智能算法在中子能谱解谱中的应用评述

doi:10.27046/j.1004-6356.202603005

摘要/Abstract

摘要： 本文首先阐述了中子解谱的基本原理。随后论述了智能算法在中子解谱领域的应用实践,涵盖神经网络算法与智能优化算法两大方向:前者重点梳理其演进脉络,从早期简单三层架构逐步迭代至BP神经网络、广义回归神经网络,再到深度学习技术;后者则介绍以遗传算法、粒子群算法为代表的典型方法。接着简要介绍了智能算法解谱效果的评价体系与方法。最后结合智能算法的独特优势,展望了其在中子解谱领域的未来发展方向与潜力,为智能解谱方法的深入研究与实际应用提供参考与借鉴。

关键词: 中子解谱, 智能算法, 智能优化算法, 神经网络算法

Abstract: This paper first describes the basic principles of neutron spectrum unfolding. Subsequently, the application practice of intelligent algorithms in the field of neutron spectrum unfolding was discussed, covering two major directions: neural network algorithms and intelligent optimization algorithms. The former focuses on sorting out its evolutionary trajectory, gradually iterating from the early simple three-layer architecture to BP neural networks, generalized regression neural networks, and then to deep learning techniques; The latter introduces typical methods represented by genetic algorithm and particle swarm optimization algorithm. Furthermore, a brief introduction was given to the evaluation system and methods for the spectral performance of intelligent algorithms; Finally, combining the unique advantages of intelligent algorithms, the future development direction and potential in the field of neutron spectrum unfolding were discussed. The article aims to provide reference and inspiration for the in-depth research and practical application of intelligent spectral analysis methods.This paper first introduces the basic principles of neutron spectrum unfolding. Subsequently, the application of intelligent algorithms in this field is discussed, covering two major categories: neural network algorithms and intelligent optimization algorithms. For neural networks, the evolutionary trajectory is sorted out, which has gradually iterated from early simple three-layer architectures to Back Propagation (BP) neural networks, Generalized Regression Neural Networks (GRNN), and then to deep learning techniques. For intelligent optimization algorithms, typical methods represented by Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) are introduced. Furthermore, the evaluation system and methods for the performance of intelligent algorithms are briefly summarized. Finally, based on the unique advantages of intelligent algorithms, this paper prospects the future development trends and potential in the field of neutron spectrum unfolding. This paper aims to provide references for the in-depth research and practical application of intelligent spectral analysis methods.

Key words: neutron spectrum unfolding, intelligent algorithms, intelligent optimization algorithms, neural network algorithms

中图分类号:

TP183-7

郝杰, 陈宝维, 李建伟, 周文明, 王远飞, 梁栋, 张艳婷, 马弢. 智能算法在中子能谱解谱中的应用评述[J]. 辐射防护通讯, 2026, 46(3): 27-36.

HAO Jie, CHEN Baowei, LI Jianwei, ZHOU Wenming, WANG Yuanfei, LIANG Dong, ZHANG Yanting, MA Tao. Review of the application of intelligent algorithms in neutron spectrum unfolding[J]. RADIATION PROTECTION BULLETIN, 2026, 46(3): 27-36.

参考文献

[1] 罗晓亮.中子通量及能谱测量中关键技术研究[D]. 长沙:国防科学技术大学, 2015.
[2] Koshiishi H, Matsumoto H, Chishiki A, et al. Evaluation of the neutron radiation environment inside the international space station based on the Bonner ball neutron detector experiment[J]. Radiation Measurements, 2007, 42(9): 1510-1520.
[3] Koohi-Fayegh R,Green S,Crout N M J,et al.Neural network unfolding of photon and neutron spectra using an NE-213 scintillation detector[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 1993, 329(1/2): 269-276.
[4] 孙光耀,甘佺,郝丽娟,等.基于多维响应矩阵的中子能谱逆向调控方法及应用[J]. 核科学与工程,2019,39(4): 556-561.
[5] Sarkar P K.Neutron dosimetry in the particle accelerator environment[J]. Radiation Measurements,2010, 45(10): 1476-1483.
[6] Aizawan N, Yamanaka M, Iwasaki T, et al. Effect of neutron spectrum on subcritical multiplication factor in accelerator-driven system[J]. Progress in Nuclear Energy, 2019, 116: 158-167.
[7] Klein H, Neumann S. Neutron and photon spectrometry with liquid scintillation detectors in mixed fields[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2002, 476(1/2): 132-142.
[8] Bertrand G H V, Hamel M, Normand S, et al. Pulse shape discrimination between (fast or thermal) neutrons and gamma rays with plastic scintillators: state of the art[J]. Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2015, 776: 114-128.
[9] Alharbi T. Distance metrics for digital pulse-shape discrimination of scintillator detectors[J]. Radiation Physics and Chemistry, 2019, 156: 205-209.
[10] 吴治华.原子核物理实验方法[M]. 北京: 原子能出版社, 2002.
[11] 郑春,李建胜,吴建华. CFBR-Ⅱ堆中子注量测量[J]. 核电子学与探测技术, 2001, 21(4): 307-309.
[12] 王松林,王琦,徐小三,等.阈探测器法测量Am-Be中子源屏蔽辐照腔内的中子能谱[J]. 原子能科学技术, 2009, 43(1): 16-21.
[13] Ishak-Boushaki G M, Boukeffoussa K, Idiri Z, et al. Thick activation detectors for neutron spectrometry using different unfolding methods: sensitivity analysis and dose calculation[J]. Applied Radiation and Isotopes, 2012, 70(3): 515-519.
[14] 史永谦.核反应堆中子学实验技术[M]. 北京:中国原子能出版社, 2011.
[15] Chiesa D, Nastasi M, Cazzaniga C, et al. Measurement of the neutron flux at spallation sources using multi-foil activation[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2018, 902: 14-24.
[16] Cherepy N J, Sanner R D, Beck P R, et al. Bismuth-and lithium-loaded plastic scintillators for gamma and neutron detection[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2015, 778: 126-132.
[17] Mabe A N, Glenn A M, Carman M L, et al. Transparent plastic scintillators for neutron detection based on lithium salicylate[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2016, 806: 80-86.
[18] 汲长松.中子探测实验方法[M]. 北京:中国原子能出版社, 1998.
[19] Gold R.An iterative unfolding method for response matrices: ANL-6984[R]. Illinois: Argonne National Laboratory, 1964.
[20] Griffin P J, Kelly J G, Vandenbury J W. User′s manual for SNL-SAND-Ⅱ code[R]. New Mexico: Sandia National Laboratories, 1994.
[21] 郑春,吴建华,李建胜,等.活化法测量CFBR-II堆中子注量和中子能谱[J]. 核动力工程, 2004, 25(1): 93-96.
[22] 王松林,孔祥忠,邓勇军,等.活化法测量中子活化在线分析系统样品处的中子能谱[J]. 原子核物理评论, 2009, 26(1): 27-32.
[23] Weise K, Matzke M. A priori distributions from the principle of maximum entropy for the Monte Carlo unfolding of particle energy spectra[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 1989, 280(1): 103-112.
[24] Reginatto M, Goldhagen P. MAXED, a computer code for maximum entropy deconvolution of multisphere neutron spectrometer data[J]. Health Physics, 1999, 77(5): 579-583.
[25] 周雪梅.钍基熔盐堆中子能谱测量方法的研究[D]. 上海:中国科学院研究生院(上海应用物理研究所), 2013.
[26] 曾丽娜.临界装置中子能谱测量[D]. 北京:中国工程物理研究院, 2014.
[27] 孟令杰,李达,周雪梅.活化法测量中子能谱实验中反应率及初始谱的检验[J]. 核电子学与探测技术,2015, 35(2): 163-168.
[28] 熊厚华.反应堆堆芯中子能谱在线测量方法研究[D]. 合肥:中国科学技术大学, 2018.
[29] Reginatto M.Overview of spectral unfolding techniques and uncertainty estimation[J]. Radiation Measurements, 2010, 45(10): 1323-1329.
[30] 范启蒙,吕宁,过惠平,等.基于多球中子谱仪的解谱算法研究现状[J]. 现代应用物理, 2023, 14(3): 94-100.
[31] Reginatto M.Bayesian approach for quantifying the uncertainty of neutron doses derived from spectrometric measurements[J]. Radiation Protection Dosimetry,2006,121(1): 64-69.
[32] 宋鸿鹄. Am-Be中子源聚乙烯泄露中子能谱测量技术研究[D]. 北京:中国工程物理研究院, 2017.
[33] Chiesa D,Cazzaniga C,Nastasi M,et al.Measurements of neutron fields in a wide energy range using Multi-Foil activation analysis[J]. IEEE Transactions on Nuclear Science, 2022, 69(7): 1659-1666.
[34] Bedogni R, Domingo C, Esposito A, et al. FRUIT: an operational tool for multisphere neutron spectrometry in workplaces[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2007, 580(3): 1301-1309.
[35] Dehimi F Z, Seghour A, Abaidia S E H. Unfolding of neutron energy spectra with fisher regularisation[J]. IEEE Transactions on Nuclear Science, 2010, 57(2): 768-774.
[36] XU Yunlin, Flaska M, Pozzi S, et al. A sequential least-squares algorithm for neutron spectrum unfolding from pulse height distributions measured with liquid scintillators[C]//Joint International Topical Meeting on Mathematics and Computations and Supercomputing in Nuclear Applications, M and C + SNA 2007. Monterey: American Nuclear Society, 2007: 1-12.
[37] Sharghi Ido A, Bonyadi M R, Etaati G R, et al. Unfolding the neutron spectrum of a NE213 scintillator using artificial neural networks[J]. Applied Radiation and Isotopes, 2009, 67(10): 1912-1918.
[38] Braga C C, Dias M S, et al. Application of neural networks for unfolding neutron spectra measured by means of Bonner spheres[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2002, 476(1/2): 252-255.
[39] Endo S, Yoshida E, Yoshitake Y, et al. Neural networks for the neutron spectrum determination based on the foil activation method[J]. Japanese Journal of Applied Physics, 2002, 41(4R): 2191.
[40] 郦文忠,谢涛,曹利国.新的三层BP神经网络算法在解谱分析中的应用[J]. 科学技术与工程,2009,9(4): 897-899, 904.
[41] ZHU Qingjun, SONG Gang, SONG Fengquan, et al. Singular value decomposition and artificial neutral network for analyzing Bonner sphere data[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2012, 687: 23-29.
[42] ZHU Qingjun, TIAN Lichao, YANG Xiaohu, et al. Advantages of artificial neural network in neutron spectra unfolding[J]. Chinese Physics Letters,2014,31(7): 072901.
[43] Del Rosario Martínez-Blanco M, Ornelas-Vargas G, Solís-Sánchez L O, et al. A comparison of back propagation and generalized regression neural networks performance in neutron spectrometry[J]. Applied Radiation and Isotopes,2016,117:20-26.
[44] Del Rosario Martinez-Blanco M,Ornelas-Vargas G,Castañeda-Miranda C L,et al.A neutron spectrum unfolding code based on generalized regression artificial neural networks[J]. Applied Radiation and Isotopes,2016, 117:8-14.
[45] WANG Jie, GUO Zhirong, CHEN Xianglei, et al. Neutron spectrum unfolding based on generalized regression neural networks for neutron fluence and neutron ambient dose equivalent estimations[J]. Applied Radiation and Isotopes, 2019, 154: 108856.
[46] 周宇琳, 王杰, 郭智荣, 等. 三种智能算法求解多球中子能谱效果比较[J]. 核电子学与探测技术, 2019, 39(3): 341-345.
[47] WANG Jie, ZHOU Yulin, GUO Zhirong, et al. Neutron spectrum unfolding using three artificial intelligence optimization methods[J]. Applied Radiation and Isotopes, 2019, 147: 136-143.
[48] 周宇琳, 曾乐, 孙光智, 等. GRNN算法多球中子谱仪解谱软件设计[J]. 核电子学与探测技术, 2021, 41(5): 897-900.
[49] 阿斯顿·张, 李沐, 扎卡里·C.立顿, 等. 动手学深度学习[M]. 北京: 人民邮电出版社, 2019.
[50] 陈舞辉. CFETR水冷包层中子学模块设计与实验研究[D]. 合肥:中国科学技术大学, 2020.
[51] Bouhadida M, Mazzi A, Brovchenko M, et al. Neutron spectrum unfolding using two architectures of convolutional neural networks[J]. Nuclear Engineering and Technology, 2023, 55(6): 2276-2282.
[52] 莫双荣,刘钰,幸浩洋,等.Elman神经网络在中子解谱中的应用[J]. 四川大学学报(自然科学版), 2020, 57(3): 531-534.
[53] Hosseini S A. Neutron spectrum unfolding using artificial neural network and modified least square method[J]. Radiation Physics and Chemistry, 2016, 126: 75-84.
[54] CAO Chenglong, GAN Quan, SONG Jing, et al. A two-step neutron spectrum unfolding method for fission reactors based on artificial neural network[J]. Annals of Nuclear Energy, 2020, 139: 107219.
[55] LIAO Shiwei, JIA Wenbao, CHENG Wei, et al. Neutron spectrum unfolding code based on iterative method combined with artificial neural networks for Bonner sphere spectrometer[J]. Journal of Radioanalytical and Nuclear Chemistry, 2024, 333(1): 557-562.
[56] CAO Chenglong, GAN Quan, SONG Jing, et al. An adaptive deviation-resistant neutron spectrum unfolding method based on transfer learning[J]. Nuclear Engineering and Technology, 2020, 52(11): 2452-2459.
[57] 曹承龙.反应堆宽能域中子能谱精准调控与测量方法研究[D]. 合肥:中国科学技术大学, 2022.
[58] 周宇琳,曾乐,任才,等.三探测器中子周围剂量当量/能谱监测仪的设计[J]. 原子能科学技术,2021,55(S1): 143-150.
[59] 王飞鹏.强流氘氚中子发生器中子特性参数测量方法研究[D]. 合肥:中国科学技术大学, 2022.
[60] Alvar A A, Deevband M R, Ashtiyani M. Neutron spectrum unfolding using radial basis function neural networks[J]. Applied Radiation and Isotopes, 2017, 129: 35-41.
[61] 王冬,何彬,张全虎.用遗传算法求解中子能谱[J]. 原子能科学技术, 2010, 44(10): 1270-1275.
[62] Hoang S M T, Sun G M, Tran H N, et al. Enhancing neutron spectral results based on the combination of genetic algorithm and activation method[J]. Journal of Radioanalytical and Nuclear Chemistry, 2018, 318(1): 631-639.
[63] 李玮,焦听雨,李晓博,等.基于遗传基因算法的中子能谱解谱程序研究[J]. 宇航计测技术,2020, 40(3): 65-68.
[64] Shahabinejad H, Sohrabpour M. A novel neutron energy spectrum unfolding code using particle swarm optimization[J]. Radiation Physics and Chemistry, 2017, 136: 9-16.
[65] 刘书焕,陈达,阿景烨,等. SAND-II解谱的不确定度分析[J]. 核动力工程, 2003(3): 204-206, 210.
[66] 陈晓亮,赵守智.基于广义最小二乘法原理的中子能谱解谱程序开发及验证[J]. 原子能科学技术,2015, 49(12): 2195-2200.
[67] LIU Bin, YANG Hongrun, LV Huanwen, et al. Study on unfolding method of neutron spectrum of BSS (Bonner Sphere Spectrometer) based on compressed sensing[J]. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2019, 925: 217-222.
[68] Masuda A, Matsumoto T, Takada K, et al. Neutron spectral fluence measurements using a Bonner sphere spectrometer in the development of the iBNCT accelerator-based neutron source[J]. Applied Radiation and Isotopes, 2017, 127: 47-51.