基于GRNN神经网络的中厚板轧机厚度预测  

Thickness prediction of medium plate mill based on GRNN neural network

作　　者：郭斌[1] 孟令启[1] 杜勇[1] 马生彪[1] 

机构地区：[1]郑州大学机械工程学院,河南郑州450001

出　　处：《中南大学学报（自然科学版）》2011年第4期960-965,共6页Journal of Central South University:Science and Technology

基　　金：国家自然科学基金资助项目(50175031)

摘　　要：根据4200轧机轧制45号钢的实测数据,利用Matlab神经网络工具箱,建立中厚板轧机厚度预测的GRNN神经网络模型。研究结果表明:GRNN神经网络模型能较好地预测轧件厚度的变化,相对误差很小;与BP网络和Elman网络模型相比,GRNN网络模型具有更高的精度和更强的稳定性。According to the measured data of rolled 45 steel obtained from 4200 medium plate mill,a GRNN neural network prediction model was established by Matlab neural network toolbox.By analyzing the results,GRNN neural network model can predict the thickness well,and it has very small relative error.Compared with BP network and Elman network,GRNN network can improve its accuracy and stability to some extent.

关 键 词：GRNN神经网络 中厚板 厚度预测 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]