锻压技术

基于ReLU激活函数的轧制力神经网络预报模型

英文标题：Neural network prediction model of rolling force based on ReLU activation function

作者：刘杰辉范冬雨田润良

单位：河北工程大学

分类号：TF334.9

出版年,卷(期):页码：2016,41(10):162-165

摘要：

平整机轧制力的预报对轧制过程的优化控制有着重要意义。针对平整机轧制力预测精度不高的问题，提出采用ReLU(Rectified Linear Units)激活函数的神经网络模型来预报平整机的轧制力。在对数据进行主成分分析后，得到影响轧制力的主要因素，并将其作为神经网络的输入层，将平整机轧制力作为输出层，通过使用Python语言编程进行实验，对神经网络模型隐层的相关参数及算法进行单一变量筛选，建立了保证轧制力预报精度最高的神经网络模型。实验结果表明，通过调整隐层层数、神经元数、传播算法、正则化方法，该模型能够将预测误差控制在10%以内，且该实验方法能够对不同输入参数下的平整机轧制力进行精确预报。

It is very important for the prediction of temper mills rolling force to control the rolling process optimally. For the problem of low precision in rolling force prediction, a neural network model with ReLU (Rectified Linear Units) activation function was proposed to predict the rolling force of temper mill. After the principal components of data were analyzed, the main factors affecting the rolling force were obtained, which were selected as the input layer of neural network, and the rolling force of temper mill was taken as the output layer. The experiment was carried out by the python programming language, and related parameters and algorithms in the hidden layer of neural network model were chosen with the principle of single variable. Then a neural network forecasting model was established to ensure the highest accuracy of rolling force prediction. The results show that by adjusting the number of hidden layers, the number of neurons, the propagation algorithms and the regularization methods, the model is able to ensure the prediction error less than 10%. And the experimental method can accurately predict the rolling force of the temper mill with different input parameters.

基金项目：

河北省自然科学基金资助项目 (E2015402112)

作者简介：刘杰辉(1968-)，男，硕士，副教授 E-mail：18230108301@163.com；通讯作者：范冬雨（1990-)，女，硕士研究生 E-mail：1533792983@qq.com

参考文献：

[1]刘杰辉,王桂霞,刘永康. 基于灰色理论和神经网络的轧制力预测[J]. 锻压技术,2015,40(10):126-129.

Liu J H,Wang G X,Liu Y K. Prediction of rolling force based on grey theory and neural network[J]. Forging & Stamping Technology,2015,40(10):126-129.

[2]陶红勇,王京,陆秀志. 神经网络在板形控制中的应用[J]. 轧钢,2003,20(4): 10-12.

Tao H Y,Wang J,Lu X Z. Application of neural network in shape control[J]. Steel Rolling,2003,20(4): 10-12.

[3]Glorot X,Bordes A,Bengio Y. Deep sparse rectifier neural networks[J]. Journal of Machine Learning Research,2010,15:315-318.

[4]Hinton G,Osindero S. A fast learning algorithm for deep belief nets[J]. Neural Computation,2006,18(7):1527-1554．

[5]Larochelle H,Bengio Y,Louradour J,et al. Exploring strategies for training deep neural networks [J]. Journal of Machine Learning Research,2009,10(12):1-40.

[6]Yu D,Deng L. Deep learning and its applications to signal and information processing[J]. IEEE Signal Processing Magazine,2011,28(1):145-154．

[7]Deng L,Yu D. Deep Learning: Methods and Applications [M]. USA, Hanover:NOW Publishers,2014.

服务与反馈：

【文章下载】【加入收藏】