User Charging Mode Extraction Model Based on Homomorphic Encryption and Clustering of Charging Time Data from Multiple Charging Stations

To solve the problem of inaccurate user charging mode extraction using single charging station data, a user charging mode extraction model based on clustering of charging time data from multiple charging stations is proposed. Firstly, the method of using charging time data instead of charging power data for charging mode clustering is proposed to solve the problem that the Euclidean distance or Manhattan distance cannot measure the temporal proximity of charging activities in the case of staggered charging power curves. Secondly, the charging mode fluctuation distance is proposed to reflect the representativeness of the charging mode to the charging data samples, based on which the method of defining the user charging mode is proposed. Next, a user charging mode extraction method based on homomorphic encryption algorithm is proposed. Finally, the arithmetic example verifies the effectiveness of the method of this paper.