Stability Analysis And Quantization Feedback Control For Networked Control System

Networked control system(NCS)is a kind of closed-loop,real time and distributed control system,which transmits and changes the information via network.New problems,like time delay,data packet dropout,disorder and etc.,are caused by the introduction of limited bandwidth network.In our thesis,stability of the NCS existing external disturbance,time delay,data packet dropout and disorder are analyzed and the design method of the controller are proposed basing on the corresponding stability analysis results.In chapter one,the study background,purpose and significance are summarily introduced.Domestic and overseas study situation of aforementioned problems existing in the NCS is given.Different disposal methods and characteristics about those problems are summarized by analyzing the existing results.In chapter three,H_? control problem for NCS,with the existence of time delay,data packet dropout and disorder in the two communication channels and external disturbance in the controlled system,is discussed.Data packet processor(DPP)is adopted to tackle the data packet disorder.Data packet dropout is seemed as a special time delay and time delay is assumed to vary in line with Markov jump law.Stochastically stability of the closed-loop Markov jump system is analyzed.Then,a H_? output feedback controller is designed basing on the stability analysis results to ensure the stochastically stability and realize the expected H_? property index.Finally,effectiveness of the obtained results is verified by a numerical example.In chapter four,quantization feedback control problem of NCS is studied with the hypothesis that only the controller to actuator communication channel has the quantization errors,time delay,data packet and disorder.A novel classification and process method is utilized to process data packet dropout,and time delay is assumed to be changing with the Markov jump law.To keep the total errors of the system in a low level,quantization density of logarithmic quantizer is specially tackled,as quantization density varies depending on the size of time delay.Then,based on the sector bound approach andMarkov jump system theory,asymptotic stability of the closed-loop system is discussed and a quantization output feedback controller is designed to realize the asymptotic stability and keep the total errors of system in a low level in line with the proposed results.Finally,effectiveness of the proposed method is proved by a numerical example.In the end of this thesis,some deficiencies and burning questions are summarized,and research direction in the future is given.