Opposition based Chaotic Differential Evolution algorithm for solving global optimization problems

Abstract

A modified differential evolution (DE) algorithm based on opposition based learning and chaotic sequence named Opposition based Chaotic Differential Evolution (OCDE) is proposed. The proposed OCDE algorithm is different from basic DE in two aspects. First is the generation of initial population, which follows Opposition Based Learning (OBL) rules; and the second is: dynamic adaption of scaling factor F using chaotic sequence. The numerical results obtained by OCDE when compared with the results obtained by DE and ODE (opposition based DE) algorithms on eighteen benchmark function demonstrate that the OCDE is able to find a better solution while maintaining a reasonable convergence rate.