Accelerator Fault Detection with Machine Learning [机器学习法检测加速器设备故障]

作者：Xiao D.-J.; Qiao Y.-S.; Chu Z.-M.^*

来源：Nuclear Electronics & Detection Technology, 2022, 42(1): 105-109.

• Accelerator Fault
• Data Acquisition System
• Fault Detection
• Machine Learning

This paper introduces a method of accelerator equipment fault detection based on machine learning. In this method, the association between device data is analyzed. A machine learning model is established by using regular accelerator operation data. The fault detection is realized by comparing the data with the model. Taking the BPM system as an example, simulation and real machine tests are carried out on BEPC II. The results show that this method can effectively detect the equipment fault for the operation of an accelerator.

• 单位
  中国科学院大学; 南京大学