Research on Community User Clustering Based on Multidimensional Power Consumption Features

In order to improve the efficiency of demand-side management in China's energy strategy, to make it easier for power supply companies to better understand the user's load conditions and power consumption, which can solve the problem of peak electricity consumption in China, this paper proposes a new method based on improved K-means algorithm for the analysis of power load characteristics. Based on the power data of community users in the smart grid to establish a multi-dimensional load feature evaluation index system, the improved K-means algorithm was used to determine the number of user categories and conduct power consumption feature cluster analysis. Simulation experiment results show that the proposed algorithm can solve the problem of selecting the number of community user categories and has a better clustering effect.