锻压技术

基于BP人工神经网络的板料自冲铆接的底切量仿真预测

英文标题：Simulation prediction on undercut for selfpiercing riveting of sheet based on BP artificial neural network

作者：周琦

单位：江阴职业技术学院

分类号：TG386

出版年,卷(期):页码：2019,44(8):61-65

摘要：

以AlMg3铝合金板料的自冲铆接作为研究对象，利用DEFORM2D有限元软件与正交试验，对两层2 mm的AlMg3铝合金板料的自冲铆接过程进行有限元仿真，基于BP人工神经网络及仿真数据，对板料自冲铆接的成形质量进行仿真预测。以模具深度、模具凸台高度、模具宽度以及铆接速度为输入层，将底切量作为输出层，建立了4111的3层BP人工神经网络。以仿真结果为样本进行反复的训练，得到BP人工神经网络的预测值，与有限元仿真值相比两者的最大误差为0.97%。此外，借助自冲铆接设备及安装模具，进行自冲铆接试验，试验值与BP人工神经网络预测值之间的相对误差为6.57%，验证了BP人工神经网络应用于板料自冲铆接成形质量预测的正确性与可靠性，为企业的实际生产提供重要的参考。

For the self-piercing riveting of AlMg3 aluminum alloy sheet, the self-piercing riveting process of two layers of AlMg3 aluminum alloy sheet with thickness of 2 mm was simulated by DEFORM-2D and orthogonal experiment, and the forming quality of self-piercing riveting for sheet was predicted based on BP artifical neural network and simulation results. Then, the die depth, height of die boss, die width and riveting speed were taken as the input layer, the under-cut was taken as the output layer, and the BP artificial neural network with three layers of 4-11-1 was established. The predictive value of BP artifical artificial neural network were obtained by using the simulation results as samples for repeated training. Compared with the finite element simulation results, the maximum error between the two methods was 0.97%. In addition, the self-piercing riveting tests were conducted by the self-piercing riveting equipment and die, and the relative error between prediction value of BP artifical neural network and experimental value was 6.57%. Therefore, the correctness and reliability of BP artifical neural network applied to the prediction on the forming quality of self-piercing riveting for sheet are verified which is an important reference on the actual production of enterprises.

基金项目：

作者简介：周琦（1981-），男，硕士，讲师 E-mail：632992423@qq.com

参考文献：

[1]Van Hall S N, Findley K O, Freis A K. Improved selfpierce rivet performance through intentional decarburization[J]. Journal of Materials Processing Technology, 2018, 251: 350-359.

[2]杜国栋, 邢彦锋, 金光灿, 等. 铝钢薄板多铆钉连接变形分析及工艺优化[J]. 热加工工艺, 2018, 47(21): 183-186.

Du G D, Xing Y F, Jin G C, et al. Deformation analysis and process optimization of multirivet connection in aluminum steel sheet[J]. Hot Working Technology, 2018, 47(21): 183-186.

[3]丁文有, 何晓聪, 邢保英，等. 铆向组合对双铆钉自冲铆接头力学性能的影响[J]. 塑性工程学报, 2018, 25(5): 234-239.

Ding W Y, He X C, Xing B Y, et al. Effect of riveting orientation combination on mechanical properties of double rivet selfpiercing riveted joint [J]. Journal of Plasticity Engineering, 2018, 25(5): 234-239.

[4]吴小丹, 王敏, 孔谅, 等. 铆速对SPR自冲铆接接头成形性能的影响[J]. 机械设计与制造, 2017, 2(2): 210-212.

Wu X D, Wang M, Kong L, et al. Influence of punch velocity on formability of selfpiercing riveted joints[J]. Machinery Design & Manufacture, 2017, 2(2): 210-212.

[5]Huang L, Guo H, Shi Y, et al. Fatigue behavior and modeling of selfpiercing riveted joints in aluminum alloy 6111[J]. International Journal of Fatigue, 2017, 100: 274-284.

[6]张学奇, 董万鹏. 半空心铆钉自冲铆接的研究进展[J]. 热加工工艺, 2016, 45(9):5-8.

Zhang X Q, Dong W P. Study on selfpiercing riveting of semihollow rivet[J]. Hot Working Technology, 2016, 45(9): 5-8.

[7]张永强, 伊日贵, 付参, 等. 钢铝自冲铆接工艺过程仿真与实验研究[J]. 电焊机, 2018, 48(10): 26-29.

Zhang Y Q, Yi R G, Fu C, et al. Simulation and experiments on self piercing riveting process of steel and aluminum[J]. Electric Welding Machine, 2018, 48(10): 26-29.

[8]张良. 基于BP神经网络的预切冲裁断面质量的仿真预测[J]. 锻压技术, 2018, 43(12): 175-179.

Zhang L. Simulation and prediction of cross section quality for precut blanking based on BP neural network[J]. Forging & Stamping Technology, 2018, 43(12): 175-179.

[9]GB/T 228.1—2010, 金属材料拉伸试验第1部分：室温试验方法[S].

GB/T 228.1—2010，Metallic materials—Tensile testing—Part 1:Method of test at room temperature[S].

[10]甘守武, 陈志军, 赵磊娜. 基于单隐藏层 BP 神经网络Ti55531 合金流变应力预测模型的建立[J]. 热加工工艺, 2018,47(11):59-61.

Gan S W, Chen Z J, Zhao L N. Construction of flow stress prediction model for Ti55531 alloy based on BP neural network with single hidden layer[J]. Hot Working Technology, 2018, 47(11):59-61.

[11]林伟路, 丁小凤, 双远华. BP神经网络对斜轧穿孔轧制力的预测[J]. 锻压技术, 2018,43(10):175-178.

Lin W L, Ding X F, Shuang Y H. Prediction on rolling force of oblique rolling piercing based on BP neural network[J]. Forging & Stamping Technology, 2018, 43(10):175-178.

[12]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.

服务与反馈：

【文章下载】【加入收藏】