A Generalized Data Recovery Model against False Data Injection Attack in Smart Grid

Recovering the compromised data after encountering a cyberattack is of prime significance for the subsequent controls and dispatches in smart grid. However, the previous recovery methods generally require some impractical assumptions or are only applicable in specific scenarios. To address it, a generalized data recovery model is proposed. It can be promptly activated to recover the pre-attack data after receiving the attack detection signal without making other assumptions. First, the measurement data inertia phenomenon is utilized to infer the preliminary values of measurements. Then, a generalized data recovery model is proposed. To analyze the recovered performance, two weighted parts are considered in the proposed recovery model, i.e., the attacked state estimation result and the inferred measurement values. The feasibility and effectiveness of the proposed recovery scheme are verified by conducting comprehensive experiments on the IEEE 30-bus benchmark.