Fault Diagnosis and Fault-Tolerant Control of Networked Control Systems with Time Delay

This article addresses the challenges of fault diagnosis and fault-tolerant control in networked control systems, introducing an observation-based approach and a corresponding control strategy. The proposed method employs HBF neural networks for fault estimation, while integrating a control prediction generator (CPG) with a network delay compensation (NDC) mechanism. This combination allows for the prediction of future control inputs and the dynamic adjustment of network delays in real-time, thereby mitigating the adverse effects of delays. This seeks to enhance the system's performance and robustness amid time-varying delays and external disturbances. In the simulation process, state equations and output equations suitable for practical applications are designed, and the effects of actuator faults are considered. The simulation results demonstrate that the hyperbasis function (HBF) neural network can effectively estimate faults, and the CPG and NDC strategies can effectively compensate for the time-varying delay effects and keep the system running stably.