Internal Model PID Control For A Class Of Networked Control System With Short Time-Delay

The appearance of Networked control system (NCS) is a mark of the wide application of computer and communication technology in industry control fields. The information in NCS is transmitted and feedback on network. NCS has advantages on installation, maintenance cost compared with the traditional control system. Yet, some ignorable problems that include network delay, packet loss and some others emerge as the insert of network. For achieving a well control performance in NCS, modify the traditional control strategies is a significant work in the progress and application of NCS.A fact that time delay always exists in NCS, and traditional proportional integral derivative (PID) controllers are still used broadly in industrial sites as their strong adaptability and robustness is well known. The control engineers must face to an unavoidable problem how to modify PID controller available for the bad influence of time delay in NCS? On the basis analysis of research background and research status at home and abroad of NCS, this paper focuses on the prospective of advanced PID controller design, some works are done mainly from the following aspects:1、In view of the influence of delay on analysis and controller design in NCS, the composition of delay, influence factors, measurement and modeling methods of NCS with short time delay are studied, and delay distribution curve of NCS is measured and draw.2、In order to simplify the application complexity of existing modified PID controllers in NCS, on the basis analysis of classic internal model control PID (IMC-PID), aim at the first and second order control plants. Orient a kind of NCS with short delay, IMC-PID controllers are designed respectively, which have only one filter factor parameter and simple structure.3、So as to have a well control performance, an improvement of IMC-PID controller is made, where the IMC PID control is combined with partial swarm optimization (PSO) to find the best value of the proportional factor added to the deduced proportional gain, and online sequential extreme learning machine (OS-ELM) to adjust the filter factor online.4、For proving the effectiveness of proposed controllers, simulations using Truetime blocks in Simulink/Matlab are designed, a discussion of the output response curves, parameter variation curves and performance indicators value under different cases are made.