Residential power user segmentation based on k-means clustering method in the context of big data

Abstract

With the deepening of the reform on the selling side of electricity, the selling company must strengthen the analysis on the electricity consumption of users and arrange the purchase and sale of electricity scheme scientifically so as to occupy the target selling electricity market and obtain the profit of selling electricity. Based on the big data background, the paper uses the k-means clustering method to divide the load curve of 2,498 residential power users into 5 categories. On the premise of considering the system load, the above five categories are classified into three types: peak load, partial peak load, and stable power users.