Improved multi-objective differential evolution for maintaining population diversity
Kezong Tang, Jun Hui Wu · 2013
Diversity-preservation mechanism in a population is a crucial task in evolutionary algorithms(EAs) because it can affect the convergence speed and quality of the final solution. In this paper, an improved multi-objective differential evolution (IMODE) is proposed based on a neighboring function criterion, which maintains diversity among population members. Also, a new selection operator is adopted to enhance the selection capability of the IMODE. The performance of the IMODE is tested on a set of benchmark problems. The comparison with reported results of other technique reveals the superiority of the proposed IMODE approach and confirms its potential for solving other multi-objective optimization problems.