OPTIMAL CONTROL OF THICKNESS AND FLATNESS FOR THE 20 ROLLS SENDZIMIR COLD ROLLING MILL

摘要

This paper presents a system for the real-time optimization of the cold rolling processes for the 20 rolls Sendzimir cluster mill. Based on an compact model of the rolling process that exploits especially its response to the control inputs for strip thickness and flatness controls [Pu, et al. 1996], a state-observer for the real-time state-estimation have been designed and experimentally verified [Pu, et al. 1998]. The Sequential Quadratic Programming (SQP) is used for the optimal control of the thickness and flatness for the cold rolling process with 10 control inputs. Technical mill restrictions are taken into consideration during the optimization. The developed system is able to run under real-time condition. Experiments with on-line data measurements during the mill operation show the efficiency of the system.