多目标强度Pareto 混沌差分进化算法

提出一种多目标强度Pareto混沌差分进化算法(SPCDE).首先利用Tent映射进行种群的混沌初始化,采用一种基于均匀排挤机制的截断排挤操作和混沌替换操作进行种群的环境选择操作;然后基于一种变缩放因子的差分变异策略进行变异操作,通过计算支配关系得到变异个体;最后通过支配关系的计算和环境选择操作进行进化选择操作并得到子代个体.以上操作不仅提高了算法的收敛性能,而且保证了Pareto最优解的均匀分布性.数值实验结果表明了该算法的有效性.

Multi-objective strength Pareto chaotic differential evolution algorithm

A multi-objective strength Pareto chaotic differential evolution algorithm(SPCDE) is proposed.Firstly,the chaotic initialization based on the Tent map is adopted to initialize the population.A truncation crowding operation based on a uniform crowding mechanism and a chaotic substitution operation are introduced to the environmental selection operation of the population.Then,the differential mutation operation is operated based on a differential mutation strategy with changing scaling factor,and the mutation individuals are obtained by computing the dominance relation.Finally,the evolutionary selection operation is operated and the offspring individuals are obtained by the dominance relation computation and the environmental selection operation.The operations above mentioned not only enhance the convergence performance of the proposed algorithm,but also maintain the uniformity of the Pareto optimal solution.Numerical experiment results show the effectiveness of the proposed algorithm.