Power grid false data injection attack detection method based on S transform and LSTM

As the power system gradually moves towards a new energy Internet, large-scale sensing measurements are deployed in the power system. This provides data support for data-driven false data injection attack detection methods. In order to solve the problem of low classification and recognition accuracy of False Data Injection Attacks (FDIAs), this paper proposes a power grid FDIAs detection method based on S Transform (ST) and Long Short-Term Memory (LSTM) network. This method uses discrete ST to perform time-frequency analysis on the power grid measurement signal. By extracting time-frequency features of power measurement data, false features are highlighted. Then, the LSTM network is combined to accurately classify FDIAs. Simulation experiments show that the attack detection accuracy of this method reaches more than 95%, and the false alarm rate is less than 5%.