Joint Interval Estimation Of State And Fault For Networked Control Systems

The introduction of an open and shared communication network in the networked control systems brings the systems the advantages of simple connection,easy expansion,and easy maintenance.It also brings a series of new problems to the system,such as data transmission delay,data packet loss and network attacks and so on,resulting in a decrease in the performance of the system.Fault diagnosis and fault-tolerant control technology provide important theoretical support for improving the safety and reliability of the system,and fault estimation provides the required fault amplitude and variation law for designing fault-tolerant control strategies.Traditional fault point estimation methods usually need to assume that interference and noise satisfy a specific probability distribution.However,in practical systems,interference and noise may be bounded but it is difficult to describe their statistical characteristics.Under this background,this topic mainly studies the joint interval estimation of the state and fault of the networked control system.The main research contents are as follows:Aiming at the networked control system where the sensor channel is under stealthy attacks,an interval observer capable of joint interval estimation of system state and fault is proposed and designed.First,an interval observer is designed to obtain the interval estimation of the covert attack signal,then the actuator failure is regarded as an augmented state,and an augmented system equivalent to the attacked system is constructed.Based on the augmented system and the interval estimation of stealthy attack signal,a TNL interval observer is designed.The parameter matrix of interval observer is obtained by using Lyapunov stability theory and Linear Matrix Inequality(LMI)method,so that the error system satisfies the given L_∞performance index.The TNL-type interval observer increases the design freedom of the observer,and all the parameters to be designed can be converted into LMI for solving,so that the estimation performance of the observer can be guaranteed.The results obtained by this observer provide information such as fault amplitude and system state for the subsequent design of fault-tolerant controllers.Aiming at the networked control system in which the sensor channel and the actuator channel are simultaneously attacked by false data injection,a set-membership estimation method that can jointly estimate the system state and fault is proposed.By considering the sensor failure as an augmented state,an augmented state observer is designed on this basis,and the parameters of the observer are obtained by using Lyapunov stability theory and LMI method,and then the interval estimation of the error signal is obtained by using the zonotpoes.The point estimate obtained by the augmented state observer and the interval estimate of the error signal obtain the interval estimate of the augmented state,thereby obtaining the interval estimate of the state and the fault.By introducing H_∞optimization technology,this method makes the obtained interval box estimation result more accurate,and the obtained fault and state information can be used for subsequent implementation of active fault-tolerant control.Aiming at the joint interval estimation problem of networked control system state and fault under unknown measurement delay,considering unknown but bounded process disturbance and unknown but bounded measurement delay.The application of TNL interval observer in the networked control system of unknown measurement delay is explored.observer.Firstly,the explicit mathematical expression of the unknown measurement delay is obtained by modeling and analysis,and it is assumed that it is unknown but bounded.Then,an augmented system with an augmented state including actuator faults and system states is constructed,and an interval observer is designed for the augmented system.Through the Lyapunov function and the LMI method,the parameter matrix of the interval observer is obtained,so that the error system satisfies the given H_∞performance index.The fault and status information obtained by the observer lays the foundation for the subsequent implementation of active fault tolerance strategies.