Multi-Estimator Framework for Robust FDI Attack Detection in Smart Grids

With the increasing incorporation of information and communication technologies, power grids have transitioned into complex cyber-physical systems. While this transformation has brought numerous advantages, it has also exposed the grid to heightened cyber risks, particularly false data injection (FDI) attacks. Traditional state estimation (SE) methods, often employed for FDI detection, face challenges due to their dependence on a single estimator, making them susceptible to sophisticated attacks. To address this, we propose a novel FDI detection method that integrates multiple state estimators within a game-theoretic framework. This approach improves estimator coordination by solving for global optimal and competitive equilibrium, and enhances detection accuracy through an anomaly detector. Validation using the IEEE-14 and IEEE-118 bus systems demonstrates the method’s increased reliability and robustness in real-time grid monitoring.