[成形过程的数据挖掘与深度学习方法]基于机器学习的管材弯曲回弹有效预测与补偿

doi:10.3969/j.issn.1004-132X.2020.22.012

摘要/Abstract

摘要： 采用基于优化的误差反向传播(BP)神经网络的机器学习算法建模，提出了考虑材料参数、几何参数等多因素的弯管回弹精确预测和高效控制方法。该方法通过引入非线性惯性权重及遗传算法的杂交算子，改进了粒子群优化(PSO)算法，进而通过改进的PSO算法对BP神经网络进行优化，构建了基于改进的PSO-BP神经网络机器学习回弹预测和补偿模型。以多种规格的铝合金数控弯管构件为对象，将实际生产中不同规格、批次、成形参数下回弹数据作为训练样本，实现了所建机器学习预测模型的应用验证。所建模型获得的预测结果平均相对误差为6.3%，与未优化的BP神经网络等传统模型相比，预测精度最大提高了18.5%，计算时间可从1.5 h缩短至300 s，同时实现了回弹预测与补偿精度以及计算效率的显著提高。

关键词: 回弹, 管材弯曲, 机器学习, 成形精度

Abstract: The machine learning algorithm modeling was adopted based on the optimized back propagation(BP) neural network and the precise prediction and efficient control method of bend springbacks was proposed. In this method, the particle swarm optimization(PSO) algorithm was improved by introducing the nonlinear inertia weight and hybrid operator of genetic algorithm, and then the BP neural network was optimized by the improved PSO algorithm, and the machine learning springback prediction and compensation model was constructed based on the improved PSO-BP neural network. Based on the springback data of different specifications, batches，and forming parameters in the actual productions, the applications of the machine learning prediction model were verified. The average relative error of the prediction results obtained by the model is as 6.3%. Compared with the traditional models, the prediction accuracy is increased by 18.5% at most, and the calculation time may be reduced from 1.5 h to 300 s. The prediction and compensation accuracy of springback and the calculation efficiency are improved significantly.

Key words: springback, tube bending, machine learning, forming accuracy

中图分类号:

TG386

陈光耀, 李恒, 贺子芮, 马俊, 李光俊, 付颖. [成形过程的数据挖掘与深度学习方法]基于机器学习的管材弯曲回弹有效预测与补偿[J]. 中国机械工程, DOI: 10.3969/j.issn.1004-132X.2020.22.012.

CHEN Guangyao, LI Heng, HE Zirui, MA Jun, LI Guangjun, FU Ying. Effective Prediction and Compensation of Springbacks for Tube Bending Using Machine Learning Approach[J]. China Mechanical Engineering, DOI: 10.3969/j.issn.1004-132X.2020.22.012.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.cmemo.org.cn/CN/10.3969/j.issn.1004-132X.2020.22.012

http://www.cmemo.org.cn/CN/Y2020/V31/I22/2745

参考文献

［1］李恒, 杨合, 宋飞飞, 等. 材料性能波动下TA18钛管绕弯回弹行为［J］. 稀有金属材料与工程, 2014(1):64-71.
LI Heng, YANG He, SONG Feifei, et al.Bending Resilience Behavior of TA18 Titanium Tube under Material Property Fluctuation ［J］. Rare Metal Materials and Engineering, 2014(1):64-71.
［2］许旭东, 李光俊. 飞机导管数字化生产线探讨［J］. 航空制造技术, 2005(9):74-76.
XU Xudong, LI Guangjun. Discussion on Digital Production Line of Aircraft Conduit ［J］. Aviation Manufacturing Technology, 2005(9):74-76.
［3］韩志仁, 吕彦盈, 刘宝明, 等. 飞机焊接导管数字化制造技术研究［J］. 航空制造技术, 2017, 60(8):95-98.
HAN Zhiren, LYU Yanying, LIU Baoming, et al. Research on Digital Manufacturing Technology of Aircraft Welding Pipe［J］. Aviation Manufacturing Technology, 2017, 60(8):95-98.
［4］詹梅, 杨合, 栗振斌. 管材数控弯曲回弹规律的有限元分析［J］. 材料科学与工艺, 2004(4):14-17.
ZHAN Mei, YANG He, LI Zhenbin. Finite Element Analysis of Springback Law in Numerical Control Bending of Pipes ［J］. Materials Science and Technology, 2004 (4):14-17.
［5］宋飞飞, 杨合, 李恒, 等. TA18高强钛管数控弯曲回弹工艺参数影响的显著性分析［J］. 稀有金属材料与工程, 2013, 42(1):43-48.
SONG Feifei, YANG He, LI Heng, et al. Significant Analysis of the Influence of the Technological Parameters of the Numerical Control Bending Rebound of TA18 High Strength Titanium Tube ［J］. Rare Metal Materials and Engineering, 2013, 42 (1):43-48.
［6］ZHOU Yingping, LI Pengfei, LI Mingzhe, et al. Residual Stress and Springback Analysis for 304 Stainless Steel Tubes in Flexible-bending Process［J］. International Journal of Advanced Manufacturing Technology, 2018, 94(1):1-9.
［7］YANG He, LI Heng, ZHANG Zhiyong, et al. Advances and Trends on Tube Bending Forming Technologies［J］. Chinese Journal of Aeronautics, 2012, 25(1):1-12.
［8］黄卫东, 陈保国, 张卫红, 等. 民用飞机构件先进成形技术［M］. 上海:上海交通大学出版社, 2016:412-418.
HUANG Weidong, CHEN Baoguo, ZHANG Weihong, et al. Advanced Forming Technology of Civil Aircraft Components［M］. Shanghai:Shanghai Jiao Tong University Press, 2016:412-418.
［9］LI Heng, MA Jun, LIU Biying, et al. An Insight into Neutral Layer Shifting in Tube Bending［J］. International Journal of Machine Tools and Manufacture, 2018, 126:51-70.
［10］ZHAN Mei, WANG Yan, YANG He, et al. An Analytic Model for Tube Bending Springback Considering Different Parameter Variations of Ti-alloy Tubes［J］. Journal of Materials Processing Technology, 2016, 236:123-137.
［11］JEONG H S, HA M Y, CHO J R. Theoretical and FE Analysis for Inconel 625 Fine Tube Bending to Predict Springback［J］. International Journal of Precision Engineering and Manufacturing, 2012, 13:2143-2148.
［12］LI Heng, YANG He, SONG Feifei, et al. Springback Characterization and Behaviors of High-strength Ti-3Al-2.5V Tube in Cold Rotary Draw Bending［J］. Journal of Materials Processing Technology, 2012, 212(9):1973-1987.
［13］MA Jun, YANG He, LI Heng, et al. Springback Prediction of Titanium Tube Bending Considering Bauschinger Effect and Youngs Modulus Variation［J］. Journal of Physics:Conference Series, 2016, 734:032113.
［14］SONG Feifei, YANG He, LI Heng, et al. Springback Prediction of Thick-walled High-strength Titanium Tube Bending［J］. Chinese Journal of Aeronautics, 2013, 26(5):1336-1345.
［15］LIAO Juan, XUE Xin, LEE M G ,et al. On Twist Springback Prediction of Asymmetric Tube in Rotary Draw Bending with Different Constitutive Models［J］. International Journal of Mechanical Sciences, 2014, 89:311-322.
［16］HASSANIN M F, SHOEB A M, HASSANIEN A E. Grey Wolf Optimizer-based Back-propagation Neural Network Algorithm［C］∥ International Computer Engineering Conference. Cairo, 2016:213-218.
［17］刘婧瑶, 唐承统, 宁汝新. 管材数控绕弯回弹实验研究及BP网络预测模型［J］. 塑性工程学报, 2009, 16(6):85-90.
LIU Jingyao, TANG Chengtong, NING Ruxin. Experimental Research and BP Network Prediction Model of Pipe Bending Springback［J］. Journal of Plastic Engineering, 2009, 16 (6):85-90.
［18］INAMDAR M, DATE P P, NARASIMHAN K, et al. Development of an Artificial Neural Network to Predict Springback in Air Vee Bending［J］. International Journal of Advanced Manufacturing Technology, 2000, 16(5):376-381.
［19］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 Xiaoli, MU Rui, ZHANG Yongqin. Springback Simulation Prediction of Sheet Metal Forming Based on BP Neural Network［J］. Forging technology, 2016, 41(6):146-149.
［20］HAN Fei, MO Jianhua, QI Hongwei, et al. Springback Prediction for Incremental Sheet Forming Based on FEM-PSONN Technology［J］. Transactions of Nonferrous Metals Society of China, 2013, 23(4):1061-1071.
［21］LIU Wenjuan, Liu Qiang, RUAN Feng, et al. Springback Prediction for Sheet Metal Forming Based on GA-ANN Technology［J］. Journal of Materials Processing Technology, 2007, 187:227-231.
［22］中国机械工程学会塑性工程学会. 锻压手册:第2卷冲压［M］. 3版. 北京:机械工业出版社, 2008:296-312.
China Society of Mechanical Engineering, Society of Plastic Engineering. Forging Manual:Volume 2 Stamping ［M］. 3rd ed. Beijing:China Machine Press, 2008:296-312.
［23］王博怀. 船用管件数控绕弯回弹及工艺研究［D］. 绵阳:西南科技大学, 2018.
WANG Bohuai. Research on Springback and Technology of Numerical Control Bending of Marine Pipe Fittings［D］. Mianyang:Southwest University of Science and Technology, 2018.
［24］DING Shifei, SU Chunyang, YU Junzhao. An Optimizing BP Neural Network Algorithm Based on Genetic Algorithm［J］. Artificial Intelligence Review, 2011, 36(2):153-162.
［25］BERGH F V D, ENGELBRECHT A P. A Study of Particle Swarm Optimization Particle Trajectories［J］. Information Sciences, 2006, 176(8):937-971.
［26］冯斌, 毛建中, 胡晖. 基于BP神经网络的条带刚凸特征回弹预测［J］.锻压技术, 2020, 45(3):20-26.
FENG Bin, MAO Jianzhong, HU Hui. Springback Prediction Based on BP Neural Network for Strip Gigid Convex Characteristics ［J］. Forging Technology, 2020, 45 (3):20-26.
［27］LIN Sen, WANG Guanglong, CHEN Yingjie, et al. Warehouse Environment Parameter Monitoring System and Sensor Error Correction Model Based on PSO-BP［J］. Transactions of Nanjing University of Aeronautics and Astronautics, 2017(3):109-116.
［28］BANSAL J C, SINGH P K, SARASWAT M, et al. Inertia Weight Strategies in Particle Swarm Optimization［C］∥ 2011 Third World Congress on Nature and Biologically Inspired Computing. Salamanca, 2011:633-640.
［29］CHEN Yonggang, LI Lixiang, XIAO Jinghua, et al. Particle Swarm Optimizer with Crossover Operation［J］. Engineering Applications of Artificial Intelligence, 2018, 70:159-169.

[1]	高嵩, 孙荧力, 李奇涵, 盈亮, 郝兆朋, 张邦成. 钛合金型材多点三维热拉弯成形工艺及其微观组织演化[J]. 中国机械工程, 2023, 34(24): 2986-2995,3014.
[2]	韩飞, 孙玮隆, 张若青. 不同弯曲方法对辊弯成形回弹的影响研究[J]. 中国机械工程, 2023, 34(19): 2353-2361.
[3]	陈昱, 项薇, 龚川. 基于数据挖掘的注塑产品质量在线故障检测及预测[J]. 中国机械工程, 2023, 34(14): 1749-1755.
[4]	聂昕, 谭天, 申丹凤. 基于深度学习的汽车梁类件冲压回弹研究[J]. 中国机械工程, 2023, 34(07): 838-846.
[5]	李守港, 刘鹏, 刘祥, 胡志力, . 高强钢热冲压结构件三维激光切割回弹规律及精度补偿方法[J]. 中国机械工程, 2022, 33(22): 2741-2747.
[6]	李方坡, 路彩虹, 赵靖宵, 李秀程, 尚成嘉. 基于机器学习的石油装备用大截面高强韧马氏体钢智能设计与性能研究[J]. 中国机械工程, 2022, 33(19): 2325-2330.
[7]	杨卓云, 樊博成, 李萧逸, 张子路, 董国疆, . 7075铝板弹性模量退化行为研究及回弹预测[J]. 中国机械工程, 2022, 33(18): 2241-2248.
[8]	屈聪, 孟智娟, 赵亮, 陈耀, 马立东. 基于变弹性模量的Ti-6Al-4V板材五点弯曲回弹预测[J]. 中国机械工程, 2022, 33(16): 1991-1999.
[9]	王顶, 吴德宇, 杨达亮, 陈立挺, 叶锦华, 吴海彬. 具有压力和位置检测功能的气囊型触觉传感器[J]. 中国机械工程, 2022, 33(14): 1691-1696,1706.
[10]	刘永兴, 唐小琦, 钟靖龙, 钟震宇, 周向东. 基于轻量级梯度提升机的非对称风险注塑成形产品尺寸预测模型[J]. 中国机械工程, 2022, 33(08): 965-969.
[11]	翁剑, 庄可佳, 浦栋麟, 丁汉, . 基于机器学习和多目标算法的钛合金插铣优化[J]. 中国机械工程, 2021, 32(07): 771-777.
[12]	秦勤1,2；何流1；李程2；臧勇1. 铜铝复合板渐进成形回弹缺陷研究[J]. 中国机械工程, 2021, 32(03): 348-356.
[13]	徐勇, 李明, 夏亮亮, 张士宏, . [材料绿色成形技术]异形排气管多向局部加载液力成形工艺[J]. 中国机械工程, 2020, 31(22): 2763-2771.
[14]	孙宇飞1;赵长财1;杨盛福2;杨卓云1. AA7075板材热处理的增塑机制及成形性能[J]. 中国机械工程, 2020, 31(17): 2128-2136.
[15]	姜少飞;邬天骥;彭翔;李吉泉;李治;孙涛. 基于XGBoost特征提取的数据驱动故障诊断方法[J]. 中国机械工程, 2020, 31(10): 1232-1239.