duced and their ap­pli­ca­tion to the ac­cel­er­a­tor fa­cil­ity at GSI/FAIR will be pre­sented. For the heavy-ion syn­chro­tron SIS18 at GSI, the multi-ob­jec­tive GA/PSO op­ti­miza­tion re­sulted in a sig­nif­i­cant im­prove­ment of multi-turn in­jec­tion per­for­mance and sub­se­quent trans­mis­sion for in­tense beams. An au­to­mated in­jec­tion op­ti­miza­tion with ge­netic al­go­rithms at the CRYRING@​ESR ion stor­age ring has been per­formed. The usage of ma­chine learn­ing for a beam di­ag­nos­tic ap­pli­ca­tion, where re­con­struc­tion of space-charge dis­torted beam pro­files from ion­iza­tion pro­file mon­i­tors is per­formed, will also be shown. First re­sults and the ex­pe­ri­ence gained will be pre­sented.  
slides icon Slides SAPAF02 [2.642 MB]  
DOI • reference for this paper ※ https://doi.org/10.18429/JACoW-ICAP2018-SAPAF02  
About • paper received ※ 16 October 2018       paper accepted ※ 27 January 2019       issue date ※ 26 January 2019  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)