Solving CEC 2015 multi-modal competition problems using neighborhood based speciation differential evolution

Abstract

In this article, a recently proposed niching algorithm called neighborhood based speciation Differential Evolution (NSDE) is used to solve CEC 2015 multi-modal competition problems. Although DE algorithm is effective in solving single global optimal, the result is not acceptalbe when solving multi-optima problems. NSDE was proposed to enable DE with the ability of handling multi-modal optimization problems. In NSDE, the mutation is performed within each Euclidean neighborhood. During the evolution the population of NSDE will evolve toward the respective global/local optimum and the neighborhood mutation can maintain the multiple optima found. The performance of NSDE is compared with the original SDE. From the simulation results, we can observe that NSDE is effective in solving multi-modal optimization problems.