Study on Temperature Rise Modeling of Main Motor of Hot Rolling Mill Based on Support Vector Machines

Abstract

Along with the ongoing process of the adjustment of industrial structure in China, overcapacity of the traditional heavy industry has become an issue of deep concern, and the direct consequence of overcapacity is energy waste. Tandem rolling mill is the typical equipment whose designed capacity is greater that the current real need. In many steel mills the practical work load of tandem rolling mill is far below the rated, while its forced-air cooling motor still runs at full capacity regardless of any change of heat load or season, leading to energy waste. Adjusting drive frequency of the air cooling motor according to its heat load is a practical and feasible measure for energy saving. As one of the important aspects of variable speed control of the air-cooling motor, to establish a precise temperature rise model for the object is critical. This research focuses on the temperature rise modeling of the main motor of hot rolling mill. The temperature rise mechanism of the main motor is first analyzed, and then the support vector machines regression algorithm is used for parameter identification of the model. To ensure the security and reliability of the production line, different penalty coefficients are adopted for positive error and negative error. Long term workshop data are collected and used for comparison with the predicted data. Comparison results demonstrate the feasibility of the established model.