Event-triggered Control And Fault Diagnosis Of Fuzzy Markov Jump Systems

With the rapid development of intelligent communication and computing technologies,a class of systems that use the network as a carrier to achieve control purposes arises at the historical moment.These systems are widely used in various fields,including manufacturing systems,economic systems,smart grids,and aerospace.Due to random disturbances and environmental uncertainty,the structure and parameters of practical systems often change suddenly,making it difficult to model them quantitatively.Markov jump systems can describe this kind of hybrid system very well,thereby receiving considerable attention from researchers.Since the introduction of the network and some technical limitations,Markov jump systems are often constrained by asynchronous mode,limited network resources,delay,and network security,resulting in difficulties in designing control and fault diagnosis methods.It should be noted that most of the existing works on the control and fault diagnosis methods of complex Markov jump systems are based on the periodic sampling transmission mechanism.This approach wastes limited network resources and causes communication congestion when network resources or bandwidth are limited.Additionally,the modeling of nonlinear systems is mostly based on the type-1fuzzy model,which is challenging in dealing with uncertainties.Therefore,this dissertation mainly focuses on studying the problems of event-triggered control and fault diagnosis for interval type-2 fuzzy Markov jump systems,including the following aspects:Focusing on the state feedback control problem of Markov jump systems with sensor saturation and actuator nonlinearity constraints,the characteristics of the two constraints are described using saturation functions and random variables.Based on interval type-2 fuzzy theory,the interval type-2 fuzzy model of the plant is established.Considering the limited network bandwidth,a dissipative control method based on event-triggered is proposed.The proposed method can not only ensure the stability of the system but also reduce communication consumption and avoid continuous communication.Focusing on the control problem of Markov jump systems with random deception attack,the characteristics of deception attack is captured by using Bernoulli random variable and energy function.An interval type-2 asynchronous fuzzy controller is designed,combining fuzzy theory and the hidden Markov model.Based on the static triggering mechanism,dynamic triggering control is proposed.The stability conditions of the closedloop system are provided by utilizing Lyapunov and dissipative theories.The proposed dynamic triggering control method can ensure the random stability of the system while satisfying the dissipative performance and can save limited network resources.For the output feedback control problem of Markov jump systems with partially unknown transition probabilities,the interval type-2 fuzzy model is established to capture the uncertainty of the system.Then,an asynchronous output feedback controller is designed,in which the hidden Markov model is used to observe the mode of the system,and the conditional transition probabilities do not need to be completely known.Furthermore,a dynamic triggering output feedback control method is proposed.Finally,the feasibility of the method is verified by numerical simulation.Aiming at the dissipative tracking control problem of Markov jump systems subject to limited communication and partially unknown transition probabilities,two dynamic event-triggered mechanisms are designed to save the network resources of sensorto-controller and controller-to-actuator channels,respectively.Considering the partial unknowns in the transition probability matrix and conditional probability matrix,an interval type-2 fuzzy asynchronous dissipative tracking control method is proposed,which ensures that the closed-loop system is stochastically stable and exhibits dissipative performance.Aiming at the fault diagnosis problem of nonlinear Markov jump systems with communication constraints,the uncertainty is dealt with by the lower and upper membership function,and an interval type-2 fuzzy model of the plant is established.After that,an asynchronous fault diagnosis filter is designed,in which the modes of the fault diagnosis filter can be obtained by using hidden Markov model to observe the modes of the plant.Notice that the transition probability matrix information of both the system and the fault diagnosis filter does not need to be completely known.By using the fuzzy basis and mode dependence method and matrix inequality scaling technique,an interval type-2 fuzzy fault diagnosis method based on a dynamic triggering mechanism is proposed,which can not only detect faults effectively but also reduce the consumption of network resources.