New Differential Evolution Algorithm with a Second Enhanced Mutation Operator

Abstract

To improve the efficiency of the original differential evolution algorithm, a new differential evolution algorithm was proposed. A new framework with a single population was used to improve its' exploration ability. And a second enhanced mutation operator was used to ensure the exploitation of previous knowledge about the fitness landscape. Numerical experiments with typical benchmark functions show the proposed new version of differential evolution performs better than original differential evolution algorithm. Performances compared with dynamic differential evolution and particle swarm optimization algorithm show its superiority.