Analysis And Design Under Imperfect Premise Matching For A Class Of T-S Model Based Networked Control Systems

This thesis deals with the problems of stability analysis,stabilization and H_?controller design under Imperfect Premise Matching for a class of T-S model based networked control systems which may has parametric uncertainty or disturbances.It consists of the following three parts.The first part is concerned with a controller design method for a class of nonlinear networked systems.A T-S fuzzy model is used to approximate the nonlinear networked control systems(NCSs),and the network-induce delays and packet dropouts are taken into account.By constructing a LyapunovKrasovskii functional,using some matrix inequalities and some slack matrices,a new criteria on the design of a fuzzy controller under Imperfect Premise Matching is derived.Simulation example has been given to illustrate the effectiveness of the proposed approach.In the second part,we investigated the H_?control problem of a class of nonlinear networked systems.A fuzzy state feedback controller under Imperfect Premise Matching is designed to stabilize the systems,and it has a prescribed H_?performance.Then by using theory of stability,Schur complements and the scaling techniques of inequalities,a sufficient condition about the problem of controller design is obtained.Finally,a method of auxiliary matrix is also used to reduce the conservatism of the design method.Our design has proven to be feasible through the numerical examples.In the third part,the stabilization problem and the H_?control problem are discussed for two class of networked control systems with linear functional structure uncertainty.Attention is focused on the designs of the T-S based controllers such that the closed-loop systems are asymptotically stable and preserves a prescribed H_?performance for the systems with disturbances.By using Lyapunov-Krasovskii functional and integral matrix inequality,two linear matrix inequalities are obtained for the two class of networked control systems without uncertainty.And a criteria on the design of systems with uncertainty are derived via some related lemmas.Finally,some numerical examples are given to illustrate effectiveness the proposed approach.