Intelligent controller design for the flatness control in a cold rolling process

Abstract

The flatness control in a cold rolling mill is an important subject because of the need for improvement in cold-rolled strip quality. It, however, is a difficult problem for a conventional approach to achieve since the cold rolling process is a highly nonlinear system in which many uncertain parameters are involved. The fuzzy controller for the flatness controller is designed by the heuristic approach that is based on the operator's experience and knowledge gained in the experiments. The feature of a neural network's learning and adapting ability is used for inverse modeling of the static model, and the error-decomposition network is developed as the inverse static model.