锻压技术

基于BP人工神经网络的油箱端盖拉深成形仿真预测

英文标题：Prediction on deep drawing of fuel-tank end cap based on BP artificial neural network

作者：李兵姜海林刘奎武高鹏

分类号：TG385

出版年,卷(期):页码：2017,42(11):177-180

摘要：

以油箱端盖作为分析对象，借助DYNAFORM仿真软件，对油箱端盖的拉深成形过程进行数值模拟，并通过拉深成形试验验证可知，板料最大减薄率与最大增厚率的试验值与模拟值之间的相对误差分别为9.26%与8.32%，验证了有限元模型的正确性。结合正交试验，进行有限元仿真试验的设计，基于BP人工神经网络，对板料的成形质量进行仿真预测。选择冲压速度、模具间隙以及压边力作为输入层，将板料成形的最大减薄率作为输出层，建立了3-11-1的3层BP人工神经网络。通过BP人工神经网络的训练与测试得知：BP人工神经网络仿真预测值与数值模拟值之间的相对误差为2.15%，验证了BP人工神经网络应用于油箱端盖拉深成形质量仿真预测的正确性。

For fuel-tank end cap, the deep drawing process of fuel-tank end cap was simulated by software DYNAFORM. The deep drawing test indicates that the relative errors between experiment value and simulation value of the maximum thinning ratio and the maximum thickening rate of sheet metal are 9.26% and 8.32% respectively, and the correctness of finite element model is verified. Then, the design of finite element simulation test was carried out by orthogonal test, and the quality of sheet metal forming was predicted based on BP artificial neural network. Furthermore, three layers of 3-11-1 for BP artificial neural network were established with the input layer of stamping speed, die clearance and blank holder force and the output layer of the maximum thinning rate of sheet metal forming. The error between predicted value of BP artificial neural network and numerical simulation value is 2.15% by the training and testing of BP artificial neural network. Therefore, the accuracy of BP artificial neural network applied to deep drawing of fuel-tank end cap was verified.

基金项目：

淮安市重点研发计划(工业及信息化)(HAG201614)

作者简介：李兵（1983-），男，硕士，讲师 E-mail：spxylb@163.com

参考文献：

[1]马国英, 黄彬兵, 苏春建, 等. 汽车翼子板拉深成形模拟及工艺参数优化[J]. 锻压技术, 2015, 40（3）: 21-24.

Ma G Y, Huang B B, Su C J, et al. Simulation and parameters optimization of deep drawing for automobile fender[J]. Forging & Stamping Technology, 2015, 40（3）: 21-24. 

[2] 朱梅云, 傅建, 崔礼春. 基于伺服技术的板材冲压成形性研究[J]. 锻压技术, 2014, 39（10）: 43-46.

Zhu M Y, Fu J, Cui L C. Research on the formability of steel plate based on servo technology[J]. Forging & Stamping Technology, 2014, 39（10）: 43-46. 

[3] 刘驰. 侧围外板冲压成形仿真技术与应用[J]. 锻压技术, 2014, 39（1）: 41-45.

Liu C. Simulation and application of stamping process for side-frame outer panel[J]. Forging & Stamping Technology, 2014, 39（1）: 41-45. 

[4] 林策, 彭艳, 孙建亮, 等. 板形缺陷板料冲压变形及回弹仿真分析[J]. 锻压技术, 2012, 37（6）: 174-178.

Lin C, Peng Y, Sun J L, et al. Stamping deformation and springback simulation analysis of sheet metal with shape defects[J]. Forging & Stamping Technology, 2012, 37（6）: 174-178. 

[5] 宫晓峰, 于仁萍. 基于Autoform汽车后围上盖板拉延成形模拟应用[J]. 锻压技术, 2014, 39（4）: 149-153.

Gong X F, Yu R P. Application of drawing simulation for automobile reinforcement upper back based on Autoform[J]. Forging & Stamping Technology, 2014, 39（4）: 149-153. 

[6] 张涛, 樊文欣, 郭代峰, 等. 基于BP神经网络的温挤压模具磨损量预测[J]. 锻压技术, 2017, 42（2）: 178-182.

Zhang T, Fan W X, Guo D F, et al. Prediction on wear loss of warm extrusion die based on BP neural network[J]. Forging & Stamping Technology, 2017, 42（2）: 178-182. 

[7] 王晓莉, 穆瑞, 张咏琴. 基于BP神经网络的薄板成形回弹仿真预测[J]. 锻压技术, 2016, 41（6）: 146-149.

Wang X L, Mu R, Zhang Y Q. Numerical prediction of springback in sheet metal forming based on BP neural network[J]. Forging & Stamping Technology, 2016, 41（6）: 146-149. 

[8] 庞敬礼. 基于BP神经网络的传力片冲裁凸模磨损仿真预测[J]. 锻压技术, 2016, 41（11）: 114-117.

Pang J L. Simulation prediction on blanking punch wear of leaf spring based on BP neural network[J]. Forging & Stamping Technology, 2016, 41（11）: 114-117.

服务与反馈：

【文章下载】【加入收藏】