不确定网络化控制系统的鲁棒容错控制

提出了网络化控制系统(NCS)的一种建模方法.该方法采用时延估计法和在线取得时延法来获取时变时延,不仅解决了采样时差和时变时延难以估计的难点,而且大大简化了NCS的建模过程.针对NCS中传感器和执行器的失效问题,利用Lyapunov法给出了一种新的鲁棒容错控制算法.运用该算法设计的控制器可使系统在网络时延、故障和建模误差的共同干扰下依然稳定运行,而且该算法接近于普通的状态反馈设计,便于工程实现.实验结果表明,算法能使系统在传感器失效和建模误差的作用下保持鲁棒稳定性.

Robust Fault Tolerant Control of Networked Control System with Time-Varying Delays

A modeling approach of networked control system (NCS) was presented. The time-delay estimation method and online time-delay acquisition method were adopted to obtain the time-delay. Therefore, the problems caused by the sample skew and the unpredictable time-variant delay were solved. Moreover, the modeling process of NCS was simplified. In order to deal with the sensor failure and the actuator failure in NCS, based on Lyapunov method, a new robust fault tolerant control algorithm is presented. With this algorithm, the stabilization of NCS can be kept even under the condition that the network induced delay, fault, and model uncertainties are interfered together. Since the state feedback is considered, the implementation of this algorithm is easy. The experimental results indicate that with this algorithm, the NCS is still robust stable even the sensor failure and norm-bounded uncertainty occur in NCS.