Data-Driven-Based Privacy-Preserving Distributed Resilient Control for Hybrid AC/DC Microgrids

Information disclosure and cyber attacks present significant challenges to the practical implementation of distributed control in microgrids (MGs). Existing works have not been well-equipped to address these two challenges simultaneously, primarily because privacy protection mechanisms increase the complexity of system modeling, which in turn makes attack detection more difficult. This article investigates the distributed privacy-preserving resilient secondary control problem for hybrid ac/dc MGs under hybrid false data injection (FDI) and denial-of-service (DoS) attacks. Specifically, a deep neural network (DNN)-based estimator is first designed to estimate the aggregated remote signals from neighboring nodes by using real-time local signals (e.g., active/reactive power) as input. Subsequently, based on the estimation of aggregated remote signals and data encryption strategy, a privacy protection resilient secondary controller is designed to mitigate the impact of cyber attacks while safeguarding the system privacy simultaneously. Finally, the effectiveness of the proposed method under hybrid attacks is confirmed through a real-time experiment in OPAL-RT.