Ensemble of Clearing Differential Evolution for Multi-modal Optimization

Abstract

Multi-modal Optimization refers to finding multiple global and local optima of a function in one single run, so that the user can have a better knowledge about different optimal solutions. Multiple global/local peaks generate extra difficulties for the optimization algorithms. Many niching techniques have been developed in literature to tackle multi-modal optimization problems. Clearing is one of the simplest and most effective methods in solving multi-modal optimization problems. In this work, an Ensemble of Clearing Differential Evolution (ECLDE) algorithm is proposed to handle multi-modal problems. In this algorithm, the population is evenly divided into 3 subpopulations and each of the subpopulations is assigned a set of niching parameters (clearing radius). The algorithms is tested on 12 benchmark multi-modal optimization problems and compared with the Clearing Differential Evolution (CLDE) with single clearing radius as well as a number of commonly used niching algorithms. As shown in the experimental results, the proposed algorithm is able to generate satisfactory performance over the benchmark functions.