Research On Analysis And Control Method For Networked Control Systems Based On Input Delay Approach

With the rapid development of information technology, the study of networked control systems has become an important research direction of control fields. Networked control systems are networked distributed feedback control systems which congregate the communication network and the control systems,the systems are mainly composed of sensors, controllers and actuators, among which the signals are transmitted through the communication network. Obviously,the introduction of communication network changes the work mode of traditional point to point control systems and brings many convenience such as low cost,easy installation and maintenance, high flexibility and reliability, and has been applied to power system, automobile industry, aerospace system and traffic system of machinery manufacturing industry fields. However, some new problems such as networked-induced delay, data packet dropouts and disordering may degrade the performance of the systems, and bring new challenges to system analysis and control.In this paper, research on stability analysis and control method for networked control systems is carried out by considering the impacts of networked-induced delay, data packet dropouts and disordering on the systems,and the conservatism is reduced. The main research work of this paper are as follows:1. Stability analysis and stabilization of linear networked control systems A class of linear networked control systems with time delay, data packet dropouts and disordering is modeled as a linear continuous system with interval time-varying delay based on input delay approach. In view of the characteristics of integrated time delay which describes time delay, data packet dropouts anddisordering simultaneously, an improved stability criterion and controller design method is derived by using delay patition method and Jensen inequlity. Then system modeling for networked control systems with nonuniform distribution delay is studied, and a stochastic system model is obtained, and a mean square stability criterion is derived, then the controller design method is given.2. Robust guaranteed cost control for networked control systems with uncertainties On the basis of the first chapter, robust garanteed cost control for networked control systems with norm bounded uncertainties is studied. A new Lyapunov-Krasovskii functional, which is different from that of the first chapter,is constructed by introducing the time delay central point method in which the information of the upper and lower bounds of the time delay is fully utilized. The By using Jensen inequality and the convexity of the matrix function, guaranteed cost stability criterion is obtained, and the design of guaranteed cost controller is given. The validity of the proposed method is demonstrated by numerical examples.3. H∞ control for networked control systems with external perturbation Considering the effects of external perturbation on the systems, the networked control systems are modled as a linear continuous system with interval time-varying delay. Sufficient conditions for the existence of a robust H∞controller is derived by using delay patition method, the time delay central point method, Jensen inequality and the convexity of the matrix function, then a sub-optimal robust H∞ controller is derived based on linearization method. On the basis of above study, H∞ control problem for networked control systems with unmeasurable states is investigate, a sufficient condition for the existence of a robust observer-based H∞ controller is derived, then a co-design method for the observer and the controller is desigened. Numerical examples are given to further show the effectiveness and the less conservative of the proposed method.4. Analysis and stabilization of nonlinear Lurie networked control systems The problems of stability analysis and state feedback control for nonlinear Lurie networked control systems are studied by modeling based on input delay method. Through using Wirtinger inequlity instead of Jensen inequality to deal with cross terms, a sufficient condition is derived, which guarantees that nonlinear Lurie networked control systems are asymptoticallystable. Then the state feedback controller is designed. By numerical examples, we can see that the results obtained by the propsed method possesses less conservativeness.