Intelligent Plate Shape Control Strategy With Batch Changeover and Its Application in Roller Quenching Process

Plate shape is a critical indicator of steel plate quality. During the roller quenching process, there are problems with significant time delays in plate shape feedback and frequent batch changeovers, leading to difficulties in plate shape control. To address the issue of time delays in feedback, a plate shape prediction model is designed to forecast the plate shape in advance. An intelligent plate shape control strategy is proposed to address the issue of batch changeover. First, by analyzing the empirical probability density of plate shape, a group of plate shape parameters that can be practically implemented is determined as the control targets. Then, when the batch changeover occurs, a case matching method is designed to ensure the stability of plate shape under drastic changes in the production environment. When there is no batch changeover, the designed plate shape prediction model and intelligent optimization method are employed to calculate optimal control parameters. Finally, the proposed method is applied to industrial sites, increasing the pass rates from 97.2% and 96.0% to 99.6% and 98.0%, respectively. The application results show that the proposed strategy has high control accuracy, providing an effective way to improve the quality and pass rate of plate shape.