基于优育子群迁徙策略的多模态遗传算法研究

通过分析小生境遗传算法(NGA)的不足,提出了用于优化计算时多模态遗传算法的完全收敛性问题.结合传统梯度进化技术,设计了基于优育子群迁徙策略的遗传算法(MBGA),并分析了其运行机理和完全收敛特性.采用新算法与NGA进行了对比仿真实验,不仅验证了所提算法的完全收敛性结论,同时验证了其求解多模态问题的有效性、完全收敛能力及其收敛的稳定性.

Study on multi-modal genetic algorithm based on excellent sub-population migrating

By the analysis of the mechanisms of niche genetic algorithm(NGA),its faults of non-complete convergence are pointed out and the concept of complete convergence of multi-modal genetic algorithm is proposed.A new migrating-based genetic algorithm(MBGA) with strategy of excellent sub-population migrating is developed,and its operating mechanism and complete convergence are analyzed using mathematical methods.The simulation experiments for MBGA and NGA are performed and the complete convergence of MBGA is proved.It is testified that MBGA has availability on solving multi-modal optimization problems,complete convergence ability and wonderful stability of search results.