Resilient control of grid-forming inverters: A model-reference adaptive damping approach against malicious attacks

Smart inverters are increasingly recognized as susceptible to malicious attacks across their hardware, control, and network layers due to heightened connectivity. This paper introduces a novel attack mitigation strategy termed resilient model reference adaptive damping (RMRAD) specifically designed for grid-tied grid-forming inverters (GFMIs). The proposed methodology incorporates a supplementary damping power control mechanism that utilizes adaptive full-state feedback gains that serve to minimize discrepancies between the system’s state variables and a reference model that embodies the desired dynamic behavior of the system. In addition, a method for constructing a reference model that improves resiliency is presented using the linear quadratic regulator (LQR) method. The stability of the controller is substantiated by Lyapunov theory. Comprehensive offline and real-time simulation studies demonstrate that RMRAD exhibits superior resiliency compared to conventional virtual synchronous machines (VSM) and state-feedback virtual synchronous generators (SFVSG) when subjected to false data injection (FDI) and destabilization attacks targeting sensors and controller settings even under weak grid conditions.