Optimization of neutron radiography using machine learning and genetic algorithm for boron concentration measurement

IF 1.4 3区物理与天体物理 Q3 INSTRUMENTS & INSTRUMENTATION

Nuclear Instruments & Methods in Physics Research Section A-accelerators Spectrometers Detectors and Associated Equipment Pub Date : 2025-09-17 DOI:10.1016/j.nima.2025.171030

M.H. Choopan Dastjerdi , M. Jafari , J. Mokhtari , H. Jafari

{"title":"Optimization of neutron radiography using machine learning and genetic algorithm for boron concentration measurement","authors":"M.H. Choopan Dastjerdi , M. Jafari , J. Mokhtari , H. Jafari","doi":"10.1016/j.nima.2025.171030","DOIUrl":null,"url":null,"abstract":"<div><div>In this study, measuring the neutron attenuation of standard samples with different boron concentrations has been performed using the Monte Carlo simulations and experimental tests. By plotting a linear calibration curve, a relationship between the attenuation and the boron concentration in each sample was found. This relationship was used to determine the boron concentration in the unknown samples. In this work, the neutron radiography (NR) method was used to measure the neutron flux passing through the samples. Recording the scattered neutrons from samples on the imaging screen posed challenges in this research. These scattered neutrons reduce the linearity of the calibration curve. By changing the length, width, and thickness of the samples and changing the distance between the imaging screen and the samples, more than 60 different configurations for performing NR have been investigated. To examine these configurations, each of them was simulated using the Monte Carlo method. The results obtained from these simulations have been optimized using neural network and genetic algorithm. The results of the optimization have been examined and confirmed using NR experiments. The use of optimized dimensions in the simulated NR increased the R-square of calibration curve from 0.841 to 0.986. In the experimental tests, the calibration curve obtained from NR in the optimal state had good linearity with an R-square of 0.964. The calibration curve detected the boron concentration in a test sample containing 0.75 % boron mass fraction with 6.33 % error.</div></div>","PeriodicalId":19359,"journal":{"name":"Nuclear Instruments & Methods in Physics Research Section A-accelerators Spectrometers Detectors and Associated Equipment","volume":"1082 ","pages":"Article 171030"},"PeriodicalIF":1.4000,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Nuclear Instruments & Methods in Physics Research Section A-accelerators Spectrometers Detectors and Associated Equipment","FirstCategoryId":"101","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0168900225008320","RegionNum":3,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"INSTRUMENTS & INSTRUMENTATION","Score":null,"Total":0}

引用次数: 0

Abstract

In this study, measuring the neutron attenuation of standard samples with different boron concentrations has been performed using the Monte Carlo simulations and experimental tests. By plotting a linear calibration curve, a relationship between the attenuation and the boron concentration in each sample was found. This relationship was used to determine the boron concentration in the unknown samples. In this work, the neutron radiography (NR) method was used to measure the neutron flux passing through the samples. Recording the scattered neutrons from samples on the imaging screen posed challenges in this research. These scattered neutrons reduce the linearity of the calibration curve. By changing the length, width, and thickness of the samples and changing the distance between the imaging screen and the samples, more than 60 different configurations for performing NR have been investigated. To examine these configurations, each of them was simulated using the Monte Carlo method. The results obtained from these simulations have been optimized using neural network and genetic algorithm. The results of the optimization have been examined and confirmed using NR experiments. The use of optimized dimensions in the simulated NR increased the R-square of calibration curve from 0.841 to 0.986. In the experimental tests, the calibration curve obtained from NR in the optimal state had good linearity with an R-square of 0.964. The calibration curve detected the boron concentration in a test sample containing 0.75 % boron mass fraction with 6.33 % error.

查看原文本刊更多论文

利用机器学习和遗传算法优化中子射线照相硼浓度测量

本研究采用蒙特卡罗模拟和实验测试的方法，对不同硼浓度的标准样品进行了中子衰减测量。通过绘制线性校准曲线，得到了衰减与各样品中硼浓度的关系。该关系式用于测定未知样品中的硼浓度。在这项工作中，使用中子射线照相（NR）方法来测量通过样品的中子通量。在成像屏幕上记录来自样品的散射中子是本研究的挑战。这些散射中子降低了校准曲线的线性度。通过改变样品的长度、宽度和厚度以及改变成像屏幕与样品之间的距离，研究了60多种不同的NR配置。为了检查这些配置，使用蒙特卡罗方法对它们进行了模拟。利用神经网络和遗传算法对仿真结果进行了优化。通过NR实验对优化结果进行了验证。在模拟NR中，优化尺寸的使用使校准曲线的r平方从0.841提高到0.986。在实验测试中，最优状态下NR得到的校准曲线线性良好，r平方为0.964。该校准曲线对硼质量分数为0.75%的试样中硼的浓度进行检测，误差为6.33%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Nuclear Instruments & Methods in Physics Research Section A-accelerators Spectrometers Detectors and Associated Equipment 物理-光谱学

CiteScore

3.20

自引率

21.40%

发文量

787

审稿时长

1 months

期刊介绍： Section A of Nuclear Instruments and Methods in Physics Research publishes papers on design, manufacturing and performance of scientific instruments with an emphasis on large scale facilities. This includes the development of particle accelerators, ion sources, beam transport systems and target arrangements as well as the use of secondary phenomena such as synchrotron radiation and free electron lasers. It also includes all types of instrumentation for the detection and spectrometry of radiations from high energy processes and nuclear decays, as well as instrumentation for experiments at nuclear reactors. Specialized electronics for nuclear and other types of spectrometry as well as computerization of measurements and control systems in this area also find their place in the A section. Theoretical as well as experimental papers are accepted.