论文部分内容阅读
针对机理模型难以刻画的热轧精轧生产过程,采用基于数据子空间的偏最小二乘方法建立热轧轧制力数据模型,并构建轧制力优化模型,利用改进的粒子群优化算法对优化模型计算求解.结果表明,使用数据驱动方法建立的轧制力数据模型能够揭示精轧过程轧制力的机理规律,可以替代机理模型在实际系统中的应用.通过对整体优化模型的求解,可以提高热轧精轧产品的质量,降低能源消耗,表明基于数据驱动的建模和优化方法在实际生产中具有较大的应用价值.
Aiming at the hot rolling finishing process, which is hard to characterize the mechanism model, the data model of hot rolling force is established by using the partial least squares method of data subspace and the rolling force optimization model is built. The improved Particle Swarm Optimization The results show that the rolling force data model established by the data-driven method can reveal the mechanism of the rolling force in the finishing rolling process and can replace the application of the mechanism model in the actual system.By solving the overall optimization model, Improve the quality of hot rolling products and reduce energy consumption, indicating that data-driven modeling and optimization methods have great practical value in practical application.