Load characteristics clustering based on an improved FCM method

It is well-known that load modeling is one of the most difficult problems in the power system of the world. It is necessary to carry out further research on the load characteristics analysis and the practical load model is constructed using clustering method to increase the precision and credibility of the power system simulation analysis. Since the output using traditional Hard C-Means (HCM) is sensitive to random selections about C initial cluster centers, this paper presents an improved Fuzzy C-Means (FCM) method on searching optimal initial cluster centers for the static load characteristics clustering of 36 220 KV substations in Hunan province power grid. The results verify the feasibility and validity of the improved FCM method for load clustering.