摘要
提出一种适合于复杂函数寻优的多群体遗传算法。该方法对搜索区域进行划分，使每个子区域具有简单的函数形态，而对每个子区域安排一个子群体进行搜索。这个过程可并行进行。仿真表明该方法速度快，可同时获得多个局部极值点。
A multiple population genetic algorithms suited for the optimization of complex functions is presented. Using genetic algorithm with sharing function, the input space can be partitioned into several subspaces in which only one output extreme point is existed. One population is assigned to search optimum output in each subspace. Through simulation examples, the multiple population genetic algorithm is proved to converge to the global optimum quickly and hardly gets stuck at alocal optimum.
出处
《控制与决策》
EI
CSCD
北大核心
1998年第3期263-266,共4页
Control and Decision
