Crowding clustering genetic algorithm for multimodal function optimization

Interest in multimodal function optimization is growing rapidly since real world optimization problems often require location of multiple optima in the search space. Within evolutionary computation community, various kinds of niching techniques, such as sharing, crowding, clustering, have been introduced to solve multimodal function optimization problems. In this paper, we propose a novel genetic algorithm which combines crowding and clustering for multimodal function optimization and prove its convergence properties. Crowding clustering genetic algorithm does not need any prior knowledge of fitness landscape and overcomes genetic drift efficiently. Numerical results on standard test functions indicate that crowding clustering genetic algorithm is superior to both standard crowding and deterministic crowding algorithm in quantity, quality and precision of multi-optima search.