Time-delay Compensation Of Networked Control System Based On Dynamic Matrix Control

With the rapid development of communication methods and technologies,networks have played more and more important roles in many industrial control applications,especially in some engineering applications,signals need to be transmitted over communication networks,such as the remote control of several mobile units,multi nodes distributed control system and wireless sensor networks,etc.Motivated by these emerging engineering applications,more and more attention has been drawn to the development of a general theory and application for networked control system(NCS).Compared with traditional feedback control systems,these components are usually connected via point-to-point cables.The main advantages of NCS are their low cost,reduced weight and power requirements.Moreover,they are simple to install and maintain,and exhibit high reliability as well.Research topics on NCS include modeling,stability analysis,control and filtering problems,etc.For NCS,there is a general issue should be taken into account due to the communication constraints.It is network communication time-delay,which is usually caused by limited bit rate of the communication channels,by nodes waiting to send out a packet via a busy channel,or by signal processing and propagation.The network induced time-delay affects the performance of the network system,and even causes stability instability.For network induced time-delay compensation problem,many scholars have conducted a lot of research.Robust control,intelligent control,fuzzy control and other time-delay compensation methods have made a lot of development and application.The dynamic matrix control algorithm is analyzed and introduced in this dissertation,and on this basis,the time-delay compensation method based on Proportion integration(PI)controller dynamic matrix control for networked control system is presented.Radial Basis Function Neural Network is used to predict the current time-delay.The predictive time-delay replaces the actual time-delay as the parameters of PI-based dynamic matrix control compensation controller.In order to improve the compensation energy efficiency of dynamic matrix control,this dissertation makes use of the advantages of PI feedback structure and predictive ability of dynamic matrix control.The optimal control incremental value can be obtained through new objective function combining with PI structure.PI controller cancorrect the output of dynamic matrix control,decrease the error between actual output and predictive output.Thus the effect of model mismatch and interference on system is reduced.The robustness and anti-interference performance of system is improved.The controller can select the appropriate control value to be transmitted to the actuator,which can compensate for the error caused by the random time-delay in the NCS.The dissertation gives the stability proof of compensation method.Through the simulation results,the effectiveness of time-delay compensation method in the dissertation is verified.