CNN-GRU based fake data injection attack detection method for power grid

Fake data injection attack (FDIA), as a new attack method, seriously threatens the safe operation of the smart grid. It can tamper with the measurement information of the power system collected by the data acquisition and monitoring system (SCADA), thus seriously affecting the security of the power grid. Based on this, a false data injection attack detection model based on an improved neural network (CNN-GRU) is proposed. By adding an entry control unit (GRU) structure to CNN, the model realizes feature extraction of spatial and temporal characteristics of data. Finally, a comprehensive case study using the Mississippi open-source dataset is conducted to evaluate the performance of the model. The results show that this mechanism can achieve satisfactory accuracy of attack detection.