Transactions of Materials and Heat Treatment

Title：Orthogonal test analysis and neural network prediction on forming defects in hydroforming for tailor rolled blank box part

Authors： Zhang Huawei Wang Yongzhe Wu Jialu Wang Xingang

Unit： Guangdong University of Petrochemical Technology Northeastern University

KeyWords： tailor rolled blanks hydroforming box part thickness thinning morement of transition area neural network model

ClassificationCode：TG386.3

year,vol(issue):pagenumber：2022,47(10):96-102

Abstract：

The hydroforming process of tailor rolled blank(TRB) box part was simulated, and the influencing laws of forming parameters on the forming defects of TRB and the optimal parameters combination were obtained by the orthogonal test. Then, on this basis, the BP neural network model was established to predict the forming defects of TRB box part in hydroforming. The results indicate that for the maximum thinning rate, the order of the influence degree of each factor is friction coefficient＞liquid pressure on thicker side> blank holder force(BHF) on thicker side>ratio of liquid pressure on thinner side to thicker side>ratio of BHF on thinner side to thicker side. For the movement of transition area at the bottom of TRB and the movement of transition area at the flange, the order of the influence degree of each factor is BHF on thicker side>ratio of BHF on thinner side to thicker side>liquid pressure on thicker side>ratio of liquid pressure on thinner side to thicker side>friction coefficient. Taking into account the thinning rate and the movement of transition area, the optimal parameters combination is the BHF on thicker side of 20 kN, the ratio of BHF on thinner side to thicker side of 1.5, the liquid pressure on thicker side of 0.5 MPa, the ratio of liquid pressure on thinner side to thicker side of 2.0, and the friction coefficient of 0.200. Thus, the BP neural network model based on the orthogonal test analysis results can precisely predict the forming defects of TRB box part in hydroforming.

Funds：

国家自然科学基金资助项目(51475086)；广东石油化工学院校级科研基金项目(2020rc020)；茂名市科技计划立项项目（2022025）

张华伟(1983-)，男，博士，副教授,E-mail：zhanghw@neuq.edu.cn

Reference：

[1]张思佳，刘相华，刘立忠. 轧制差厚板变厚度区的应力应变关系表征[J]. 机械工程学报,2018, 54(18): 49-54.

Zhang S J, Liu X H, Liu L Z. Characterization of stressstrain relationship of tailor rolled blank′s thickness transition zone[J]. Journal of Mechanical Engineering,2018, 54(18): 49-54.

[2]Zhang H W, Wu J L, Wang X G. Crack defect of tailor rolled blank in deep drawing process[J]. Journal of Iron and Steel Research International, 2018, 25(12): 1237-1243.

[3]Han S W, Hwang T W, Oh I Y, et al. Manufacturing of tailorrolled blanks with thickness variations in both the longitudinal and latitudinal directions[J]. Journal of Materials Processing Technology, 2018, 256: 172-182.

[4]Palumbo G, Zhang S H, Tricarico L, et al. Numerical/experimental investigations for enhancing the sheet hydroforming process[J]. International Journal of Machine Tools & Manufacture,2006, 46(11): 1212-1221.

[5]Nakamura K, Nakagawa T, Amino H. Various application of hydraulic counter pressure deep drawing[J]. Journal of Materials Processing Technology,2010, 71(1): 160-167.

[6]汪建敏，王健，朱先忠，等. 差厚拼焊板充液拉深焊缝移动及厚度的研究[J]. 热加工工艺,2011, 40(17): 118-120,123.

Wang J M, Wang J, Zhu X Z, et al. Research on weldline movement and thickness of different thickness twbs in hydroforming deep drawing[J]. Hot Working Technology,2011, 40(17): 118-120,123.

[7]石磊. 拼焊板在径向辅助压力下充液拉深工艺及数值模拟研究[D]. 镇江: 江苏大学, 2010.

Shi L. Numerical Simulation of Tailor Welded Blank in Hydromechanical Deep Drawing with Independent Radial Hydraulic Pressure[D]. Zhenjiang: Jiangsu University, 2010.

[8]Krux R, Homberg W, Kleiner M. Properties of largescale structure workpieces in highpressure sheet metal forming of tailor rolled blanks[J]. Steel Research International,2005, 76(12): 890-896.

[9]Kleiner M, Homberg W, Krux R. Highpressure sheet metal forming of large scale structures from sheets with optimized thickness distribution[J]. Steel Research International,2005, 76(2-3): 177-181.

[10]Urban M, Krahn M, Hirt G, et al. Numerical research and optimisation of high pressure sheet metal forming of tailor rolled blanks[J]. Journal of Materials Processing Technology,2006, 177(1-3): 360-363.

[11]Van Putten K, Urban M, Kopp R. Computer aided product optimization of highpressure sheet metal formed tailor rolled blanks[J]. Steel Research International,2005, 76(12): 897-904.

[12]张渝，顾栩，巫洪亮，等. 过渡区参数TRB管液压胀形性能的影响及预测[J]. 锻压技术,2017, 42(11): 99-104.

Zhang Y, Gu X, Wu H L, et al. Influence and prediction of transition zone parameters on hydraulic bulging properties for TRB tube[J]. Forging & Stamping Technology,2017, 42(11): 99-104.

[13]王智，谢延敏，胡静，等. 基于改进灰色神经网络模型的板料成形缺陷预测研究[J]. 中国机械工程,2013, 24(22): 3075-3079.

Wang Z, Xie Y M, Hu J, et al. Research on defect prediction in steel metal forming based on improved gray neural network model[J]. China Mechanical Engineering, 2013, 24(22): 3075-3079.

[14]张华伟, 郑晓涛. 基于遗传算法优化神经网络的拼焊板压边力预测[J]. 东北大学学报:自然科学版, 2020, 41(2): 241-245.

Zhang H W, Zheng X T. Blank holder force prediction of tailor welded blank based on neural network optimized by genetic algorithm[J]. Journal of Northeastern University: Natural Science, 2020, 41(2): 241-245.

[15]Zhang H W, Liu X H, Liu L Z, et al. Study on nonuniform deformation of tailor rolled blank during uniaxial tension[J]. Acta Metallurgica Sinica: English Letters, 2015, 28(9): 1198-1204.

[16]姜银方，王飞，李新城，等. 基于正交试验和神经网络的激光拼焊板回弹预测[J]. 塑性工程学报, 2009, 16(3): 40-44.

Jiang Y F, Wang F, Li X C, et al. Study on the springback prediction in laser TWBs forming based on orthogonal experiment and neural network[J]. Journal of Plasticity Engineering, 2009, 16(3): 40-44.

[17]刘伟，杨玉英. 基于FEA的板料成形工艺优化及评价函数研究[J]. 材料科学与工艺, 2006, 14(2): 159-161.

Liu W, Yang Y Y. Study on process optimization and objective functions of sheet metal forming based on FEA[J]. Materials Science & Technology, 2006, 14(2): 159-161.

[18]林忠钦，刘呈，李淑慧. 应用正交试验设计提高U形件的成形精度[J]. 机械工程学报, 2002, 38(3): 83-89.

Lin Z Q, Liu C, Li S H. Application orthogonal experiment design in increasing dimensional accuracy of Ushaped parts[J]. Journal of Mechanical Engineering, 2002, 38(3): 83-89.

[19]官英平，马向东，张登谦. 拼焊板方盒件拉深成形过程变压边力曲线预测[J]. 塑性工程学报, 2017, 24(2): 17-21.

Guan Y P, Ma X D, Zhang D Q. Variable blank holder force curve prediction in deep drawing for square box of tailorwelded blanks[J]. Journal of Plasticity Engineering, 2017, 24(2): 17-21.

[20]李捷菲. 基于BP神经网络的PID控制系统研究与设计[D]. 长春: 吉林大学, 2019.

Li J F. Research and Design of PID Control System Based on BP Neural Network[D]. Changchun: Jilin University, 2019.

Service：

【This site has not yet opened Download Service】【Add Favorite】