Analysis and Identification of Power Consumption Behavior of Industrial Users Based on Data Mining

In this paper, the Shanghai industrial users in some areas is studied to analyze its behavior of electricity through using data mining techniques. Based on the acquisition and integration of user profile data and electricity data, the data undergo repair and normalization processes. To enhance the K-means algorithm, the study considers two key factors: the number of clusters and the selection of initial cluster centers. The improved K-means algorithm is then applied to data classification to extract clustering characteristic curves for all user types. Subsequently, the typical electricity behavior characteristics are analyzed and compared with those obtained from the traditional K-means algorithm. Relevant indicators are introduced to evaluate the clustering performance. The results demonstrate that the improved K-means clustering algorithm effectively classifies different types of users and provides a more accurate and efficient analysis of user electricity behavior.