CLGAIN-KF: A Data Recovery Method for False Data Injection Attacks in Power Systems

False Data Injection Attacks can compromise the integrity of power system measurement data, mislead state estimation, and cause operational instability. To effectively recovery attacked data, this paper proposes a data recovery method (CLGAIN-KF). The method uses Generative Adversarial Imputation Networks to reasonably estimate attacked data based on incomplete normal measurement data. The generator and discriminator in GAIN are replaced with CNN-LSTM network structures to enhance the modeling capability of spatiotemporal features. Additionally, Kalman Filter is introduced to dynamically correct the recovered data in real-time, further reducing recovery errors. Experiments conducted on the IEEE 14-bus test system demonstrate that the proposed method can effectively recover the attacked data.