An improved species conserving genetic algorithm for multimodal optimization

Abstract

A new method for finding multiple solutions of multimodal optimization problems is proposed in this paper. To avoid the necessity of specifying a niche radius, the proposed method adopts hybrid of species conservation and hill-valley detection mechanism. The proposed method is compared with classical Species Conservation Genetic Algorithm (SCGA) on a number of standard benchmark problems. The experimental results show that the new approach performs better in finding all optima with no additional parameters introduced.