摘要
为了解决多模态函数优化问题中全局搜索和局部优化的矛盾，本文提出一种模仿社会分工现象的观群体遗传算法，该算法用一个群体搜索，另一个群体优化，仿真结果表明：和现有方法相比，该算法不仅不会陷入局部极小点，而且收敛速度极快，是一种多模态函数寻优的有效方法。
The ambivalence hetween global search and local optimization exists in most traditional opti-mization algorithms for multimodal function. To solve this problem .a dual population genetic algorithm simu-lating the cooperation phenomenon in human society is presented in this paper. The key idea lies on that one population is used for global search while the other is used for local optimization. From the result of simula-tion, the conclllsion can be drawn that compared with other methods,this algorithm will converge to the glob-al optimum more quickly,and hardly gets stuck at a local oplimum.
出处
《控制理论与应用》
EI
CAS
CSCD
北大核心
1997年第5期765-769,共5页
Control Theory & Applications
