一类非线性网络控制系统的故障检测

针对远程被控对象为非线性模型，且具有输出诱导时延的一类网络控制系统（NCS），假设可能发生执行器故障，对系统进行故障检测：利用T—S模糊模型将对象线性化，建立了模糊观测器，并给出了观测器系统渐近稳定的条件。进一步地，又考虑到满足该观测器稳定条件的矩阵难以寻找到，因此利用模糊模型中的主导子系统思想，重新设计了观测器，并基于Lyapunov稳定性定理，推导出了整个观测器系统全局渐近稳定的充分条件。最后，通过一个仿真例子验证了方法的有效性。

Fault Detection for a Class of Nonlinear Networked Control Systems

To a class of networked control systems with nonlinear remote controlled object and output induced-delay,the fault detection is carried out when actuator faults are supposed.Using T-S fuzzy model,the object is linearized and the fuzzy observer is constructed.And the stability condition of the observer system is given as well.Furthermore,it is not straightforward to find the matrix to satisfy the stability condition of this observer,so the observer is redesigned using the concept of the dominant subsystem in fuzzy models.And on the basis of Lyapunov stability theory,the sufficient condition of globally asymptotic stability of the whole system is deduced.Finally,the result of simulation shows the effectiveness of the proposed method.