锻压技术

BP神经网络对斜轧穿孔轧制力的预测

英文标题：Prediction on rolling force of oblique rolling piercing based on BP neural network

作者：林伟路丁小凤双远华

单位：太原科技大学

分类号：TP183

出版年,卷(期):页码：2018,43(10):175-178

摘要：

为了有效预测AZ31镁合金管坯在三辊斜轧穿孔变形区的轧制力，保证穿孔后镁合金管材的优良性能，借助MATLAB工具箱建立了三辊斜轧穿孔的BP神经网络模型。结合穿孔过程中影响轧制力的因素，从实际生产过程中抽取了325个数据作为试验样本；根据AZ31镁合金在不同穿孔参数下的变形特点和三辊穿孔的有限元结果分析，利用轧制力的经验公式实现了理论计算，并将预测结果与理论结果进行了对比。结果表明：实际值与计算值的误差为14％，网络预测的最大误差为5％，平均误差为2.4％，最小误差为1.4％。因此，网络预报精度高，操作简洁，可以代替复杂的数学计算模型。

In order to predict the rolling force of AZ31 magnesium alloy tube billet in the oblique rolling piercing deformation zone with threeroller effectively and ensure the excellent performance of magnesium alloy tube after piercing, the BP neural network model of oblique rolling piercing with threeroller was established by MATLAB toolbox. Combined with the factors affecting the rolling force in the piercing process, three hundred and twentyfive data extracted from the actual production process were applied into experiment simples. According to the deformation characteristics of AZ31 magnesium alloy with different piercing parameters and the finite element analysis result of threeroller piercing, the theoretical calculation was realized by the empirical formula of rolling force, and the predicted results were compared with the theoretical results. The results show that the error between the actual value and the calculated value is 14%, the maximum error of the network prediction is 5%, the average error is 2.4% and the minimum error is 1.4%. Thus, the network prediction has high precision and simple operation, and the complex mathematical calculation model can be replaced.

基金项目：

山西省留学基金资助项目（2017-084）

林伟路（1990-），男，硕士研究生，E-mail：1499794610@qq.com；通讯作者：双远华（1962-），男，博士，教授，E-mail：2465752485@qq.com

参考文献：

[1]刘欣玉,潘露,帅美荣. 基于Matlab的BP神经网络轧制力预报模型及应用[J].重庆科技学院学报：自然科学版, 2016, 18(6): 96-98,103.

Liu X Y, Pan L, Shuai M R. Prediction model and its application of BP neural network rolling force based on Matlab[J].Journal of Chongqing University of Science and Technology: Natural Sciences Edition,2016,18(6):96-98,103.

[2]杨景明,顾佳琪,闫晓莹,等.基于改进遗传算法优化BP网络的轧制力预测研究[J].矿冶工程,2015,35(1):111-115.

Yang J M, Gu J Q, Yan X Y, et al. Rolling force prediction based on BP network optimized by an improved genetic algorithm[J].Mining and Metallurgical Engineering,2015,35(1):111-115.

[3]马臣,李慕勤,闫振林.利用BP人工神经网络实现对轧机轧制力的预测[J].佳木斯大学学报：自然科学版,2008,26(6):785-788.

Ma C, Li M Q, Yan Z L. Mill rolling force prediction using BP artificial neural network [J]. Journal of Jiamusi University: Natural Science Edition, 2008, 26(6): 785-788.

[4]王智,张果,王剑平,等.基于PSOBP神经网络双机架炉卷轧机轧制力的预测[J].钢铁研究,2017,45(3):23-26.

Wang Z, Zhang G, Wang J P, et al. Prediction of rolling force for twostand steckel mill based on PSOBP neural network[J].Journal of Iron and Steel Research,2017,45(3):23-26.

[5]沈花玉, 王兆霞, 高成耀,等. BP神经网络隐含层单元数的确定[J]. 天津理工大学学报, 2008,24(5):13-15.

Shen H Y, Wang Z X, Gao C Y, et al. Determining the number of BP neural network hidden layer units [J]. Journal of Tianjin University of Technology, 2008, 24 (5): 13-15.

[6]刘莉,李传峰,张广军.基于BP神经网络斜轧穿孔轧制力的预测[J].山东冶金,2013,35(2):43-44.

Liu L, Li C F, Zhang G J. Rolling force prediction of rotary piercing based on BP neural network [J]. Shandong Metallurgy, 2013,35 (2): 43-44.

[7]Hu X L, Wang Z D, Yu J M, et al. Prediction of rolling load by BP neural networks integrating with selfadaption of traditional model[J]. Journal of Northeastern University, 2002, 23(11):1089-1092.

服务与反馈：

【文章下载】【加入收藏】