实验技术与管理2025，Vol.42Issue(3)：1-8,8.DOI:10.16791/j.cnki.sjg.2025.03.001
基于宇宙线缪子透射成像的建筑材料甄别方法初步研究
Preliminary study on the building material discrimination method based on cosmic ray muography
摘要
Abstract
[Objective]This study introduces a new method for material discrimination in cosmic ray muon transmission imaging,addressing a current limitation in the field that primarily focuses on identifying void volumes or density anomalies within large architectures.This method will be more crucial for the structural inspection and protection of large architectures and will broaden the applications and analytical modes of muon imaging technology.[Methods]The proposed material discrimination method integrates Monte Carlo simulation and Bayesian inference to achieve three-dimensional reconstruction of the structures and materials of large architectures.First,the researchers summarized the basic principles of cosmic ray muon transmission imaging into forward and inverse models.The forward model uses measured or simulated muon counts from various directions to derive the minimum energy matrix.By combining this minimum energy matrix and the track length matrix of voxels,the minimum stopping power matrix representing material information can be calculated in the inverse model using the algebraic reconstruction technique algorithm.Furthermore,a material identification process is established through Monte Carlo simulation and Bayesian inference,utilizing the minimum stopping power matrix and the prior hypothesis of structural materials.This process yields posterior probabilities corresponding to different prior hypotheses and quantitatively analyses the most probable architecture materials using the statistical testing quantity ptest.To validate the feasibility of the new method,an architecture composed of rock,wood,and air was constructed within an ideal simulation experiment.The simulated cosmic ray muons were generated using Gaisser's formula,and their transportation and detection were simulated using an ideal detector.[Results]The application of the forward and inverse models on the simulated data effectively reconstructed the intended geometric structure and enabled basic identification of the materials(such as rock and wood).For the quantitative analysis of the material composition,we executed the material identification process and computed the distribution of posterior probabilities under four prior hypotheses and the posterior probability corresponding to the ideal simulation experiment.From the perspective of posterior probability and the statistical test quantity ptest,the analysis result exhibits the highest posterior probability and no obvious contradiction when the prior hypothesis is consistent with the ideal experiment.This finding shows that the proposed method effectively distinguishes between different materials,especially softwood and hardwood with similar densities and stopping powers,and thereby provides preliminary evidence that the method has the feasibility to differentiate between materials.[Conclusions]In summary,the material discrimination method combining Monte Carlo simulation and Bayesian inference presented in this study fills a critical gap in material identification within muon transmission imaging technology.In the future,with the combination of efficient and suitable cosmic ray muon detection techniques,this method is poised to develop a new imaging mode with material discrimination and density differentiation,thereby providing a novel methodology for the structural inspection of large architectures.Furthermore,this study will encourage broader applications and deeper investigations of muon transmission imaging in various fields,including volcanology,archaeology,geology,and mineral exploration,and will improve the utility and applicability of muon imaging technology.关键词
宇宙线缪子/缪子透射成像/材料识别/蒙特卡洛抽样/贝叶斯推断Key words
cosmic ray muon/muography/material discrimination/Monte Carlo sampling method/Bayesian inference分类
数理科学引用本文复制引用
徐仲行,代文翰,曾志,程建平..基于宇宙线缪子透射成像的建筑材料甄别方法初步研究[J].实验技术与管理,2025,42(3):1-8,8.基金项目
北京市自然科学基金资助项目(L242004) （L242004）
