Study Of Control Model Of Laminar Cooling Based On Neural Networks

Laminar cooling is one part of the technique for the control of rolling and cooling, it effects the tissue and properties of plate directly. So laminar cooling process plays an important role in hot plate production. The control accuracy of the coiling temperature of the hot plate laminar cooling system is the plate mill to guarantee the better quality and flatness of plate. This paper takes a laminar cooling system of TISC hot plate mill for background, and has a deep study for how to raise the accuracy of cooling temperature of the hot rolled plate.As the mathematic model of hot plate laminar cooling system is based on the theory of calorifics, so the paper first analyze the models of heat transfer,and particular analyze the mathematic model of the process of this hot plate laminar cooling system. This mathematic model mainly includes setup model and adaptation model, its calculated precision influences the last erect of control cooling. So the proper heat transfer model is very important thing which can improve the accuracy of the coiling temperature.Study and analyze the laminar cooling control strategy of a hot mill plant, and particular analyze the all functions of control strategy. The paper has established a simulate system which can offer an experiment tool to improve the control strategy and the model.Point out the major reason why coiling temperature isn’t accurate and temperature equality on the same plate isn’t good is the temperature model is not accurate enough. ANN has the ability to analytic nonlinear complex process, so the paper makes a study on the ANN, adopt improved BP neural network to predict the plate’s coiling temperature.The result of the simulation shows that this method is very effective.