Horizon-Varying Model Predictive Control for Accelerated and Controlled Cooling Process

摘要

In accelerated and controlled cooling (ACC) processes, the relationship between the plate velocity and the final temperature of plate point is complicated and implicitly expressed, with undetermined ending time of the plate cooling. It is difficult to obtain the optimal plate velocity to accurately control the final temperature. A novel control methodology is proposed to improve the accuracy of the final temperature. The technical targets are transformed first into several location-dependent temperatures to simplify the ACC process as a linear system. Then, a horizon-varying model predictive control algorithm is developed to obtain the optimal final temperature of the entire plate. The optimization horizon is estimated according to the plate length and the prediction of future input sequence in each control period. The experimental results show the effectiveness of the proposed method.