Network-based event-triggered security control for cascade switched systems under hybrid attacks

This paper presents a resilient control framework for networked cascade systems (NCCSs) subject to transmission delays and hybrid cyber-attacks involving false data injection (FDI) and Denial-of-Service (DoS) attacks. The hybrid attack model is defined by the asynchronous operation of these two threats: DoS attacks disrupt data availability by blocking the communication channel, while FDI attacks compromise data integrity by injecting false signals during the DoS dormant periods. Accordingly, an event-triggered mechanism and a neural network-based estimator are co-designed to reduce the network transmission burden and identify and compensate for unstructured FDI attack signals. The closed-loop system is modeled as a switched system, and sufficient stability conditions under hybrid attacks are derived through a Lyapunov-based neural network approach to facilitate the design of an H_∞ controller with guaranteed robust performance. Simulations on a steam temperature cascade control system validate the effectiveness of the proposed method, demonstrating 20% faster convergence and a 67% reduction in oscillations compared with a conventional method while maintaining stability and security under hybrid attacks.