A Unified Cyber Attack Detection and Mitigation Framework for an Islanded AC Microgrid
Abstract
This article proposes a novel three-step framework to accurately detect, estimate and mitigate cyber-attacks like unauthorized data manipulation and hijacking controller attacks which can jeopardize the entire frequency and voltage stability of an autonomous microgrid (MG). Step-1 proposes a novel maximum mean discrepancy (MMD)-based index to detect and locate the attacked distributed energy resources (DERs); in Step-2, an unknown input observer (UIO) is proposed to coarsely estimate the unknown false data injection attack (FDIA) parameters; and Step-3 develops a backstepping integrated sliding-mode control (BSMC) to compensate the attack by injecting reverse control input bias. The efficacy of the proposed cyber-attack detection and mitigation framework is rigorously tested under various types of cyber attacks on the modified IEEE-13 bus distribution test feeder operated in an islanded mode, modeled in RSCAD and is validated with real-time digital simulator (RTDS). The performance and superiority of the proposed detection scheme are compared with an existing method through hardware-in-the-loop (HIL) simulation control environment.
