Neural control of a steel rolling mill

The application of nonlinear neural networks to control of the strip thickness in a steel-rolling mill is described. Different control structures based on neural models of the simulated plant are proposed. The results for the neural controllers, among them internal model control and model predictive control, are compared with the performance of a conventional proportional-integral controller. By exploiting the advantage of the nonlinear modeling technique, all neural approaches increase the control precision. In the application considered, the combination of a neural model as a feedforward controller with a feedback controller of integral type gives the best results.<>