摘要
厚度预测模型的精度是影响厚度控制的重要因素。针对本项目国内水平领先、最宽幅的"1+4"热连轧生产线,根据生产现场获取的5083宽幅铝合金中厚板实测数据,在研究分析关键影响因素的基础上,运用人工神经网络技术建立了铝合金宽幅中厚板厚度预测的BP神经网络模型。其相对误差在0.5%之内,高于已有模型预测精度,能实现高精度预报。应用模型预测了5052宽幅铝合金中厚板的出口厚度,结果表明,模型能较好的预测轧件厚度的变化,有很好的泛化能力。
The accuracy of the thickness prediction model is an important factor which was influenced the thickness control. Aiming at the '1 + 4'hot tandem rolling line which was domestic leading and most wide,According to the processed measured data of as-rolled 5083 wild aluminum medium plates,a thickness prediction model was developed by artificial neural network based on the analysis of key factors. The relative error of the model is within 0. 5%,which is better than that of previous models. And the high-precision prediction for rolling thickness was achieved. The developed model was successfully to predict the thickness of 5052 wild aluminum medium plate and exhibited good generation ability.
出处
《功能材料》
EI
CAS
CSCD
北大核心
2015年第6期6102-6105,6110,共5页
Journal of Functional Materials
基金
科技部国家科技支撑计划资助项目(2012BAF09B04)
科技部国际科技合作计划专项资助项目(2011DFR50950-03)
科技部国际合作资助项目(2011DFR50950)
支撑计划资助项目(2012BAF09B04)
关键词
人工神经网络
热连轧
中厚板
厚度预测
artificial neural network
hot tandem rolling
mediumplate
thickness prediction
