The Research Of Modified Internal Model Control In The Networked Control System

In the traditional industrial control, it used the mode of point-to-point to make the regulator device, the actuator device, the sensor device and the controlled plant connected to a closed loop by cable. Owing to the technique of production has become more complicated and more diversified, the problems of complex cabling, high control cost and difficult to expand in the traditional industrial control is also increasingly exposed. In this case, with the rapid development of communication technology, control theory and computer technology, Networked Control System (NCS) came into being. In NCS, the information transmission and feedback is implemented through the network, and it is easy to realize remote control, share system resources, expand the system and decrease the cost. However, it also brings us some problems including the packet loss, networked time-delay and discarded time sequential etc.In NCS, as the adverse influence of networked time-delay, the traditional control strategy may not be up to the people’s control requirements, while Internal Model Control has the advantages of simple structure, strong anti-interference and strong robustness. How to improve the original Internal Model Control (IMC), so as to reducing the adverse effects of the time-delay to obtain better control effect, is this paper’s initial point. In this paper, these works are carried out from the following three aspects.1、The networked control system in the case of short time-delay and long time-delay is modeled when the sensor adopts time-driven, the controller and the actuator both adopt event-driven. And some experiments based on one degree freedom control and two degree freedom control are carried out on the TrueTime2.0 platform. The simulation results show that IMC has a good tracking performance and anti-disturbance.2、In view of the uncertainty of the NCS, it is very difficult to obtain the model of the process. It uses SVM to model the process online in this paper and gains the inverse model to design an internal model controller. At last, the experimental results verify the feasibility of the control algorithm.3、In order to get a better control effect, some improvements on the internal model controller are done in this paper based on the principle of bionics and artificial intelligence, as the introduction of the endocrine ultra-short feedback, it greatly reduces the delay effect, and improves dynamic effect, and it has a better anti- disturbance performance.