两阶段遗传算法的结构及性能分析
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
针对基本遗传算法(简称BGA)常常存在局部收敛以及收敛解精度不高等方面的不足,提出了一种改进的算法——两阶段遗传算法,给出了算法的结构及具体的实施策略,进而利用Mark-ov链理论和仿真技术分析了该算法的收敛性能,结果表明该算法具有操作简单、鲁棒性强等特点,不仅可以有效地避免寻优过程中的“早熟”现象,而且在很大程度上能提高最优解精度,适合于大规模、高精度的优化问题。
In view of poor convergence and partially converge of basic genetic algorithm(BGA),an new improved algorithm,two-stage genetic algorithm is proposed.The algorithm structure and its implementation strategies are demonstrated.Then,its convergence is analysed by using Markov chain theory and simulation technology.All results indicate that this new improved algorithm will help avoid "premature" phenomenon,and improve the precision of the optimal solution.It is applicable to large scale optimization problems that demand high accuracy.
引文
[1]SRINIVAS M,PATNAIK M.Genetic algorithm:A survey[J].IEEE Computer,1994,27(6):17-26.
    [2]FOGE D B.An introduction to simulated evolutionary optimization[J].IEEE Trans on SMC,1999,(1):3-14.
    [3]ATMAR W.Noteson the simulation of evolution[J].IEEE Trans on SMC,1994,(2):130-147.
    [4]巩敦卫,孙小燕,郭西进.一种新的优胜劣汰遗传算法[J].控制与决策,2002,11(6):908-912.
    [5]韩万林.遗传算法的改进[J].中国矿业大学学报,2001,3(1):102-105.
    [6]韩炜,廖振鹏.关于遗传算法收敛性的注记[J].地震工程与工程振动,1999,19(4):13-16.
    [7]周明,孙树栋.遗传算法的原理及应用[M].北京:国防工业出版社,1999.
    [8]KIM K,PARK K S.Ranking fuzzy numbers with index of optimism[J].Fuzzy Sets and Systems,1990,35:143-150.

版权所有:© 2023 中国地质图书馆 中国地质调查局地学文献中心