Event-triggered distributed privacy-preserving security control of source-network-load-storage microgrid

Information leakage and cyber-attacks pose significant challenges to the reliable implementation of distributed secondary control in source-grid-load-storage (SGLS) microgrids. To address these problems, this paper proposes a privacy-preserving resilient secondary control strategy against hybrid cyber-attacks, including false data injection (FDI) and denial-of-service (DoS) attacks. First, a distributed event-triggered controller incorporating a state-decomposition mechanism is developed to reduce the communication burden while protecting private information. Furthermore, a deep neural network (DNN)-based estimator is designed to reconstruct the aggregated neighboring communication signals when communication is compromised, thereby mitigating the effects of cyber-attacks. Compared with conventional methods, the DNN is trained solely on data pairs consisting of locally available signals and their corresponding aggregated neighboring communication signals, thereby avoiding privacy leakage during the offline training stage. In addition, the closed-loop stability under hybrid attacks is established using Lyapunov theory while accounting for bounded estimation errors. Finally, simulation studies verify the effectiveness of the proposed method.