强激光与粒子束2025，Vol.37Issue(6)：90-97,8.DOI:10.11884/HPLPB202537.240271
CSNS残余气体电离型束流剖面测量畸变校正
Ddistortion correction of CSNS Ionization Profile Monitor measurement
摘要
Abstract
The ionization profile monitor(IPM)can provide critical beam distribution information required for real-time debugging and stable operation of high-current proton accelerators.The IPM system of the China Spallation Neutron Source(CSNS)Linac adopts a compact structural design.It collects data in ion mode and performs one-dimensional transverse beam distribution measurement through an optical imaging system.However,the honeycomb mesh structure at the electrode plate apertures blocks some ions or electrons from entering the microchannel plate,causing imaging shadows and introducing beam distribution distortion.Offline numerical algorithms must be used for correction.In this paper,partial differential equation(PDE)restoration and machine learning algorithms are used to correct the imaging shadows and beam distribution distortion caused by the honeycomb mesh of the IPM in the CSNS linac.The unsupervised machine learning method DIP(Deep Image Prior)was employed,and the corrected beam size deviates from the theoretical expectation by less than 10%,while maintaining a good signal-to-noise ratio.关键词
残余气体电离型束流剖面探测器/机器学习/图像校正Key words
ionization profile monitor/machine learning/image correction分类
核科学引用本文复制引用
刘孟宇,孙纪磊,徐智虹,杨涛,聂小军,黄蔚玲,杨仁俊,康玲,刘华昌..CSNS残余气体电离型束流剖面测量畸变校正[J].强激光与粒子束,2025,37(6):90-97,8.基金项目
广东省自然科学基金(2021A1515010269) （2021A1515010269）
国家自然科学基金项目(12275294) （12275294）
