UKF-based State Estimation for Smart Grids Under False Data Injection Attacks

The false data injection attacks (FDIAs) targeting smart grids compromise the integrity of supervisory control and data acquisition systems (SCADA), posing a significant threat to the safe operation of the smart grids. The residual threshold method is usually used to detect whether the power system is under attack. However, the attack sequence carefully constructed by the attacker has concealed characteristics and can avoid false data detection mechanisms. Therefore, based on the unscented Kalman filter (UKF) state estimation, it is judged whether the power supervisory control and data acquisition system is attacked. The standard IEEE-14 node test system is used for simulation experiments, and the results show that this method can effectively estimate FDIAs for smart grids.