基于神经网络的冷连轧机轧制力预报模型
来源期刊：中南大学学报(自然科学版)2006年第6期
论文作者：周富强 曹建国 张杰 尹晓青 贾生晖 曾伟
文章页码：1155 - 1160
关键词：冷连轧机;轧制力;神经网络;数学模型 中图分类号:TG333.71
Key words：tandem cold rolling mill; rolling force; neural network; mathematic mode
摘 要：为了提高冷连轧机轧制力预报精度,提出一种解析数学模型结合神经网络校正模型的计算方法,建立冷连轧机轧制力预报模型。采用径向基函数的局部映射和全局线性映射相结合的神经网络校正模型求解带钢变形抗力和轧制变形区的摩擦因数;并采用轧制变形区离散化方法分析轧制变形区内张力、摩擦力及金属变形抗力等在带钢轧制方向上的分布规律,从而建立轧制力在线计算数学模型。现场实测数据离线仿真结果表明,采用此基于神经网络的冷连轧机轧制力预报模型预测轧制力,其预测误差小于8.9%,此模型能用于指导生产实践。 附件: z2006-06-27
Abstract: A prediction model of the rolling force for tandem cold rolling mill based on neural networks and mathematical models was put forward to improve the pre-calculation precision of the rolling force for tandem cold rolling mill. The neural networks correction method combining the local map with radial based function (RBF) networks and the global map with linear transform was applied to calculate the deformation resistance of the strip and the friction coefficient in the rolling deforming region. And the distributions of the tension and the friction force and the deformation resistance of the strip along the rolling direction were discussed by the discretization of the rolling deformation region. Then the online calculation model of the rolling force for tandem cold rolling mill was set up. Contrasting the measuring data with the simulation result, the prediction error can be limited within 8.9%, and the practical production can be instructed by using the prediction model of the rolling force for tandem cold rolling mill based on neural networks.
基金信息：北京科技大学科技发展专项基金资助项目
