PREDICTION OF THE MEASURED TEMPERATURE AFTER THE LAST FINISHING STANDUSING ARTIFICIAL NEURAL NETWORKS

In this report the development of an artificial neural network, capabl e of predicting the temperature after the last finishing stand of a ho t strip mill for a certain class of steels, is described. Three neural networks with different numbers of hidden nodes (3, 5 and 7) were tra ined. The relative standard deviation in finish temperature as predict ed by the best performing neural network model (7 hidden nodes) was ju st over 25% smaller than that of the linear Hoogovens model. This impr oved accuracy can be explained by the incorrect assumption in the Hoog ovens model of linear dependence of the finishing temperature on some input parameters. With the trained neural network, the influence of th e various input parameters on the finishing temperature could be exami ned. The dependencies predicted by the neural network can be approxima ted by a linear fit and are a factor 2 lower for all input parameters. it is conceivable that operation of the mill using an artificial neur al network for the prediction of the finishing temperature would have resulted in smaller operational fluctuations.