Rolling Force Prediction Method Based on Fuzzy Identification

Abstract

To improve the precision and efficiency of rolling force prediction on hot rolled strip, a new rolling load prediction of finishing stands method was set up by fuzzy identification. It was based on T-S fuzzy model using clustering subjection functions to calculate the grade of membership for each given pattern, and using recursive least squares method to identify the consequent parameters of fuzzy model. On the basis of the measured data of the 1580 mm, the relation between the main hot strip mill parameters and rolling force was established using fuzzy model. Experimental results show that the prediction precision is higher, responds quickly and steady. The method can satisfy on line control requirements in a hot mill strip rolling process.