Network Control System Dispatch And Dynamic Quantization Control With Communication Constraints

In recent years, with the rapid development and cross-penetration of control theory, computer networks and electronic technology, networked control systems have had a profound impact on automation technology and brought new challenges. In the paper, an integrated design scheme of scheduling, bit rate allocation and dynamic quantized feedback control for NCSs with communication constraints is proposed and investigated. The main achievements are as follows:Fist of all, for a class of NCSs with medium access constraints and random time delay, the problem of quantized feedback stabilization based on dynamic scheduling is investigated in the paper. Logarithm quantizers and TOD scheduling strategy are applied to forward and feedback channels, quantization error is dealt with sector-bounded method. By employing switched system theory and Lyapunov stability theory, a controller design method based on MEF-TOD scheduling strategy is given to guarantee stability and robust performance of the closed-loop system.Secondly, a simultaneous design scheme of H-infinity feedback controller based on observer and dynamic quantizers for a class of NCSs is proposed in this paper. In order to overcome the quantization error by static quantizer can not convergence to origin point, the mid-tread uniform quantizer is transformed to a dynamic quantizer which is composed of an adjusted "zoom" parameter. The quantizers update its parameters which take on a finite set of values based on the quantized inputs(the information of the control outputs and the system measurement outputs). Then a sufficient condition is given to guarantee the robust stable of closed-loop networked control systems, a H-infinity feedback controller and a quantized parameter update strategy is designed afterwards.Finally, an integrated design scheme of nodes scheduling, bit rate assignment and quantized feedback control for a collection of unstable NCSs using dynamic quantizer is proposed in this paper. Due to the restriction of communication constraints, a dynamic scheduling policy based on average dwell time strategy is raised to accommodate a limited number of plant-controller connections. A bandwidth scheduler is empployed in order to guarantee the convergence of quantization error and improve the performance of control system. Transform the dynamic quantization error to quantization range, a dynamic state feedback controller design is then given to guarantee exponentially stable of all controlled subsystems based on quantization-dependent Lyapunov function. For the proposed ideas and methods, the simulation examples demonstrate their efficiency by Truetime toolbox.