Research on self-learning of hot strip finishing setup model for rolling force

To solve the problem of self-learning of finishing model in some hot rolling, a deep research on the function of short-term self-learning and long-term self-learning models was made. The point in the short-term self-learning is how to premise the smooth modulus in the method of exponential smoothing. A multi-variable control exponential smooth model was proposed. The results show that the result obtained by the proposed model is more better than the result of only using the single modulus self-learning model. In the aspect of long-term self-learning, the main mission is the launch conditions of the self-learning, and how to select the strategy of learning coefficient about the first piece of steel after changing the layer. Firstly, the judge conditions about the degree of the changing of the standard is joined into the launch conditions of the long-term self-learning, the forecast accuracy is ensured, it effectively reduces the launch times in long-term self-learning, the continuity of the self-learning is improved. Secondly, the method of tendency learning modulus is put into the original rolling force learning model. It effectively renews the equipment message belonged to the self-learning modulus in layer table, improves the forecast accuracy in the layer never rolled. Finally, the strategy of the self-learning modulus in the layer never rolled is proposed by using the similar self-learning modulus in the rolled layer, which effectively improved the accuracy of the initial self-learning modulus in the never rolled layer. ©, 2014, Central South University of Technology. All right reserved.