基于机器学习的多中子探测技术
Multi-neutron Detection Based on Machine Learning
杜泽宇 1黄思维 1杨再宏 1李奇特 1边佳伟1
作者信息
- 1. 北京大学物理学院,北京 100871;北京大学核物理与核技术国家重点实验室,北京 100871
- 折叠
摘要
中子滴线区的丰中子原子核结构是当前放射性核束物理研究的前沿热点之一.通过直接探测这些不稳定原子核衰变中发射的中子,不仅能提取核内部的多中子关联,也为丰中子核物质的性质研究提供重要的线索.为满足开展多中子探测实验的需求,本工作发展了基于机器学习的多中子识别算法,以大量的模拟数据作为训练样本,构建深度神经网络来逐事件判定反应的中子数,并进一步挑选出真实中子.本工作的结果表明,机器学习算法的四中子探测效率为～15％,传统算法为～1％,机器学习算法能将四中子探测效率显著提升10倍以上,有望应用到多中子探测实验中.
Abstract
The structure of neutron-rich nuclei in the neutron drip line region is one of the frontiers of the Radioactive Ion Beam physics.By directly detecting the neutrons emitted during their decay,the multi-neutron correlations of the nucleus can be extracted,which also provides critical information for the study of the properties of neutron-rich nuclear matter.In order to meet the requirements of conducting multi-neutron detection experiments,we developed a machine-learning-based multi-neut-ron recognition algorithm.We constructed a deep neural network to determine the number of incident neutrons event by event,and to further select the real neutron signals.The results of this work indicate that the detection efficiency of the machine learning algorithm is～15％,whereas that of the traditional algorithm is～1％.The machine learning algorithm can signific-antly improve four-neutron detection efficiency by more than 10 times,and is expected to be applied to multi-neutron detec-tion experiments.
关键词
丰中子核/多中子探测/机器学习/深度神经网络Key words
neutron-rich nuclei/multi-neutron detection/machine learning/deep neural network引用本文复制引用
基金项目
国家重点研发计划项目(2022YFA1605100)
国家重点研发计划项目(2023YFE0101500)
国家自然科学基金资助项目(12275006)
北京大学核物理与核技术国家重点实验室自主科研课题资助项目(NPT2022ZZ02)
北京大学核物理与核技术国家重点实验室自主科研开放课题(NPT2020KFY06)
State Key Laboratory of Nuclear Physics and Technology,Peking University(NPT2022ZZ02)
State Key Laboratory of Nuclear Physics and Technology,Peking University(NPT2020KFY06)
出版年
2024
NETL
NSTL
万方数据