The Filtering And Control For Networked T-S Fuzzy Stochastic Systems

Networked control systems?NCSs? possess significant advantages, like lower costs, more convenient installation, and higher reliability. However, the sharing characteristic of the net-work brings new challenges to system stabilization and performance analysis. Due to the limit-ed communication capacity packet dropouts and time-delay happen randomly.In the practical industrial applications, the systems are more complex and almost cannot described by the traditional linear system models. Fuzzy model adopts a set of fuzzy rules to approximate any smooth nonlinear functions. This merit makes the T-S fuzzy model become a simple and effective tool to analyze the nonlinear systems. Researchers have been paying attentions to the performance of the robust filter or estimator in the past decades.Inspired by the aforementioned works, in this paper, we discuss the filtering design and controller design for T-S fuzzy stochastic systems under non-ideal network environment. The main results are as follows:1. The problem of non-fragile filtering for nonlinear stochastic systems subject to channel fading is investigated. The nonlinear stochastic systems are represented by the Takagi-Sugeno ?T-S? model and the channel fading is described by a random process. By using the Lyapunov method, a sufficient condition which guarantees that the filtering error system is stochastical-ly stable and satisfies the passive performance is established. Simulation results are used to illustrate the availability of the method.2. The l₂-l_? filter design for fuzzy Markov stochastic systems with sensor failures and packet dropouts is addressed. A multi-failure model is used for Markov systems to better de-scribe the sensor failures, and the packet dropouts are assumed to obey a series of Bernoulli processes which are formed into a diagonal matrix. By using Lyapunov method, sufficient con-ditions which guarantee that the system is stochastically stable and achieves the l2- l? perfor-mance is obtained. The parameters of the filter are derived by solving linear matrix inequalities. Simulation results are given to demonstrate the usefulness of the method.3. The l₂-l_? state feedback controller design for a class of discrete T-S fuzzy stochastic systems with logarithmic quantization and channel fading is studied. By using the Lyapunov method, sufficient condition which guarantees that the closed-loop system is asymptotically sta-ble and achieves the l2-l? performance is obtained. Simulation result is given to demonstrate the availability of the method.