Predictive Controller Design for Networked Control System

Abstract

A networked control system where sensor ( ), controller ( ) and an actuator ( ) are interacting to the process model or system through a communication channel. The model reliability depends on time delay of data transmission from to , to , controller processing time and packet dropouts. In this proposed work, model based network system is designed to achieve the desired response of the controlled system by minimizing the network time delay and improves con-troller performance in term of stability. Time delay mainly occurs due to time sharing of the medium as well extra overload of signal coding and processing phase. The time delays could be steady, limited, or even random, depends upon protocol used in the network and hardware. The network predictive control tech-nique effectively works on random delay. The predictive controller is used to analyses the future output conditions of the system based on the previous values. The simulated result shows that if time delay increases, the response of the system degraded and therefore for improving it the correct choice of prediction and control horizon is needed. The designed networked control system based on predictive functional control is simulated using True Time tool on Matlab. The ob-tained outcomes from the proposed technique accomplishes that the technique is effective and feasible.