基于RBF网络的金属应力状态系数模型  

The Model of Influential Coefficient in Stressed State of Medium and Heavy Plate Rolling Mill Based on RBF Neural Network

作　　者：雷明杰[1] 孟令启[1] 

机构地区：[1]郑州大学机械工程学院,郑州450001

出　　处：《微计算机信息》2009年第15期187-188,135,共3页Control & Automation

基　　金：基金申请人:孟令启;国家自然科学基金资助项目:中厚板轧机控制模型研制与开发(50175031)

摘　　要：以4200轧机轧制的大量实测数据为基础,利用Matlab人工神经网络工具箱,建立了轧制变形区的应力状态系数与轧前厚度、轧后厚度的RBF神经网络预测模型。通过分析应力状态系数的影响因素,结合传统的数学模型,确立了网络的输入层参数,并对函数newrb()中宽度系数spread的试验调整,确定了最佳的网络结构形式,提高了模型的预测精度,并与BP和Elman神经网络模型相比较,结果表明,RBF神经网络能更好地适用于金属应力状态系数模型。On the basis of the data obtained from large scale experiments in 4200 rolling mill, RBF neural network prediction model are established for the relationship between influential coefficient in stressed state and thickness before rolling and thickness after rolling by Matlab neural network toolbox. By analyzing influential factors of Coefficient of Stressed State and taking into account the traditional mathematical model, this paper affirmed the parameters of input layer, and By selecting suitable spread in function-newrb 0,this paper affirmed the best form of the network, as a result, it improved the prediction accuracy,and compared with BP and Elman network, the result indicated that RBF neural network can be better applied in the model of influential coefficient in stressed state of metal.

关 键 词：应力状态影响系数 RBF神经网络 预测 

分 类 号：TF3[冶金工程—冶金机械及自动化]