A differential evolution optimized fuzzy clustering algorithm with adaptive adjusting strategy

This paper presents a differential evolution optimized fuzzy clustering algorithm (DEOFCA), which combines differential evolution (DE) algorithm and fuzzy clustering theory. Since DE algorithm has strong global search ability and good robustness, DEOFCA uses DE to replace the iteration process of fuzzy C means clustering algorithm, by which the global optimization capability is greatly improved. An adaptive adjusting strategy for control parameters is integrated with the algorithm to eliminate negative effects of the control parameters setting to algorithm performance and efficiency. The proposed algorithm is applied to a case of power system, and the results demonstrate the feasibility and efficiency of this novel method.