基于共享机制的自适应混合遗传算法

Sharing Scheme-based Adaptive Hybrid Genetic Algorithm

共享机制小生境遗传算法常由于保持算法种群的多样性而减缓了全局收敛速度。针对共享机制的这个缺陷,提出了一种基于共享机制的自适应混合遗传算法。将熵的概念引入共享机制,提出了用以度量种群多样性的小生境熵的概念;构造了小生境半径和进化参数(交叉、变异概率)的自适应计算方法;设计了用于增强算法局部搜索寻优能力的扩展突变算子。最后实验表明,该算法对于解决多模态函数优化问题具有很好的全局搜索能力和较快的收敛速度,能够有效避免早熟收敛。

The sharing scheme niche genetic algorithm usually slows down its global convergence rate to keep the population＇s diversity. To solve the problem of the sharing scheme, a sharing scheme - based adaptive hybrid genetic algorithm was proposed. The algorithm introduces the entropy into the sharing scheme, and presents a concept of niche entropy to measure the population＇s diversity. Adaptive methods for calculating the niching radius and the evolutionary parameters of crossover probability and mutation probability are designed in the algorithm. And in order to enhance the algorithm＇s ability in the local searching, an operator of expansion mutation is also designed. Experi- ments show that the algorithm can solve the multimodal function optimization problems with good global search ability and fast convergence rate, and can avoid premature convergence effectively.