为了提高Pareto解集的收敛性，平衡多目标优化的全局搜索和局部寻优的能力，提出一种混合精英策略的元胞多目标遗传算法。该算法在分析元胞种群结构的特点基础上，融入一种混合精英策略，提高算法的收敛性能。为了更好的平衡算法的全局搜索和局部寻优的能力，加入一种差分进化交叉算子。通过与同类算法在21个基准函数上对比实验，结果表明，引入混合精英策略和差分进化策略能够提高算法的性能，与其他优秀算法进行比较的结果说明，新算法有更好的收敛性和多样性。工程实例求解结果表明了算法的工程可行性。%In order to maintain better convergence of Pareto sets and to balance the global search and local optimiza-tion ability,the cellular multi-objective genetic algorithm based hybrid elite strategy ( CMOGA-HES) was introduced.The algorithm is integrated into a hybrid elitist strategy to improve the convergence performance,which is based on analyzing the cellular population structure characteristics.For better balance between exploitation and exploration,a differential evolution crossover operator is proposed.Comparing with the similar cellular genetic algorithm by testing 21 benchmark functions, CMOGA-HES can improve the algorithm performance and outperform several state-of-the-art multi-objective metaheuristics in terms of convergence and diversity.The results of engineering example showed the feasibility of the proposed algorithm.
展开▼
