Author: Oonuma, T.
Paper Title Page
S17AIA01 Development of a Diagnostic System for Klystron Modulators Using a Neural Network 558
 
  • M. Mutoh, T. Oonuma, Y. Shibasaki
    Tohoku University, School of Science, Sendai, Japan
  • I. Abe, K. Nakahara
    KEK, Ibaraki, Japan
 
  The diagnostic system for klystron modulators using a neural network has been developed. Large changes in the voltage and current of the main circuit in a klystron modulator were observed just several ten milli-seconds before the modulator experienced trouble. These changes formed a peculiar pattern that depended on the parts with problems. Diagnosis was possible by means of pattern recognition. The recognition test of patterns using a neural network has shown good results. This system, which is built in a linac control system, is presently being operated so as to collect new trouble patterns and to carry out tests for practical use.  
DOI • reference for this paper ※ doi:10.18429/JACoW-ICALEPCS1991-S17AIA01  
About • Received ※ 11 November 1991 — Accepted ※ 20 November 1991 — Issued ※ 04 December 1992  
Cite • reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)