Author: Carpenter, A.
Paper Title Page
TUXXPLM2 SRF Cavity Fault Classification Using Machine Learning at CEBAF 1167
  • A.D. Solopova, A. Carpenter, T. Powers, Y. Roblin, C. Tennant
    JLab, Newport News, Virginia, USA
  • K.M. Iftekharuddin, L. Vidyaratne
    ODU, Norfolk, Virginia, USA
  The Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab is the first large high power CW recirculating electron accelerator which makes use of SRF accelerating structures configured in two antiparallel linacs. Each linac consists of twenty C20/C50 cryomodules each containing eight 5-cell cavities and five C100 upgrade cryomodules each containing eight 7-cell cavities. Accurately classifying the source of cavity faults is critical for improving accelerator performance. In addition to archived signals sampled at 10 Hz, a cavity fault triggers a waveform acquisition process where 16 waveform records sampled at 5 kHz are recorded for each of the 8 cavities in the effected cryomodule. The waveform record length is sufficiently long for transient microphonic effects to be observable. Significant time is required by a subject matter expert to analyze and identify the intra-cavity signatures of imminent faults. This paper describes a path forward that utilizes machine learning for automatic fault classification. Post-training identification of the physical origins of faults are discussed, as are potential machine-trained model-free implementations of trip avoidance procedures. These methods should provide new insights into cavity fault mechanisms and facilitate intelligent optimization of cryomodule performance  
slides icon Slides TUXXPLM2 [4.404 MB]  
DOI • reference for this paper ※  
About • paper received ※ 14 May 2019       paper accepted ※ 23 May 2019       issue date ※ 21 June 2019  
Export • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)