2000 Volume 40 Issue 8 Pages 771-778
This paper describes a new mill-setup system for a hot strip finishing mill in the iron and steel making industry, that integrates process models and operator expertise. Operator expertise is expressed in terms of roll gap differences between neighboring stands to avoid excessive loads that may lead to unstable rolling. The problem is formulated as that of minimizing deviation between the know-how-based roll gap difference pattern and the actual one. In order to realize such mill set-up considering the roll gap differences in actual plant, in which the computational time is strictly restricted, an estimation method of the Jacobian matrix for solving the problem is proposed. The efficiency and validity of the method are discussed, based on the linear system theory and case study results of simulations. The method is found to have valid computational properties for applying to the actual set-up of the process. Applying the system to the actual plant, it was found that operator intervention has been markedly reduced compared with the conventional setup method. Furthermore the stability of operation, i. e. a decrease in the frequency of defective rolls, is also achieved.