Quantized Control And Filtering Of Networked Control Systems With Time-delay And Packet Dropout

With the development of network technology and integrated communications, the network based automatic control systems are becoming a new application trend. Closed feedback control systems where in the control loops are closed through a real-time network are called networked control systems (NCSs). The enormous advantages of NCSs, including lower cost, higher reliability, simpler installation and maintenance, etc., which have made NCSs more applications than the traditional point to point control systems. However, owing to the limited communication ability, some new problems occurred, such as communication delays, packet dropouts, etc., also, the communication signals should be quantized for saving network resources, therefore the effect of the quantization on the performance of the NCSs should be considered. Which make the analysis and synthesis more complex than ever.In this paper, the problems of modeling, stabilization analysis, and method of designing controller and filter for NCSs subject to the major problems in NCSs using Lyapunov stability theory, stochastic system theory and LMI method.Firstly, for the NCSs with time-varying delays and multiple packet dropouts, the quantized Hx controller design problem is considered. Both the sensor-to-controller and controller-to-actuator random packet dropout and signal quantization problems are considered. Sufficient conditions for the existence of an observer-based delay-dependent controller are established to ensure the exponentially mean-square stable of the closed-loop system and achieve the optimal H∞disturbance attenuation level.Secondly, the observer-based H∞controller design problem for NCSs with signal quantization, sector-bounded nonlinearity, Ito stochastic disturbance as well as random communication delays is studied. The discrete-time stochastic nonlinearity systems model with both the measured output and control input signal quantization and random communication delays are established. Based on Lyapunov stability theory combining with linear matrix inequalities, sufficient conditions for the exponentially mean-square stable of the closed-loop system which satisfy the optimal H∞performance index are given, and also the observer-based H∞control design method.Finally, the H∞filtering problem is investigated for NCSs subject to quantization, discrete and distributed delays as well as missing measurements over unreliable communication channel. Sufficient condition for the existing of a filter is given such that the filtering error system is asymptotically mean square stable, and also the H∞filtering performance index. The filter parameters are designed by solving a linear matrix inequality.