Improved event-triggered security control for nonlinear networked stochastic systems under hybrid attacks
摘要
The problem of security controller design and synthesis for nonlinear networked stochastic system (NNSS) under hybrid attacks interference is investigated. Firstly, the NNSS is modeled as interval type-2 fuzzy networked stochastic system. Aiming at the complex network environment in networked control systems, random variables with uncertain characteristics are introduced to describe the occurrence of random hybrid attacks. To further reduce the loss of network resources, an improved adaptive event-triggered mechanism is proposed. It dynamically adjusts the triggered threshold parameters based on the following two conditions: 1) the change of the latest sampled-instant data and the previous triggered-instant data and 2) the weighting relationship between the correlation threshold of the above two types of signals and the triggered error. Then, the observer-based security controller is designed to ensure that both the stability and the expected performance index of the system are satisfied under the random hybrid attacks. Furthermore, the co-design conditions for obtaining the observer/controller gains and the event-triggered parameters are given. Finally, a nonlinear model of permanent magnet synchronous motor is established to verify the effectiveness of the proposed method in alleviating the impact of hybrid attacks and optimizing network communication channel resources.