基于GRNN网络的中厚板轧制温度的预测

 针对中厚板轧机控制模型中的轧制温度精度的提高问题，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，建立了中厚板轧制温度的GRNN神经网络预测模型。通过分析影响钢板温度变化的各种因素，调整神经网络的光滑因子，确定了最佳的网络结构形式，提高了模型的预测精度，并与传统的BP神经网络模型相比较。结果表明，GRNN网络具有更高的精度和更好的泛化能力。该神经网络模型可应用于中厚板轧制温度的预测，也可为人工神经网络在其它自动控制方面的应用提供参考。

Prediction of Rolling Temperature of Medium and Heavy Plate Based on GRNN Neural Network

For the improvement of the accuracy of rolling temperature in the medium and heavy plate rolling mill control model, on the basis of the data obtained from large scale experiments on 4200 rolling mill, a GRNN (generalized regression neural network) neural network prediction model of rolling temperature is established by Matlab neural network toolbox. By analyzing influencing factors of rolling temperature and by selecting suitable neural network, the best architecture of the network can improve the prediction accuracy, and compared with BP network, the result indicates that GRNN neural network has better accuracy and adaptability of the network. The neural network model can be used to predict the medium and heavy plate rolling temperature, it can also be used for artificial neural networks in other automatic control.