MPC controller used with ARX methods via control and modelling with ANN for a wire coating thickness

Author Bekir Cirak
DOI

Country : Turkey
Subject : Mechanical Engineering

Keywords : Wire coating Processes, Coating thickness, artificial neural network PVC, Matlab Extrusion, Model Predictive Control

Abstract

Nowadays, wire coating process is used in almost every area especially in the electrical and electronic industry. Wire coating is a process which utilizes a variety of complex equipment. This consists of a plasticating extruder fitted with a wire coating die through which the wire is continuously drawn. A three layer back propogation artificial neutral network (ANN) model was used for description of wire coating thickness. On comparing the experimental data, the predictions the ANN model predictions, it is found that the ANN model is capable of predicting the coating thickness. In this paper, the results of experimental investigation are presented by comparing the coating quality on galvanized mild steel wire using EP 58 PVC molten is used as the coating material in a wire coating extrusion unit at different extruder temperatures and extruder speeds.

Simulations of MPC Control algorithms for coating process have been made and the results of these simulations was observed. There are a comparison of PID Controller and Generalized Predictive Controller results and there are comments about this comparison in these studies. The simulations and calculations of the algorithms are done in MATLAB Packaged Program Environment.MPC is especially suitable for controlling these types of systems. MATLAB Packaged Program is utilized in all these studies.

Download