摘要
本研究旨在分析和解决小型热管冷却反应堆系统中核-热-力物理场计算的复杂多物理场耦合现象的不确定性问题。研究针对中子输运、功率温度、膨胀与热应力三个模块完成数值计算建模,进一步分别训练神经网络代理模型构建多物理场耦合分析框架,使用拉丁超立方抽样方法对几何尺寸、材料温度等关键堆芯参数进行扰动,采用Sobol敏感性分析方法、蒙特卡罗不确定性分析方法进行分析。结果显示,有效增殖系数对燃料棒直径与燃料富集度敏感,基体温度对燃料棒直径、基体几何外边界参数较为敏感,膨胀后基体边界则对堆芯高度与初始几何边界参数敏感。进一步对高敏感参数进行单因素扰动试验:在燃料棒直径扰动下,有效增殖系数预测值平均下降571pcm(1pcm=10^(-5)),标准差增加159pcm,表明其存在一定负反馈及潜在不确定性放大风险;而功率温度与结构响应输出在单一几何扰动下均值小幅增加,不确定性无显著增加。研究结果为后续热管冷却反应堆设计优化提供了理论支持。
This study aims to analyze and address the uncertainty issues arising from the complex multiphysics coupling of neutronics,thermal,and mechanical fields in a compact heat pipecooled reactor system.Numerical models for neutron transport,power-temperature distribution,thermal expansion and thermal stress response were developed,and neural network surrogate models were subsequently trained for each module to construct an integrated multiphysics coupling analysis framework.Key core parameters such as geometric dimensions and material temperatures were perturbed using the Latin Hypercube Sampling method.Sensitivity and uncertainty analyses were conducted using Sobol indices and Monte Carlo simulations.The results indicate that the effective multiplication factor is sensitive to fuel rod diameter and fuel enrichment,while the matrix temperature is sensitive to fuel rod diameter and matrix geometric outer boundary parameters.The expanded matrix boundary,in turn,is sensitive to core height and initial geometric boundary parameters.Further single-factor perturbation tests were conducted on highly sensitive parameters:under fuel rod diameter perturbation,the predicted effective multiplication factor decreased by an average of 571pcm(1pcm=10^(-5)),with the standard deviation increasing by 159pcm,indicating a certain degree of negative feedback and potential risk of uncertainty amplification.In contrast,the mean values of power-temperature and structural responses slightly increased under geometry perturbation,with no significant increase in uncertainty.These findings provide theoretical support for future design optimization of heat pipe-cooled reactors.
作者
陈依诺
张滕飞
王思成
柴翔
刘晓晶
Chen Yinuo;Zhang Tengfei;Wang Sicheng;Chai Xiang;Liu Xiaojing(Institute of Nuclear Science and Safety,Shanghai Jiao Tong University,Shanghai,200240,China;Shanghai Digital Nuclear Reactor Technology Integration Innovation Center,Shanghai,200240,China;College of Smart Energy,Shanghai Jiao Tong University,Shanghai,200240,China)
出处
《核动力工程》
北大核心
2026年第1期25-32,共8页
Nuclear Power Engineering
基金
国家自然科学基金(12175138)
上海市青年科技启明星项目
中核集团领创科研项目。
关键词
热管冷却反应堆
多物理场耦合
不确定性分析
敏感性分析
Heat pipe-cooled reactor
Multiphysics coupling
Uncertainty analysis
Sensitivity analysis
