论文部分内容阅读
为提高多目标进化算法的收敛性,提出一种基于空间距离的多目标进化算法.定义一种密度估计指标——树聚集距离,在考虑非支配前沿的同时,利用个体的空间距离及树聚集距离进行个体选择操作.另外,在外部种群的非支配解个数超过规定的种群规模时,用基于个体邻近距离的维护方法对其进行维护.通过6个测试问题和5个方面的测试标准,与NSGA-II和SPEA2进行比较,该算法在拥有更好收敛度的同时,保持良好的均匀性和分布广度.
In order to improve the convergence of multi-objective evolutionary algorithms, a multi-objective evolutionary algorithm based on spatial distance is proposed.A densification index-tree aggregation distance is defined, taking into account the non-dominant frontier, the spatial distance and tree aggregation Distance and individual selection.In addition, when the number of non-dominated solutions of external population exceeds the specified population size, it is maintained by the maintenance method based on individual proximity distance.Through six test questions and five aspects of test standards, Compared with NSGA-II and SPEA2, the algorithm has better convergence and maintains good uniformity and distribution.