Resilient load frequency control of cyber–physical power systems with off-the-shelf redundant communication channels under FDI attacks

The ICT-dependent cyber–physical power systems are exposed to increasing cyber security threats. It has been revealed that cyber-attacks may significantly deteriorate control performances or even destroy system stability. Since the load frequency control (LFC) application plays a vital role in the frequency stability of power systems, multiple redundant communication channels have been designed for an existing LFC system in advance. This paper considers the real multi-channel feature of the power communication network and proposes a secure signal extraction method for the LFC system against false data injection (FDI) attacks. To ensure the reliability of the measurement and control signals, we adopt the median filtering (MF) algorithm to extract the intact signal from multiple redundant signal channels. Furthermore, we prove that the MF algorithm can always obtain the secure signal from multi-channel signals when the number of normal signals is greater than that of abnormal signals. Eventually, we use the benchmark IEEE 30-bus system to validate the effectiveness of the proposed method.