Abstract
Multimodal optimization is to find and maintain as many global and local optima of a function as possible. Niching techniques based on multi-population and clustering have been proved to be effective and efficacious for tackling the multimodal optimization problems. How to enhance the diversity of the population and improve the global search ability of the algorithm to locate more optima is a big challenge. To address this objective, a Double-Layer-Clustering Differential Evolution (DE) based on Speciation (SDLCDE) and integrated it with Self-adaptive strategy (SSDLCDE) is proposed to solve the multimodal optimization problems. The first layer clustering based on speciation is used to divide the entire population into multiple subpopulations to locate the global and local optima. Then all the species seeds from each species form a subpopulation for the global search. To test the performance of our proposed algorithms, both SSDLCDE and SDLCDE are compared with 17 state-of-art niching algorithms on 29 multimodal problems with different dimensions. The Experiment results demonstrate that both the proposed algorithms outperform or perform comparably to the 17 niching algorithms on all the test functions.