State Estimation Of Networked Markov Jump Systems

In recent years,with the rapid development and demand in the fields of manufacturing,power systems and networked control systems.More and more scholars are devoted to the work of applying the Markov jump system theory to the networked control of the above systems.However,in the networked control systems,the quality of the signal is often greatly reduced in the process of transmission due to the phenomena of packet loss and networked delays resulted from the limited bandwidth.In this case,how to establish a reasonable channel model is crucial to achieving the desired control and filtering effect.To this end,this thesis applies the Rice fading channel model and combines the hidden Markov model(HMM)to optimize the modeling of the networked control systems,then studying the problem of robust state estimation for networked Markov jump systems with complex structures and the major aspects are as follows:1.For a class of Markov jump linear parameter varying systems,the passive state estimation problem with fading channels is studied.Firstly,a hidden Markov process is employed to describe the exchange of mode information between the system and the presented estimator.Then,by the aid of stochastic analysis theory,some conditions that make the estimation error system meet the passive performance index are obtained.Finally,a numerical simulation example is used to indicate the effectiveness of the designed method.2.For a class of Markov jump memristive neural networks,the problem of the H_? state estimation is considered.The goal of the addressed problem is to design a state estimator based on HMM such that,the estimation error system is stochastically stable and satisfies the H_? performance level.Firstly,using the Rice fading model to describe the phenomenon of fading channels.Then,using the mode-dependent Lyapunov functional method,the stability criterion and the gains of the estimator are obtained.Finally,a numerical example is given to demonstrate the availability of the proposed method.3.For a class of Markov neural network systems,the finite-time H_? estimation problem is addressed.The purpose is to design an H_? state estimator under the consideration of fading channels.Firstly,with the help of the stochastic analysis theory,the criterion for ensuring the finite-time boundedness and H_? performance level of the estimation error system is presented.Then,by using an improved matrix decoupling approach,the gains of the designed estimator are obtained.The utilizability of the estimator designed is explained by a simulation example.