Observer-Based Fuzzy Event-Triggered LFC for Wind Power Systems Under Multiple Cyberattacks via an NN Approximation Approach

This paper presents an observer-based fuzzy event-triggered load frequency control (LFC) strategy and a neural network (NN) approximation approach for wind power systems (WPSs) facing network resource constraints and multiple cyberattacks. First, a Takagi-Sugeno (T-S) fuzzy model is developed to capture the nonlinear dynamics of the wind turbine. A novel multiple cyberattack model is then proposed, relaxing stringent assumptions about denial-of-service (DoS) and false data injection (FDI) attacks prevalent in prior studies. An observer-based dynamic event-triggered controller is subsequently designed to reduce network burden in WPSs while mitigating DoS effects. This controller incorporates an NN-based online weight adjustment algorithm to approximate and counteract the false data injected by FDI attacks. Lyapunov analysis demonstrates that the augmented error system achieves asymptotic stability with ${\mathcal {H}}_{\infty }$ performance in the absence of DoS attacks and maintains uniformly ultimate boundedness under multiple cyberattacks. Finally, a numerical example validates the effectiveness of the proposed control strategy and NN approximation approach. Note to Practitioners—Integrating WPSs with communication networks enables remote control but exposes them to cyberattack threats and resource constraints. This paper proposes a low-bandwidth, high-reliability LFC strategy to ensure frequency stability of WPSs under these conditions. Unlike existing methods, our dynamic event-triggered LFC and NN approach not only reduces communication demands but also mitigates performance degradation from cyberattacks. Additionally, the proposed cyberattack model relaxes strict boundary assumptions, enhancing the adaptability of LFC strategy to real-world attack scenarios.