遗传算法用于MJTR堆芯装载方案优化  

MJTR Core Loading Pattern Optimization Using Genetic Algorithms

作　　者：彭凤[1] 彭钢[1] 傅蓉[1] 

机构地区：[1]中国核动力研究设计院,成都610005

出　　处：《核科学与工程》2002年第3期212-216,共5页Nuclear Science and Engineering

摘　　要：应用遗传算法 ,编制了MJTR堆芯装载方案优化计算程序 ;设计了与问题特性有关的编码与交叉算子 ,用两维六角形栅元节块扩散程序SIXTUS 2作适值计算。为提高程序的质量与效率 ,通过试算找出多个计算方法与参数的优化选择。MJTR各炉装载方案的优化计算结果表明 ,遗传算法的优化方案全都好于已被采用的专家方案。Genetic Algorithms are used in MJTR core loading pattern optimization programming. Encoding and crossover operators were designed relating the characteristics of the real problem. The two dimensional hexagonal geometric multigroup diffusion code SIXTUS 2 is used in fitness calculation. For improving the quality and efficiency of the program, many math experiments have been done to find the optimal methods and parameters. Results of the optimization calculation show that all genetic algorithms optimized loading pattern are better that expert′s loading pattern. The optimized schemata and optimized loading pattern of MJTR core are given.

关 键 词：堆芯装载方案 优化 遗传算法 MJTR 5MW池式研究堆 

分 类 号：TL351.1[核科学技术—核技术及应用] TL41