Model Predictive Control for Line Pacing in Steel Processing Lines

Line pacing control in steel processing lines is known as a difficult task for operators since operators have to control the speed and length of steel sheets while keeping constraints and predicting the future situation. In order to avoid a shutdown of the line, operators usually takes conservative control policy. Therefore, the processing capacity of the line has not been fully utilized yet.For this problem, this paper presents a new modeling and control techniques for line pacing in steel sheet processing lines based on a Hybrid Petri Nets (HPN) and Mixed Logical Dynamical Systems (MLDS) expressions. The proposed modeling method can harmoniously integrate continuous (e.g., strip length, flow speed, and so on) and discrete (e.g., welding and cutting of sheets,) aspects of the system. Moreover, the proposed method can provide a suitable formulation for numerical optimization of the line pacing. Based on this hybrid system modeling, the closed loop control can be realized by receding horizon scheme. Numerical examples demonstrate that the processing capacity could be fully utilized by the proposed method.