Research On Fault Detection For Networked Control System With Random Delay Via Delta Operator

With the rapid development of communication technology and computer technology and the complexity of control system,the network control systems(NCSs)emerge as the times require.Compared to the traditional point to point control system,the NCSs use network to transmit signals,which brings the advantages of resource sharing,distributed control,reducing system wiring,easy maintenance,expansion and so on.On the one hand,the advantages of communication network make NCSs widely applied in telemedicine,intelligent transportation,intelligent building,industrial manufacturing process,etc.On the other hand,the introduction of network also brings some new problems,such as time delay,data packet loss,communication limitation.Therefore,once the NCSs failure will bring great loss,so the researches on the fault detection of the NCSs are particularly necessary.At present,most of the researches on NCSs are based on the shift operators.However,when the sampling frequency becomes larger,the system model discretized by the shift operator does not approach to the continuous system model,even the system becomes unstable.While the delta operator can avoid the above disadvantages and makes the discrete time model and continuous system model into a formulation.This thesis studies the problem of fault detection for network control systems with random time delay via delta operator.The major contribution of the thesis are as follows:Firstly,the problem of H-infinity fault detection for network control systems with random delay and packet loss is considered.The stochastic time-delay is described by a Markov process,then the conducted system is modelled as a Markov jump system using delta operator.Then the fault detection filter is set up for the system.The problem of fault detection is converted to solve parameter of fault detection filter.The sufficient conditions for the asymptotic stability of the fault detection filter are obtained by Lyapunov-Krasovskii functional stability theory and linear matrix inequality(LMI)approach.The parameters of the filter are obtained by the linear matrix inequality toolbox of Matlab.Then,on the basis of the above research,this thesis considered the fault detection for uncertain networked control system with random delay and packet losses.The random time delay is less than a sampling period and satisfies the Markov process,packet loss meets Bernoulli distribution,and uncertainty exists in the system model.The system is discretized by the delta operator,and the fault detection filter is designed for the system.By using LMI and Lyapunov-Krasovskii functional stability theory,sufficient conditions for the stochastic stability and H-infinity performance of system and fault detection filter are given.The numerical simulation shows the effectiveness of the proposed method.At last,the problem of fault detection is concerned with the network control system with random delay under the condition of limited communication.The communication constraints between sensors and controllers is dealt with the communication sequence.The stochastic short-time delay is modeled as a Markov chain,and the system is modeled as a Markov jump model.Then,a residual system is established for the system.By using the LMI approach and Lyapunov-Krasovskii functional stability theory,sufficient conditions for the stochastic stability and performance of the residual system are obtained.The effectiveness of the proposed algorithm is shown by a numerical example.