Digital Twin-Enhanced Adaptive Control for Robust Power System Stability and Security

The modern power grid is enhanced by increased renewable generation and cyber-physical technologies to provide decentralized green energy solutions. These communication technologies become one of the vulnerabilities to false data injection (FDI) attacks that threaded the cyber-security of power grid. The cyber-security compromise when coupled with load variations and renewable intermittent that further compromise the power system stability. To overcome this limitation, a digital adaptive control technique is proposed to detect FDI attacks and ensure stable frequency regulation. The FDI attack detection is done using the digital twin, a real time virtual model of the power grids. The anomaly detection is done by comparing predicted state using digital twin with the actual state of the system. After digital twin detection, the adaptive controller based on detected states dynamically adjust control action to encounter both cyber-attacks and sudden load variation. The efficacy of the proposed design is assessed in comparison with the conventional proportional Integral (PI) control method, through simulations using a realistic step-variable load profile and a sustained sinusoidal FDI attack. The proposed design improves FDI detection accuracy, achieves rapid frequency stabilization, and increases system resilience.