锻压技术

基于BP神经网络的温挤压模具磨损量预测

英文标题：Prediction on wear loss of warm extrusion die based on BP neural network

作者：张涛樊文欣郭代峰史永鹏

单位：中北大学

分类号：TG376

出版年,卷(期):页码：2017,42(2):178-182

摘要：

连杆衬套毛坯在生产过程中会出现凸模磨损严重。根据连杆衬套毛坯的温挤压成形原理和加工成形特点，得到了影响温挤压凸模磨损寿命的4个主要因素，即模具初始硬度、摩擦系数、挤压速度和模具预热温度。以凸模磨损量最小为目标，设计了4因素3水平标准正交试验表。利用Archard磨损理论，通过Deform-3D软件，进行了温挤压磨损正交模拟试验。基于试验数据，建立了4-15-1的3层BP神经网络预测模型，得到预测值和数值模拟值误差小于3%，此方法可以用于快速预测温挤压模具的磨损量。

For punch badly worn in the process of connecting rod bushing blank production, according to the warm extrusion principle and processing characteristics, four major factors influencing the wear life of warm extrusion punch were obtained, namely die initial hardness, friction coefficient, extrusion speed and pre-heating temperature．Then, the standard orthogonal experiment table with four factors and three levels was designed to minimize punch wear loss. Furthermore, based on the theory of Archard wear, the orthogonal simulation experiments of warm extrusion punch wear were executed by software Deform-3D. Finally, according to data from the experiment, three-layer BP neural network predicted model of 4-15-1 was established, and the error between the predicted value and the numerical simulation value was less than 3%. Therefore, the above method could predict the wear loss of warm extrusion die quickly.

基金项目：

山西省自然科学基金资助项目(2012011023-2);山西省高校高新技术产业化项目（20120021）

张涛(1992-)，男，硕士研究生樊文欣(1964-)，男，博士，教授

参考文献：

［1］樊文欣，张涛，宋河金，等. 强力旋压加工的铜合金连杆衬套[J]. 车用发动机，1997,（2）:32-35.Fan W X，Zhang T，Song H J，et al. A connecting rod bushing of high speed diesel engines by the powerful swivel press technique [J]. Vehicle Engine, 1997,（2）:32-35.
［2］Kang J H, Park I W, Jae J S, et al. A study on a die wear model considering thermal softening (I): Construction of the wear model [J]. Journal of Materials Processing Technology, 1999, 96:53-58.
［3］龚小涛，杨帆，陈荷，等.基于Archard模型的热摆辗模具磨损研究[J]. 锻压技术,2015,40(10): 119-121.Gong X T，Yang F，Chen H, et al. Research on wear of hot rotary forging die based on Achard's model[J]. Forging & Stamping Technology, 2015, 40(10):119-121.
［4］孙宪萍，刘强强，杨兵，等.基于磨损正交试验的温挤压模具优化设计[J]. 润滑与密封，2016，41(6):73-76.Song X P，Liu Q Q，Yang B, et al. Optimization design on warm extrusion die based on orthogonal experiments on wear [J].Lubrication Engineering, 2016, 41(6):73-76.
［5］李伟伟，余心宏.GH4910合金反挤压成形模具磨损[J]. 航空材料学报,2016, 36(1): 12-17.Li W W，Yu X H. Research on wear of GH4910 alloy backward extrusion forming die [J].Journal of Aeronautical Materials, 2016, 36(1):12-17.
［6］佘勇，樊文欣，陈东宝，等.基于RBF神经网络的强力旋压连杆衬套力学性能预测能力[J].锻压技术,2016,41(6):128-132.She Y，Fan W X，Chen D B，et al. Study on mechanical property prediction of power spinning connecting rod bushing based on RBF neural network [J].Forging & Stamping Technology,2016,41(6):128-132.
［7］胡建军，李小平.DEFORM-3D塑性成形CAE应用教程[M]. 北京：北京大学出版社,2011. Hu J J，Li X P. DEFORM-3D Plastic Forming CAE Application Tutorial [M].Beijing: Peking University Press, 2011.
［8］谢晖，凌鸿伟. 基于Archard理论的热冲压模具磨损分析及优化[J]. 热加工工艺,2016,45(1):100-104.Xie H，Ling H W. Analysis and optimization of hot stamping die wear based on Archard theory[J].Hot Working Technology,2016,45(1):100-104.
［9］王惠梅.40Cr长轴件的温挤压成形工艺研究[D].哈尔滨：哈尔滨工业大学，2010.Wang H M. The Warm Extrusion Forming Technology of a 40Cr Steel Long Shaft Part [D].Harbin: Harbin Institute of Technology, 2010.
［10］周杰，赵军，安治国.挤压模磨损规律及磨损对模具寿命的影响[J]. 中国机械工程,2007,18(17):2112-2115.Zhou J，Zhao J，An Z G. Wear rule and effect on the die service life during hot extrusion [J].China Mechanical Engineering,2007, 18(17):2112-2115.
［11］王小川，史峰，郁磊，等.MATLAB神经网络43个案例分析[M]. 北京：北京航空航天大学出版社,2013.Wang X C，Shi F，Yu L，et al. 43 Cases Analysis of MATLAB Neural Network [M].Beijing: Beihang University Press, 2013.
［12］曹存存，樊文欣，杨华龙. 基于BP神经网络的连杆衬套磨损量预测[J]. 组合机床与自动化加工技术,2016,(8):50-53.Cao C C，Fan W X，Yang H L. Prediction of wear of connecting rod bushing based on BP neural network[J].Modular Machine Tool & Automatic Manufacturing Technique,2016,(8):50-53.

服务与反馈：

【文章下载】【加入收藏】