Model-based predictive control of twin-roll strip casting thermal mode

Abstract

Twin-roll strip casting is one of the most promising technologies for the production of hot rolled steel strip. Liquid steel from mold is cast in 5 to 8 mm thick strip. After casting, the strip enters the roll stand, where it undergoes final plastic forming. The main goal of thermal mode control system is mass-average temperature stabilization at the hot rolling stand. The control of thermal mode is chiefly realized by changing the speed of casting. Significant time delay requires using the model-based predictive control systems. Block diagram of the twin-roll strip casting model-based predictive control system developed is provided in this paper. Distinctive features of the developed system are based on the previously developed mathematical model of the process and specialized command module. This module provides a real-time optimization of performance and conformity of the used models with current conditions. Modeling of the system performance in dynamic operating conditions of the unit is realized. Results of the modeling are presented in the form of transient process control system.