%0 Journal Article
%T 机器学习预测Cs在缓冲材料和岩石上的分配系数<br>Prediction of Cs Distribution Coefficients on Buffer Materials and Rocks by Machine Learning Methods
%A 何文科
%A 陈涛
%J Nuclear Science and Technology
%P 210-217
%@ 2332-712X
%D 2023
%I Hans Publishing
%R 10.12677/NST.2023.113022
%X 缓冲材料和天然岩石是高放废物深层地质处置库多重屏障系统的重要组成部分。核素在缓冲材料和围岩中的分配系数(K<sub>d</sub>)是处置库安全评价中的重要参数之一。本研究采用4种常见的基于树的机器学习集成算法预测Cs在缓冲材料和岩石上的K<sub>d</sub>。结果表明，RF模型的预测性能最好，其中对准确预测K<sub>d</sub>最敏感的因素是t。LSR、CEC和T对K<sub>d</sub>的影响都很小。本工作证明了机器学习在促进高放废物处置库安全评价方面的应用前景。<br />
Buffer materials and natural rocks are important components of the multiple barriers system of the deep geological repository for high-level radioactive waste (HLW). Distribution coefficient (K<sub>d</sub>) of nuclides in buffer materials and surrounding rocks plays an important role in the performance and safety assessment of HLW repository. In this study, four common tree-based machine learning integration algorithms were used to predict the K<sub>d</sub> of Cs on buffer materials and rocks. The results indicate that RF model has the best predictive performance, and t is the most sensitive factor to accurately predict K<sub>d</sub>. The effects of LSR, CEC and T on K<sub>d</sub> were small. This work demonstrates the great application prospect of machine learning in the performance and safety assessment of HLW repository.
%K 机器学习，吸附，Cs，分配系数，安全评价<br>Machine Learning
%K Adsorption
%K Cs
%K Distribution Coefficient
%K Safety Assessment
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=68458