摘要
本文分析了多种先进蒙特卡罗程序的CSG粒子追踪算法,对JCOGIN粒子追踪模块进行了优化。粒子定位算法采取了边界粒子定位和源粒子位置预估,能减少一定量的粒子定位计算;径迹求交算法采取了安全距离优化和布尔二叉树展开,其中安全距离优化可减少电子输运径迹求交次数,布尔二叉树展开能使布尔体求交算法的时间复杂度降为O(n)。应用4个典型算例测试了优化效果,结果表明,粒子定位算法优化对于一般问题具有一定的优化效果,安全距离优化显著提升了电子输运效率,布尔二叉树展开大幅提升了JMCT对于非规则复杂几何的计算速度。
The latest optimizations for particle tracking algorithms of JCOGIN were presented in this paper.Boundary particle locating and position estimation of source particles optimizations were adopted for particle locating algorithm,which can reduce a certain amount of particle locating calculations.Safety distance optimization and Boolean binary tree expanding method were applied in intersection algorithm.The safety distance optimization can reduce the intersection calculation frequency of electron transport,and the time complexity of Boolean solids intersection algorithm is cut down to O（n）by using the Boolean binary tree expanding method.Four typical models were used to verify the optimization effect.The optimizations of particle locating algorithm have a certain effect on general problems.The safety distance optimization significantly enhances the electron transport efficiency,and the Boolean binary tree expanding method greatly improves the calculation speed of JMCT for irregular and complex geometries.
作者
王鑫
邓力
李刚
付元光
李瑞
张宝印
WANG Xin;DENG Li;LI Gang;FU Yuanguang;LI Rui;ZHANG Baoyin(Institute of Applied Physics and Computational Mathematics,Beijing 100088,China;CAEP Software Center for High Performance Numerical Simulation,Beijing 100088,China)
出处
《原子能科学技术》
EI
CAS
CSCD
北大核心
2018年第8期1530-1536,共7页
Atomic Energy Science and Technology
基金
国家重点研发计划高性能计算重点专项资助项目(2016YFB0201303)
能源局重大专项子项资助项目(2015ZX06002008)
关键词
JCOGIN
粒子追踪
算法优化
布尔二叉树展开
JCOGIN
particle tracking
algorithm optimization
Boolean binary tree expanding
作者简介
王鑫（1985-），男，山东青州人，博士，核科学与技术专业;通信作者：张宝印，Email：zby@iapcm．ac．cn
