Robust Filtering For Discrete-time Networked Systems With Random Delay And Packet Dropout

Since 1960s,since the filter theory and filter design method was founded by scholars,it has become a hot issue in the field of automatic control and has been paid more and more attention by more and more scholars.With the continuous progress of the society,the demands of communication and computer technology are constantly improving,because in order to realize remote control and sharing of resources among parts of different regions,more and more control systems are designed to be closed loop control system implemented by network,called "networked control systems?NCSs?".However,due to the introduction of the network,there are many adverse factors such as network induced delay,data packet loss,parameter uncertainty and so on,as well as the influence of external interference and external environment on some sensitive components in the actual engineering design.The problem of filtering in networked control systems is becoming more and more urgent and urgent.Therefore,it is of practical value and theoretical significance to study the robust filtering problem of networked control systems.The main contents of this paper are as follows:First,the mathematical model of the discrete networked control system with random time delay is established,and on this basis,the H_? and l₂-l_? robust filtering problem with random delay in the network environment is analyzed.Based on Lyapunov stability theory,the Bernoulli distribution sequence of 0 and 1 is used to describe the random time delay in the network environment.At the same time,the proper Lyapunov function is selected,and the linear matrix inequality is used to solve the stability problem in the networked filtering error system,and the mean square step of the error system is proposed.A sufficient condition for stability and satisfying a given H_? and l₂-l_? performance index is achieved.Secondly,based on a typical class of discrete systems with time lag,the Bernoulli probability distribution sequence is introduced to describe the random loss of the measured data caused by the network.The robust H_? and l₂-l_? performance indexes of the system are analyzed,and the full order filter which corresponds to the performance is designed.The sufficient conditions for the existence of full order filters are derived by selecting suitable Lyapunov functions and using linear matrix inequalities.At the same time,the designed filter can make the network filtering error system asymptotically stable and have the corresponding performance index.Finally,the mathematical model of the discrete CSTR system with random delay and packet loss in the network environment is established with the application background of the continuous stirred reactor,and the H_? robust filtering problem is studied.Two Bernoulli distribution sequences with independent values of 0 and 1 are used to describe the random time delay and the measurement output in the networkenvironment.At the same time,the appropriate Lyapunov function is selected,and the correlation definition and lemma are used,and the stability problem in the networked filtering error system is solved by using the linear matrix inequality method.The sufficient condition for the mean square asymptotic stability of the error system satisfies the H?given performance index and achieves the design of robust filter.