Non-intrusive identification and privacy-preserving of residential electric vehicle

Abstract

Masses of user-side smart meter data provide a foundation for utilities to acquire users’ electricity consumption patterns and promote refined energy management. Therefore, the charging load of electric vehicles can be identified non-intrusively through the smart meter data of residential users, which provides flexible resources for the demand side response of power grid enterprises and is beneficial to the power dispatching decision making. Meanwhile, revealing residential energy consumption behavior from smart meter data is treated as energy privacy leakage. Extensive data collection from smart meters has raised privacy risk, which could lead to the reduction of trust in utilities from households. Aiming at this problem, the paper proposes a non-intrusive method to extract residential electric vehicles, and on this basis, discusses the privacy protection. In this paper, DWT method is used to decompose the total power of smart meter to extract the charging load of electric vehicle, and similar charging load patterns are added to smart meter data to preserve the residential EV charging behavior privacy. Practical households with EVs are used to test the proposed method and the result shows its feasibility for the performance of privacy-preserving.