Radiation field reconstruction method based on non-uniform source activity distribution inversion

被引：0

作者：

Liu, Liye ^{[1
]}

Fan, Qing ^{[1
]}

Li, Hua ^{[1
]}

Li, Hui ^{[1
]}

Jin, Haijing ^{[1
]}

Zhao, Yuan ^{[1
,2
]}

机构：

[1] China Institute for Radiation Protection, Taiyuan

[2] Shanxi Key Laboratory for Radiation Safety and Protection, Taiyuan

来源：

He Jishu/Nuclear Techniques | 2025年 / 48卷 / 04期

关键词：

3D radiation field; Information Criterion; Non-uniform distribution; Radiation protection; Source activity inverse;

D O I：

10.11889/j.0253-3219.2025.hjs.48.240184

中图分类号：

学科分类号：

摘要：

[Background] The accurate reconstruction of γ radiation fields is fundamental to the digitalization of radiation protection and is a prerequisite for radiation dose assessment and visualization simulation. Traditional interpolation methods and uniform source activity inversion methods struggle to accurately reconstruct high-dose-rate gradient 3D radiation fields in scenarios with high-gradient, non-uniform activity distributions within large volume source terms inside nuclear facilities. [Purpose] This study aims to develop an inversion method for non-uniform source activity distribution and apply it to accurately reconstruct 3D gamma radiation fields of the aforementioned types. [Methods] Based on multi-objective source activity inversion and the Bayesian Information Criterion, an innovative inversion method for non-uniform source activity distribution was proposed. Then, the accuracy of radiation field reconstruction results in a pipeline simulation case obtained by using this method and ordinary Kriging interpolation method were compared under different source activity distribution conditions in various regions. Finally, the effectiveness of this method was further validated using measured data from inside a nuclear facility. [Results] Under four different activity distribution conditions in the pipeline simulation case, the proposed method achieves an ARD (average relative deviation) of less than 5% for radiation field reconstruction results in all regions, significantly outperforming the ordinary Kriging interpolation method, especially in high-dose-rate gradient areas. In the real nuclear facility scenario, the ARD between 77 reconstructed dose rate values calculated from 30 measured values and the actual measurements is only 12.69%, much lower than the result of 85.40% by ordinary Kriging interpolation. [Conclusions] The inversion of non-uniform source activity distribution achieved by introducing the Bayesian Information Criterion in this study is very suitable for gamma radiation field reconstruction under complex source term conditions. It provides advanced technical support for the digitalization and simulation of radiation protection based on dynamic data in nuclear facilities, enhancing its effectiveness in practical applications. © 2025 Science Press. All rights reserved.

引用

共 23 条

[1]

Masiello I, Herault R, Mansfeld M, Et al., Simulation-based VR training for the nuclear sector—a pilot study, Sustainability, 14, 13, (2022)

[2]

Zhao Y, Liu L, Wang C, Et al., SMARP: a 3D ALARA planning tool based on the virtual reality technology[J], Annals of the ICRP, 52, 1–2, pp. 171-176, (2023)

[3]

WU Qiang, TIAN Shujian, YANG Zihui, Et al., Design and application of training simulation system for nuclear accident emergency drill, Journal of Radiation Research and Radiation Processing, 41, 2, (2023)

[4]

XIE Weiping, ZHAO Xihuan, CHEN Hao, Et al., Radiation protection for foreign body salvage work in primary of nuclear power plant, Journal of Radiation Research and Radiation Processing, 41, 3, (2023)

[5]

LIU Li, ZUO Yinghong, NIU Shengli, Et al., Application of global variance reduction methods for the calculation of γ radiation field in a large space, Nuclear Techniques, 47, 2, (2024)

[6]

LI Hua, ZHAO Yuan, LIU Liye, Et al., Research on interpolation reconstruction and visualization of radiation dose field based on Kriging theory[J], Radiation Protection, 39, 6, pp. 475-482, (2019)

[7]

Xie X W, Cai J J, Tang Z H., The reconstruction of 3D radiation field based on sparse measurement data, Annals of Nuclear Energy, 179, (2022)

[8]

Zhu S Z, Zhuang S H, Fang S, Et al., 3D gamma radiation field reconstruction method using limited measurements for multiple radioactive sources, Annals of Nuclear Energy, 175, (2022)

[9]

West A, Tsitsimpelis I, Licata M, Et al., Use of Gaussian process regression for radiation mapping of a nuclear reactor with a mobile robot, Scientific Reports, 11, (2021)

[10]

Zhou W, Sun G M, Yang Z H, Et al., BP neural network based reconstruction method for radiation field applications, Nuclear Engineering and Design, 380, (2021)

← 1 2 3 →