A Two-Stage Protection Method for Detection and Mitigation of Coordinated EVSE Switching Attacks

Abstract

The new surge of interest towards mass integration of Electric Vehicles (EVs) in distribution smart grids can expose the high-voltage grid to instability conditions, for instance, through cyber threats initiated from the residential or public EV Supply Equipment (EVSE). This paper (i) investigates the impact of switching attacks on EV charging infrastructure and their impacts on the inter-area stability of the transmission grid, and (ii) proposes a two-stage detection and mitigation technique for those attacks. Initially, we demonstrate that leveraging the existing vulnerabilities in charging stations' cyberspace and the topology of the grid, an adversary can switch the injected or absorbed power of EVs with inter-area frequency and cause a blackout by destabilizing the angular speed of the grid's generators. Then, a Back Propagation Neural Network (BPNN) scheme is designed and hosted at the central management system (CMS) of a public EVSE network. Using this BPNN scheme, the switching attacks are accurately detected by analyzing the features of charging/discharging requests. Moreover, the detected attacks are mitigated by delaying or discarding the request execution. Finally, to cope with the conditions where the residential chargers are under-attack, or when the BPNN fails to provide accurate detection, a wide area H <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sup> controller is designed to keep the angular speed of the synchronous generators within the acceptable limits. The effectiveness of the proposed techniques is evaluated using two-area Kundur and 5-area Australian grids.