H_∞ Control Of Networked Control Systems Based On Time-Delay System Theory

Networked Control Systems (NCSs) are distributed control systems in which sensors, actuators, controllers, and other units are connected via a common communication network. The primary advantages of NCSs are reduced system wiring, ease of system diagnosis and maintenance, increased system agility and sharing information resources. However, the insertion of the common communication network in the control loop brings us many new problems, such as the time-varying network-induced delays, the uncertain packet-dropouts, communication constraints, et al.Recently, these new problems arisen in NCSs which have attracted considerable research interest. Especially, the networked-induced delay, which is either constant or time-varying, needs to be fully addressed in NCSs. Various approaches have been presented in the existing literature to deal with the delay issue, such as the stochastic optimal control approach, the Asynchronous Dynamical System (ADS) approach, fuzzy control method, predictive control method and so on. However, the great mass of issues only localized the NCSs with single sensor node and single actuator node. The aim of importing communication networks to control systems is not to improve control performance of the systems, but to reduce wiring and break the limit of region space of the control system, and ultimately to decrease the complexity of the system, and to realize control system centralization and highly effective operation. Thereby, the superiority and value of the control systems can be manifested by applying networks in control system with many sensor and actuator nodes.In this thesis, we investigate the modeling, stability and H_∞control for NCSs with multiple sensor and actuator nodes by applying the theory of time-delay systems and the Linear Matrix Inequality (LMI) techniques. The main contributions of the thesis are as follows:1) The survey of the networked control systems, which includes the basic contents, issues, and recent research on networked control systems are presented. Then, the main work of the thesis is stated.2) For linear time-invariant normal controlled plant, the NCSs with state feedback controller is modeled as a class of time-delay systems with time-varying delays. Specifically, a class of multi-input multi-output (MIMO) NCSs with time-varying delays is described a time-delay system. Based on the obtained system model, the stabilization problem is studied by applying a newly proposed method involving the Lyapunov approach and the LMI technique. Further, the design method of H_∞controller is presented. Two numerical examples are given to show the effectiveness of the proposed approach.3) Based on Lyapunov-Krasovkii functional and LMI technology, the stabilization problem is investigated for the MIMO NCSs with dynamic output feedback controllers, and a new stability condition which is less conservative than some of the existing results in the literature is derived. The MIMO NCSs with time-varying delays which have two independent sensors and two independent actuators are constructed to show the effectiveness of our results. Moreover, the design method of H_∞control is also presented based on the obtained system model and the stability analysis results. In the end, an example is given to demonstrate the effectiveness of our results.