Differential Evolution with Clustering-based Niching and Adaptive Mutation for Global Optimization

This paper proposes a differential evolution with clustering-based niching and adaptive mutation (CMDE) for global optimization. In the proposed algorithm, a clustering method is first employed to adaptively divide the population into niches according to the stage of evolution as well as the diversity of population. Based on the obtained niches, an adaptive mutation scheme is then devised such that encouraging high potential niches for exploitation while low potential niches for exploration, thus appropriately search the space. The performance of proposed algorithm has been evaluated on CEC'2015 test suit and compared with related methods. The results show that the proposed clustering-based niching and adaptive mutation schemes could be promising to enhance the DE for optimization.