Aperiodic Sampled-Data Control For T-S Fuzzy Markov Jump Systems

As a kind of important stochastic hybrid system,T-S fuzzy markov jump systems(FMJSs)play a very important role in describing different types of systems,which has attracted wide attention in industry and academia.This paper will focus on several types of T-S fuzzy markov jump systems,and on the basis of dissipation,deeply study their aperiodic sampled-data control,non-fragile control and many other problems.The specific contents mainly include:1.For a class of T-S fuzzy markov jump systems,the problems of dissipative analysis and non-fragile aperiodic sampled-data controller design are discussed.Firstly,by constructing a Lyapunov function consisting of a two-sided closed-loop function,the stochastic stability and strictly(7)Q,S,R(8)-?-dissipative conditions are provided in the form of linear matrix inequalities(LMIs).Secondly,according to the strict dissipative condition,the design methods of non-fragile state feedback controllers satisfying additive uncertainty and multiplicative uncertainty are given respectively.Finally,the effectiveness of the method is checked by a truck-trailer simulation example.2.For a class of T-S fuzzy markov jump systems with actuator saturation,the aperiodic sampled-data control problem is investigated.In the first place,by constructing a mode-dependent Lyapunov function satisfies the form of two-sided closed-loop function,the stochastic stability condition is given in the form of LMIs.Secondly,built on the stochastic stability condition,the mode-dependent aperiodic sampled-data controllers are developed.At the same time,stability criteria and controller design scheme are given for aperiodic sampled-data linear systems with actuator saturation.In order to estimate the range of attraction,an improved linear matrix inequality is presented.And last,numerical examples are offered to illustrate the effectiveness of the method.3.For a class of T-S fuzzy semi-markov jump systems(FSMJSs)with actuator saturation,the aperiodic sampled-data control problem is investigated.First,by constructing a mode-dependent Lyapunov function satisfies the form of two-sided closed-loop function,the stochastic stability condition is given in the form of LMIs.Then,built on the stochastic stability condition,the mode-dependent aperiodic sampled-data controllers are developed.At the same time,stability criteria and controller design scheme are given for T-S fuzzy semi-markov jump systems without actuator saturation.The improved LMI-based optimization conditions are obtained to estimate the region of attraction.Finally,the single-link robot arm and other simulation examples are offered to illustrate the effectiveness of the method.