Analysis And Design Of Networked Control Systems With Time-delay And Quantization

The optimal control and stability analysis for networked control systems with time-delay and quantization parameters design of systems with finite quantizers studied in this dissertation when the network-induced delay and signal quantization as two basic questions are study objects.Firstly, design of compensation algorithm and optimal control law of networked control system with time delays are studied. In view of the system existence random time delays, we design state observers separately under the full condition feedback system and the output feedback system to realize the time delays compensation, thus we can obtain the more accurate forecast estimate. Meanwhile, we present the optimal control law of the networked control system based on LQG performance, to weaken the influence of time delays and system noise on system stability, so we can ensure the stability of the system. Finally, passed the example simulation, we confirmed the validity of the time delay compensation strategy and the optimum control design.Secondly, based on existent result on control network performance and scheduling in protocol layer, the multi-task networked control systems for scheduling simulation in virtual platform in considered. In addition, technologies for integration of control and scheduling design are explored. The influence of sampling period to NCS is discussed. Based on the proposed quadratic cost function and network scheduling, an optimization model and sampling period selection algorithm are proposed. An example shows that the algorithm satiafies performance requirement for NCS, optimizes the network scheduling, and improves the network efficiency.Finally, concerned with asymptotic stability analysis and quantization parameters design of discrete-time linear system subject to output feedback and finite logarithmic quantizers. The asymptotic stability analysis and its corresponding stability conditions on observation error system and plant system are given, respectively. Then, the numerical relationships of stability between plant system and error system are presented explicitly under finite logarithmic quantizers. Further, the upper bound interval value of each finite quantizers and their data rate ratio are achieved. Last, a hybrid quantized control strategy based on state observation is proposed as well as a numerical example is given. So, the design is feasible and effective.