Application of artificial bee colony algorithm to select architecture of a optimal neural network for the prediction of rolling force in hot strip rolling process

In the face of global competition, the requirement for dimensional accuracy, mechanical properties and surface properties has become a major challenge on aluminum manufacturing industries. Conventional rolling force formulas, however, provide not more than reasonably exact approximations. The mathematical modeling of the hot rolling process has long been recognized to be a desirable approach to aid in rolling operating practice and the design of mill equipment to improve productivity and quality. However, many factors make the theoretical analysis of the rolling process very complex and time-consuming. This paper presents a prediction method based on the Artificial Bee Colony algorithm and BP neural network, which was developed in order to improve the prediction of rolling force in hot strip rolling process. The architecture of BP neural network is optimized by Artificial Bee Colony algorithm. Comparing with the Sims mathematical model and BP neural network, the experimental results show that the prediction accuracy and error of rolling force is superior to the other two methods, and the predicted rolling force is very close to the practical rolling force.