Development of parameterized roll pass design based on a hybrid model

Abstract

Hot steel rolling is an important manufacturing process used to efficiently provide a wide range of products of high quantity and quality. In order to meet the continuously increasing demands, both an improved quality and a broader variety of products must be delivered along with improvements in efficiency, reliability and sustainability of rolling mill systems. Further development of roll pass design presents one of the central aspects in these efforts. However, to optimize roll pass design, numerous combinations of system parameters must be analyzed and correlated. This cannot be done through a single deterministic model. Therefore, a parameterized hybrid model based on combining cross-disciplinary knowledge is proposed to improve the quality and efficiency of roll pass design. Application of artificial intelligent algorithms is envisaged for the roll pass design optimization and a methodology for constructing the relevant hybrid model is described.