改进的多目标规划遗传算法

An Improved Genetic Algorithm for Multiobjective Programming

本文讨论了 [1]中多目标规划遗传算法存在的缺陷 ,并提出了相应改进策略 .这些策略包括 :引进精粹策略 ,杂交限制 ,终止条件 ,个体表示改进等方面 .利用这些策略使算法能克服终止准则和小生境聚集的缺陷 ,使得算法能更快的收敛到 Pareto最优解集同时又有好有分布的

This paper discussed some problems of multiobjective programming genetic algorithm (MPGA) in [1],at the same time,gives some new improvements to MPGA.These improvements include:elitism,mating restrictions,stop criteria,individual representation.By using these technical,we can overcoming the shortcomings in algorithm stopping,niche mass and so on,on the other hand we can make the algorithm converge more quickly and achieve a well distributed and we spread non-dominated set (Pareto-optimal set).