Abnormal Detection System Design of Charging Pile Based on Machine Learning

Abstract

Abstract With the exhaustion of fossil energy and people’s increasing attention to environmental protection, electric vehicles began to be popularized around the world. As an important infrastructure, the EV charging network is faced with the risk of a series of network attacks, which may cause economic losses to the power grid and car owners, and even endanger the stable operation of the power grid. In order to solve the security problem of charging piles, we designed an abnormal detection system for charging piles based on the power consumption side channel and machine learning. By collecting power consumption information of the charging control unit of charging piles, the abnormal detection system determines whether charging piles are facing attacks or not. We have verified three kinds of attacks, proving that our anomaly detection system can effectively detect attacks and protect the security and stable operation of charging piles.