Fuzzy clustering algorithm-based classification of daily electrical load patterns

As the development of the smart grid data acquisition system, many smart meters are loaded in the smart grid to acquire the power users´ real-time load data. These real-time data are very important. Because these data contain users electrical behavior features. However the smart grid has too many users and data, it is in-feasible to process each user´s data and analyze each user´s behavior. To solve this problem, this paper uses the fuzzy clustering algorithm to classify smart grid users before analyzing users´ power consumption behavior. And this paper calculates two cluster validity indexes to determine the optimal number of clusters. At last, the simulation result shows that the fuzzy clustering algorithm can play an important role for solving the smart grid users´ clustering question.