Defense strategy against cyber-physical coordinated false data injection attacks in cyber-physical power systems

With the deep integration of cyber-physical power systems (CPPS), cyber-physical coordinated false data injection attacks (FDIAs) have emerged as a new risk threatening the security of power grids. Accordingly, this paper proposes an active defense strategy for CPPS based on dynamic game theory. First, the coordinated mechanism by which attackers use false data to conceal physical line-disconnection attacks is thoroughly analyzed, and a defense-attack-operation tri-level dynamic game model involving defenders, attackers, and operators is built. To address the difficulty of solving multi-level game models, the proposed model is transformed into a two-stage optimization problem. In the first stage, the particle swarm optimization (PSO) algorithm is employed to exhaustively search all possible attack scenarios without defense constraints, obtaining the attacker’s false data injection strategy, load shedding amount, and load loss cost under different cyber-physical coordinated FDIAs. In the second stage, based on the attack set and results from the first stage, a mixed-integer linear programming (MILP) model is solved to determine the optimal defense resource allocation strategy under resource limitations. Finally, simulation results based on improved IEEE 14-bus and 39-bus systems demonstrate that the proposed method effectively identifies critical lines and important load nodes. The proposed defense resource allocation strategy significantly reduces the system’s expected load loss under resources constraints, thereby enhancing the resilience of CPPS against cyber-physical coordinated FDIAs.