Improving Differential Evolution with Ring Topology-Based Mutation Operators

Abstract

Differential evolution (DE) has been proven to be a simple and powerful evolutionary algorithm, and obtains many successful applications in scientific and engineering fields. The mutation strategy plays the key role in DE for finding global optimal solutions. In most of the DE algorithms, the base and difference vectors are randomly selected from the current population. Furthermore, both the neighborhood and direction information are not fully and simultaneously exploited in the evolutionary process of DE. In order to alleviate this drawback and enhance the performance of DE, we employ the ring topology to construct neighborhood for each individual and introduce the direction information with the neighbors into the mutation operator of DE. The proposed DE is named as ring-DE in this paper. By this way, ring-DE can utilize the neighborhood and direction information simultaneously to guide the search of DE. In order to evaluate the effectiveness of the proposed method, ring-DE is incorporated into several original DE algorithms. Experimental results clearly show that ring-DE is able to enhance the performance of the DE algorithms studied.