Stability Analysis Of System With Quantize Output Feedback

In control system, system's main performance indexes as well as the stability are much affected by performance of real-time and accuracy of signal transmission. While in the modern network control system, there are many various intelligences sensory equipments and the low cost digitized intelligence measuring appliances, which could deal the data and transmit them between each kind of equipments. So those make the structure of the network control system much complex. Along with information transmission between kinds of digital equipments increasing, this brings the huge correspondence pressure to the network transmission path, leading to network congestion or time delay which will influence the real-time of signal. Therefore the advantage of quantizing the state signal and control signal is appear, Not can only enhance the network the used efficiency, reduces the number of the transmission signal quantity, moreover declines the investment and allow the long-distance network monitor and control. However, the signal quantification is lossy compression that will lead to the information loss and bring quantized error, In addition some unstable factors of the system self as well as noise signal disturbance, all of those affect to system's performance index and the system stability. So studying the stability of quantized output feedback control system have important theoretical and practical significance.This paper studies the stability of quantized output feedback control systems without noise disturbance subjected to communication constraints firstly. We can obtain the estimate error system through by the differential of plant system and state estimate system. For state estimate error system, a invariant region is constructed using the Lyapunov equation, and obtain the system stable condition through the control policy of dynamic adjusting the scaling parameter. Meanwhile for the plant system we can obtain the similar conclusion using the same method. Then the paper analyses the connection between state estimate error system and the plant system, and get some corresponding conclusions. The digital simulation experiment has given in order to prove the validity of the control policy.Secondly, Then the article transits from the non-noise system model to the system model with noise. We use the method similar to the second chapter to analyses the stability of new system. Also we obtain the quadratically attractive conditions and the minimal invariant region near the equilibrium points respectively. At the same tine, the connection of the two invariant regions is given too.Finally, a method of designing state estimator of quantize output feedback network system with two quantizers is given. First, in order to get the optimize estimator, it analyses the astringency of the covariance matrix subjected to estimate error system. When the covariance matrix of estimate error system is minimal, then it can guarantee that the estimator could track state of the plant system accurately, and also reduce the error, have been advantageous to system's monitoring.