基于深度学习的^(252)Cf源驱动核材料浓度识别技术  

^(252)Cf-source-driven nuclear material concentration identification based on deep learning

作　　者：陈乐林 魏彪[1] 李鹏程[1] 冯鹏[1] 周密[1] Chen Lelin;Wei Biao;IA Pengcheng;Feng Peng;Zhou Mi(Key Laboratory of Opto electronics Technology and System of Ministry of Education,College of Opto Electronics Engineering,Chongqing University,Chongqing 400044,Chin)

机构地区：[1]重庆大学光电工程学院光电技术及系统教育部重点实验室,重庆400044

出　　处：《强激光与粒子束》2018年第9期111-117,共7页High Power Laser and Particle Beams

基　　金：国家自然科学青年基金项目(11605017)

摘　　要：针对核武器/核材料识别系统中核材料浓度识别的关键技术问题,采用Monte Carlo方法,通过建立^(252)Cf源驱动核材料裂变中子信号样本库,模拟分析了随探测器距离和角度及核材料浓度变化的裂变脉冲中子信号特点,基于深度学习之卷积神经网络,构建了一种^(252)Cf源驱动核材料浓度识别方法,实现了对测试样本浓度的识别,且还与BP神经网络和K最近邻方法进行了对比试验研究。结果表明,卷积神经网络算法进行核材料浓度识别,得到了高达92.05%识别准确率,不仅解决了因探测器距离和角度变化时对核材料浓度识别准确率影响的难题,而且还获得了优于BP神经网络和K最近邻算法对核材料浓度识别的认识,这为^(252)Cf源驱动核材料浓度识别提供了一种新的途径。For the problem of concentration identification of nuclear material in nuclear weapon/material identification system, we used the Monte Carlo method, established a database of neutron signal obtained by fission of nuclear material driven by ^252Cf-sourcc undcr thc condition of diffcrcnt distancc and angle of dctcctors. Bascd on thc convolutional ncural nctwork in dccp learning area, a method for ^252Cf source driven nuclear material concentration identification was constructed, thereby, the identifi cation of test samples was realizcd. Then acontrast cxperiment was conducted withthe BP neural network and K nearest neighbor method. The experimental results show that using the constructed method, a high identification rate of 92.05 % is got. The prob lcm of the accuracy of the nuclear material concentration identification was affected by the change of the distance and angle of the detector is solved, and the accuracy of this method is better than that of the BP neural network and K nearest neighbor methods. This paper provides a new idea for the ^252 Cf source driven nuclear material concentration identification.

关 键 词：核武器/核材料 裂变中子信号库 深度学习 卷积神经网络 浓度识别 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]