Open AccessDevelopments of control system for ion source using machine learning
Author(s) -
Yoshimitsu Morita,
Mitsuhiro Fukuda,
T. Yorita,
H. Kanda,
K. Hatanaka,
T Saitou,
H Tamura,
Yoshitaka Yasuda,
T Washio,
Y Nakashima,
M Iwasaki,
Hui Wen Koay,
Keijiro Takeda,
Takafumi Hara,
T H Chong,
Zhao Hongwei
Publication year - 2022
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2244/1/012105
Subject(s) - brightness , computer science , bayesian optimization , point (geometry) , ion beam , beam (structure) , control (management) , bayesian probability , ion source , artificial intelligence , ion , machine learning , optics , mathematics , physics , geometry , quantum mechanics
Various factors influence each other in an ion source. Therefore, when operating an ion source, it is necessary to optimize and adjust various parameters. This time, we performed an experiment to automize adjustment that maximizes the brightness of the beam using machine learning. By automatically adjusting 4 parameters, we succeeded in finding a point with a beam brightness of 4.32 × 10 -6 mA/(imm mrad) 2 in 25 steps. This shows that automatic adjustment using Bayesian optimization is feasible.