Modified species-based differential evolution with self-adaptive radius for multi-modal optimization

In real world optimization, many problems are not only target on finding one global peak, but also multiple global/local peaks. These problems are referred as multi-modal optimization problems. Various techniques that commonly known as niching are proposed to solve multi-modal problems. Species-based differential evolution (SDE) is one of the recent algorithms that use the notion of speciation for solving multimodal optimization problems. In this paper, a modified SDE with a self-adaptive radius is proposed to overcome the difficulty of selecting the proper radius and improve the performance of SDE. The proposed algorithms was tested on a set of classical benchmark multi-modal optimization problems and compared with the original SDE and several other niching algorithms in literature. As shown in the experimental results, the proposed algorithm outperforms these algorithms on the benchmark problems.