锻压技术

基于改进的BP神经网络无缝钢管连轧轧制力的预测

英文标题：Prediction on rolling force in continuous rolling of seamless steel pipe based on improved BP neural network

作者：张坚双远华胡建华穆佳浩赵铁林

分类号：TP183

出版年,卷(期):页码：2022,47(5):153-160

摘要：

无缝钢管连轧过程具有多变量、强耦合、非线性等特点，传统的数学模型无法对一些参数进行精确地预测。为了提高连轧过程中轧制力预测的精度，采用改进初始值选取方法来优化BP神经网络，建立改进的BP神经网络的轧制力预测模型。首先，采集某钢厂历史生产数据，进行预处理，通过灰色关联度确定影响轧制力的主要因素；然后，对初始值进行设置，利用MATLAB编写仿真程序对连轧机组轧制力进行预测。结果表明：基于改进的BP神经网络的轧制力预测模型具有很强的学习能力和表达能力，轧制力预测精度得到了很大的提高，对实际的生产具有重要意义。

The seamless steel pipe rolling process has the characteristics of multivariable, strong coupling and nonlinear, and the traditional mathematical model can not predict some parameters accurately. Therefore, in order to improve the accuracy of rolling force prediction during rolling process, BP neural network was optimized by the improved initial value selection method, and the improved BP neural network rolling force prediction model was established. First, the historical production data of a steel mill were collected and preprocessed, and the main factors affecting rolling force were determined by grey correlation degree. Then, the initial value was set, and the simulation program was written by MATLAB to predict the rolling force of continuous rolling mill. The results show that the rolling force prediction model based on the improved BP neural network has strong learning ability and expression ability, and the rolling force prediction accuracy is greatly improved, which is of great significance to the actual production.

基金项目：

山西省科技重大专项（20191102009）；山西省重点研发项目（201903D121049）

作者简介：张坚（1994-），男，硕士研究生，E-mail：S20190365@stu.tysut.edu.cn；通信作者：双远华（1962-），男，博士，教授，E-mail：yhshuang@tysut.edu.cn

参考文献：

[1]肖成. 无缝钢管连轧过程建模与轧制规程优化[D].长沙：中南大学, 2014.

Xiao C. Modeling and Rolling Schedule Optimization of Seamless Steel Tube Rolling Process [D].Changsha: Central South University,2014.

[2]汤涛. 无缝钢管连轧过程数值模拟及参数优化[D].衡阳：南华大学，2013.

Tang T. Numerical Simulation and Parameter Optitmzat-ion for Tandem Rolling Process of Seamless Steel Tube [D]. Hengyang：University of South China， 2013.

[3]孟代江. 人工神经网络技术及其应用[J]. 电子技术与软件工程, 2016, (23):16.

Meng D J. Artificial neural network technology and its application [J]. Electronic Technology & Software Engineering,2016, (23):16.

[4]杨源杰, 黄道.人工神经网络算法研究及应用[J].华东理工大学学报:自然科学版，2002，28（5）：551-554.

Yang Y J, Huang D.Research and application of artificial neural networks [J].Journal of East China University of Science and Technology:Natural Science Edition, 2002,28(5):551-554.

[5]刘翰培, 汪宇轩,王亚琴,等.深度神经网络轧制力建模及其并行优化研究[J/OL].控制工程:1-8[2021-03-24].DOI:10.14107/j.cnki.kzgc.20200277.

Liu H P, Wang Y X, Wang Y Q, et al.Research on rolling force modeling and parallel optimization of deep neural network [J/OL] Control Engineering of China: 1-8[2021-03-24].DOI:10.14107/j.cnki.kzgc.20200277.

[6]马威, 李维刚,赵云涛,等.基于深度学习的热连轧轧制力预测[J].钢铁研究学报，2019，31（9）:805-815.

Ma W, Li W G, Zhao Y T, et al. Prediction of hot-rolled roll force based on deep learning [J]. Journal of Iron and Steel Research，2019，31 (9):805-815.

[7]Bagheripoor M, Bisadi H . Application of artificial neural networks for the prediction of roll force and roll torque in hot strip rolling process[J]. Applied Mathematical Modelling, 2013, 37(7):4593-4607.

[8]陈鑫, 朱明杰,吴敏,等.结合机理计算与神经网络预测的无缝钢管轧制力建模[J].冶金自动化,2015,39(4):32-37.

Chen X, Zhu M J, Wu M, et al. Rolling force modeling for seamless steel pipe combining mechanism model and neural network prediction [J]. Metallurgical Industry Automation, 2015,39(4):32-37.

[9]宋勇, 苏岚,荆丰伟,等.热轧带钢轧制力模型自学习算法优化[J].北京科技大学学报，2010，32（6）：802-806.

Song Y, Su L, Jing F W, et al. Self - learning algorithm optimization for the rolling force model of hot strips [J].Journal of University of Science and Technology Beijing, 2010,32(6)：802-806.

[10]周富强, 曹建国,张杰,等.基于神经网络的冷连轧机轧制力预报模型[J].中南大学学报:自然科学版,2006,37(6):1155-1160.

Zhou F Q, Cao J G, Zhang J, et al.Prediction model of rolling force for tandem cold rolling mill based on neural networks and mathematical models [J].Journal of Central South University: Science and Technology,2006，37(6):1155-1160.

[11]郭立伟, 杨荃,郭磊.冷连轧过程控制轧制力模型综合参数自适应[J].北京科技大学学报，2007,29（4）：413-416.

Guo L W, Yang Q, Guo L. Comprehensive parameters self-adapting for a rolling force model of tandem cold rolling process control [J]. Journal of University of Science and Technology Beijing,2007,29(4):413-416.

[12]何飞, 石露露,黎敏,等.基于多模态和加权支持向量机的热轧轧制力智能预报[J].工程科学学报,2015,37(4):517-521.

He F, Shi L L, Li M, et al.Intelligent prediction of rolling forces in hot rolling based on a multi-modal and weighted support vector machine [J].Chinese Journal of Engineering, 2015,37(4):517-521.

[13]曹建国，张杰，张少军.轧钢设备及自动控制[M].北京：化学工业出版社，2010.

Cao J G, Zhang J, Zhang S J. Rolling Equipment and Automatic Control[M]. Beijing: Chemical Industry Press,2010.

[14]胡贤磊, 王昭东,于解民,等.结合模型自学习的BP神经元网络的轧制力预报[J].东北大学学报:自然科学版,2002，23(11):1089-1092.

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:Natural Science,2002,23(11):1089-1092.

[15]朱其萍, 徐红玉,王晓强,等.基于PSO-BP的超声滚挤压轴承套圈表面加工硬化程度预测[J].锻压技术,2021,46(11): 190-196.

Zhu Q P, Xu H Y, Wang X Q, et al. Prediction on degree of work hardening for surface of bearing ring by ultrasonic rolling extrusion based on PSO-BP[J].Forging & Stamping Technology,2021,46 (11): 190-196.

服务与反馈：

【本网站尚未开通全文下载服务】【加入收藏】