Stability And Control Of Networked Control Systems

The research contents of this thesis are about stability analysis and time-delay of networked control systems(NCSs). With the rapid development of control technology and correspondence network technology, and increasingly extending of system scale, more and more control systems adopt the network structure, using networks as a media to interconnect the different components in control system and to realize information to exchange and deliver of control signal. Advantages of NCSs include low cost, high reliability, less wiring ,ease of system diagnosis and maintenance, etc. However, NCSs bring a lot of opportunities and challenges along with a great deal of benefits. The insertion of communication network in the feedback control loop complicates the application of standard results in analysis and design of the NCSs because many ideal assumption made in the traditional control theory can not be applied to NCSs directly. NCSs have some questions including the network-induced delay, data packet dropout and disorder, the way of transmission which made it more complex. This dissertation is concerned with the stability of networked control systems, and computes the network induced delay of networked control systems when it is stable.Based on the Lyapunov theory of stability, the major contributions of this dissertation are as follows in terms of LMI (linear matrix inequality ):(1) This paper is concerned with the design of robust controller for networked control systems (NCSs)with uncertainties. A new analysis method of NCSs based on a discrete model is provided by introducing some variational matrix. The guaranteed cost control controller design is obtained by the guaranteed cost control law.(2) The stability analysis for the networked control system with dependent delay and time-varying uncertainties is considered, and the condition of stability is given. An appropriate type of Lyapunov functional is used to obtain our results. Applying the theorem that we have proposed, the time delay of NCSs can be computed. Then compared with previous papers, the conclusion of this part is predominant, extensive, and less conservative.(3) The stability analysis for a class of networked control systems with multi-step delay is considered. The criteria for the asymptotical stability of networked control systems is proposed to calculate the maximum delay by using a linear matrix inequality approach.In order to show the superiority of our results, we give some numerical examples and simulation results to prove that our method is less conservative.