Stability Analysis And Synthesis Of Nonlinear Networked Systems With Bandwidth Constrained

The definition of networked control systems(NCSs) is that the communication network is regarded as a bridge and can connect different component units(such as actuator, controller and sensor) which be distributed in different geographical locations, so it can come true data exchange and information sharing. Compared with the traditional control system, the NCSs have more advantages such as better reliability, resource sharing,less wiring, low cost, easy to install and extension, which prompted the application of NCSs obtains rapid development in military, medical treatment, aerospace and industrial areas. Although the introduction of the network drives the rapid development of the control field, there are still some new problem and new challenges, such as data packet loss,quantization and network delay,etc, which results in a decline in the stability of the control system and even cause direct economic losses. Therefor, the traditional method of thinking and analysis is difficult to directly used in the networked control system, we must start with a new visual angle to research, in order to ensure the normal operation of the network control system.This paper mainly studies the problem of data packet loss and quantization for the bandwidth constrained network control system which based on linear matrix inequality and lyapunov stability theory. In addition, the paper also takes the nonlinear and saturation phenomenon of the NCSs in attention. From what has been discussed above, the main research content of the article is divided into the following several aspects:The third chapter, it mainly studies the state feedbackH_∞ control problem of networked control systems with random packet loss and quantization. This chapter adopts logarithm quantizer and consider random packet loss which the stochastic variable conformed to the bernoulli distribution so that we can construct a complete networked control model including random packet loss and the signal quantization. Then based on the linear matrix inequality(LMI) and lyapunov theory, the design method of state feedbackH_∞ controller is given for a class networked control systems which exist random packetloss and the signal quantization simultaneously. Moreover, the sufficient condition based on LMI is proposed to prove the closed-loop system is exponential mean-square stability and satisfy specifiedH_∞ performance index.The fourth chapter, it mainly researches the quantizedH_∞ filtering problem for networked control systems including non-linearity. For a given model of networked control system containing non-linearity, it will be rebuild and transform into linear networked control system to the linear systems through nonlinear decomposition technique. This chapter uses a dynamic quantizer which is consist of a dynamic adjusting parameter and a static quantizer. Based on the LMI technology, a design method of quantizedH_∞ filter about nonlinear network control system is proposed and prove that the filtering error system is asymptotically stable and satisfy given filtering target.The fifth chapter, it mainly research quantizedH_∞ filtering problem of NCSs whichas sensor saturation and random nonlinear. Firstly, this chapter builds a new and more realistic networked control model including sensor saturation, quantization and random nonlinear. Then combining the dynamic quantized strategy and using lyapunov theory and LMI method, a quantizedH_∞ filter is designed successfully and can ensure system meet the index of mean square stability andH_∞ performance.