Anomaly Detection and State Correction in Smart Grid Using EKF and Data Compensation Techniques

The smart grid is built on an integrated and high-speed communication network, which renders it more vulnerable to cyber-attacks. False data injection attacks (FDIAs) and sudden load changes (SLCs) can bias state estimation (SE), causing the system to make incorrect decisions or provide incorrect outputs. For this reason, this article proposes a novel outlier detection and state correction strategy. To detect outliers, we propose an extended Kalman filter (EKF) based on forecast-assisted SE (FASE) to generate estimations. By modifying the prediction step in FASE into the form of adaptively adjusting the smoothing parameters, the accuracy of the state estimation can be improved. Then, outliers are detected using the deviation of the state estimate between the EKF and the weighted least squares (WLS) compared with a detection threshold determined offline at a specific false alarm rate. In addition, to recover the measurements and correction states in time, this article proposes a data compensation strategy with a sliding window to compensate the measurements. Then, the compensated measurements are used to reconstruct the state equation and the measurement equation of the system. The IEEE 14-bus system and the IEEE 30-bus system are simulated to demonstrate the performance of the proposed algorithm in anomaly detection and state correction. Numerical examples show that our algorithm can accurately detect and correct states when anomalies are present.