Research on EV loads clustering analysis method for source-grid-load system

Abstract

This report does some work on big data analysis of EV loads and extends automatic generation control (AGC) to automatic active power control (APC) by combining the current operating conditions of the grid and the spatial and temporal distribution characteristics of the adjustable capacity of EV loads. One year of user data from three EV charging stations provided by Chongqing is processed, cleaned, and clustered to analyze several typical EV load behaviors and the assessment of their adjustable characteristics. The data processing methods and the comparison of the K-means algorithm, the direct clustering, and the hierarchical clustering algorithm DPC clustering algorithm are also shown for this example.