Analysis of User Power Consumption Characteristics and Behavior Portrait Based on KS-RF Algorithm
Hongbin Hu, Yuqiang Wang, Junfei Han, Yifan Zhang, Yan Qin · 2021
In order to adapt to the development trend of the interaction between supply and demand of power grid, the user’s electricity behavior is fully excavated based on the actual data collected by smart meters, and the research on the method of resident electricity portrait is carried out. In this paper, a method based on KS-RF algorithm is proposed to analyze the user’s power consumption characteristics and behavior portrait. Firstly, the cluster analysis method of user electricity consumption behavior is proposed based on k-shape clustering algorithm. Secondly, the random forest algorithm is used to extract four key feature tags from several common users’ electricity consumption features, and a multi-dimensional electricity consumption feature tag system is established. Thirdly, the cluster partition results and feature tags are applied to make comprehensive portrait and visual presentation of users’ electricity consumption. Finally, the actual electricity consumption data of residents is taken as an example to analyze the application effect.