电子学报2016，Vol.44Issue(6)：1472-1480,9.DOI:10.3969/j.issn.0372-2112.2016.06.031
基于动态种群多策略差分进化模型的多目标进化算法
MuIti-objective EvoIutionary AIgorithm Based on Dynamic Popu Iation Mu Iti-strategy DifferentiaI ModeIs
摘要
Abstract
According to the characteristics of differential evolution,a multi-objective evolutionary algorithm based on dynamic population multi-strategy differential models and decomposition (MOEA/D-DPMD)is proposed to solve the ex-pensive problems.The algorithm divides the population into three sub-populations and each sub-population is corresponding to a differential evolution strategy.In order to improve the performance of the algorithm,the size of sub-population is adjus-ted dynamically on the basis of a differential evolution strategy contribution.Each strategy is adopted to participate in coordi-nation during the evolution process.Through the test simulation on the LZ09 benchmarks with complicated Pareto Set (PS), MOEA/D-DPMD shows a best performance with a neighborhood size of 25 .Via the comparative analysis of different schemes of differential strategy,MOEA/D-DPMD also performs well.The experimental results indicate that MOEA/D-DPMD has a better performance in terms of convergence and diversity compared with MOEA/D and NSGA-II,which is an effective way for solving complex multi-objective optimization problems.关键词
分解机制/多策略差分进化/动态种群/多目标优化Key words
decomposition mechanism/multi-strategy differential evolution/dynamic population/multi-objective op-timization分类
信息技术与安全科学引用本文复制引用
王亚辉,吴金妹,贾晨辉..基于动态种群多策略差分进化模型的多目标进化算法[J].电子学报,2016,44(6):1472-1480,9.基金项目
国家自然科学基金 （）
