基于SPSO-BP神经网络的带筋壁板蠕变成形高精度快速回弹预测

doi:10.3901/JME.2025.20.081

机械工程学报 ›› 2025, Vol. 61 ›› Issue (20): 81-90.doi: 10.3901/JME.2025.20.081

• 材料科学与工程 • 上一篇

扫码分享

基于SPSO-BP神经网络的带筋壁板蠕变成形高精度快速回弹预测

湛利华^1,2, 赵帅¹, 杨有良^1,2, 陈赞冲¹, 谢豪¹, 刘长志³, 鄢东洋³

1. 中南大学轻合金研究院长沙 410083;
2. 中南大学极端服役性能精准制造全国重点实验室长沙 410083;
3. 北京宇航系统工程研究所北京 100076

收稿日期:2024-11-25 修回日期:2025-07-12 发布日期:2025-12-03
作者简介:湛利华，女，1976年出生，博士，教授，博士研究生导师。主要研究方向为复杂薄壁构件形性协同制造。E-mail:yjs-cast@csu.edu.cn
杨有良(通信作者)，男，1988年出生，博士，讲师。主要研究方向为复杂薄壁构件形性协同制造。E-mail:yangyouliang@csu.edu.cn
基金资助:
国家自然科学基金(U2341273，U22A20190，52205435)、湖南省自然科学基金(2022JJ40621)、湖南省科技创新计划(2020RC4001)和极端服役性能精准制造国家重点实验室自主课题(ZZYJKT2022-07)资助项目。

High Precision and Rapid Springback Prediction for Creep Forming of Ribbed Panel Based on SPSO-BP Neural Network

ZHAN Lihua^1,2, ZHAO Shuai¹, YANG Youliang^1,2, CHEN Zanchong¹, XIE Hao¹, LIU Changzhi³, YAN Dongyang³

1. Research Institute of Light Alloy, Central South University, Changsha 410083;
2. State Key Laboratory of Precision Manufacturing for Extreme Service Performance, Central South University, Changsha 410083;
3. Beijing Institute of Astronautical Systems Engineering, Beijing 100076

Received:2024-11-25 Revised:2025-07-12 Published:2025-12-03

摘要/Abstract

摘要： 蠕变时效成形技术可实现大型复杂薄壁构件的形性协同制造，但其工艺制度、模具型面等关键参数的确定需要大量繁琐的有限元计算迭代，导致开发周期长、成本高。在此背景下，提出基于SPSO-BP神经网络的带筋壁板蠕变时效成形高精度快速回弹预测方法。通过对蠕变时效成形过程关键建模参数提取以及带筋壁板结构特征离散点表征，建立了材料特性、时效工艺、壁板结构、模具型面与构件回弹型面关联关系的BP神经网络回弹预测模型；进一步通过考虑混沌Logistic映射函数的SPSO粒子群算法优化BP神经回弹模型初始权值和阈值，实现了神经网络回弹模型精确训练和数据集扩充。试验结果表明，SPSO-BP预测模型RMSE误差仅为0.68 mm，计算效率是常规有限元仿真的90倍。研究结果可用于大型带筋类壁板构件蠕变时效成形工艺参数范围快速定位，大幅缩短开发周期。

关键词: 蠕变时效, 回弹预测, 粒子群算法, BP神经网络, 带筋壁板

Abstract: Creep age forming can realize the shape-properties integrated manufacturing of large and complex thin-walled components. However, the determination of key parameters such as process system and die surface requires complicated finite element calculation iterations, which leads to long development cycle and high cost. Based on SPSO-BP neural network, a high precision and rapid springback prediction method for creep age forming of ribbed panel is proposed. By extracting key modeling parameters of forming process and characterizing discrete points of structure features of ribbed panel, a BP neural network model for predicting the relationship between material properties, ageing process, panel structure, die surface and component’s springback surface was established. Furthermore, the SPSO particle swarm optimization considering chaotic Logistic mapping function was used to optimize the initial weight and threshold of the BP neural springback model, and the accurate training and data set expansion of the neural network springback model were realized. The experimental results show that the RMSE error of SPSO-BP prediction model is only 0.68 mm and the calculation efficiency is 90 times higher than that of conventional finite element simulation. The research results can be used to rapidly locate the processing parameters of creep age forming for large components and shorten the development cycle significantly.

Key words: creep age forming, springback prediction, particle swarm optimization, BP neural network, ribbed panel

中图分类号:

TG156

湛利华, 赵帅, 杨有良, 陈赞冲, 谢豪, 刘长志, 鄢东洋. 基于SPSO-BP神经网络的带筋壁板蠕变成形高精度快速回弹预测[J]. 机械工程学报, 2025, 61(20): 81-90.

ZHAN Lihua, ZHAO Shuai, YANG Youliang, CHEN Zanchong, XIE Hao, LIU Changzhi, YAN Dongyang. High Precision and Rapid Springback Prediction for Creep Forming of Ribbed Panel Based on SPSO-BP Neural Network[J]. Journal of Mechanical Engineering, 2025, 61(20): 81-90.

参考文献

[1] ZHAN Lihua，LIN Jianguo，DEAN T A. A review of the development of creep age forming：Experimentation，modelling and applications[J]. International Journal of Machine Tools and Manufacture，2011，51(1)：1-17.
[2] 陈俊，张玉成，王博，等. 5系铝合金车门内板冲压成形及回弹研究[J]. 精密成形工程，2023，15(12)：106-115. CHEN Jun，ZHANG Yucheng，WANG Bo，et al. Stamping forming and springback of 5 series aluminum alloy door inner panel[J]. Journal of Netshape Forming Engineering，2023，15(12)：106-115.
[3] 樊博成. 7075高强铝板弹塑性成形回弹预测与试验研究[D]. 秦皇岛：燕山大学，2021. FAN Bocheng. Elastoplastic forming springback prediction and experimental study of 7075 high-strength aluminum sheet [D]. Qinhuangdao：Yanshan University，2021.
[4] 胡轶敏，林忠钦，徐伟力，等. 车身覆盖件冲压成形动态仿真的研究进展[J]. 力学进展，2000(2)：252-271. HU Yimin，LIN Zhongqin，XU Weili，et al. Research progress on dynamic simulation of stamping forming for body coverings[J]. Advances in Mechanics，2000(2)：252-271.
[5] PATHAPATI S K，BRANDON G，RAM D. LocateMe：Magnetic-fields-based indoor localization using smartphones[J]. ACM Transactions on Intelligent Systems and Technology (TIST)，2013，4(4)：1-27.
[6] WANG C T，KINZEL G L，ALTAN T. Process simulation and springback control in plane strain sheet bending[J]. Springs，1993，1：45-54.
[7] CHUN B K，JINN J T，LEE J K. Modeling the Bauschinger effect for sheet metals，part I：Theory[J]. International Journal of Plasticity，2002，18(5)：571-595.
[8] SIEGERT K，ALTAN T，NAKAGAWA T. Development and manufacture of dies for car body production[J]. CIRP Annals-Manufacturing Technology，1997，46(2)：535-543.
[9] 马振武，童国权，徐宝林. 蠕变时效成形回弹预测及工装型面优化[J]. 航空制造技术，2017(1)：66-69，78. MA Zhenwu，TONG Guoquan，XU Baolin. Springback prediction and tool profile optimization of creep aging forming[J]. Aeronautical Manufacturing Technology，2017(1)：66-69，78.
[10] 温正略，李卫东，万敏，等. 7475铝合金板蠕变时效成形回弹研究[J]. 塑性工程学报，2016，23(3)：77-81. WEN Zhenglue，LI Weidong，WAN Min，et al. Study on springback of 7475 aluminum alloy sheet during creep aging forming[J]. Journal of Plasticity Engineering，2016，23(3)：77-81.
[11] LU R，LIU X，XU Z，et al. Simulation of springback variation in the U-bending of tailor rolled blanks[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering，2017，39(11)：4633-4647.
[12] FEI H，MO J，QI H，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.
[13] SONG Y，YU Z. Springback prediction in T-section beam bending process using neural networks and finite element method[J]. Archives of Civil and Mechanical Engineering，2013，13(2)：229-241.
[14] AGUIR H，BELHADJSSLAH H，HAMBLI R. Parameter identification of an elasto-plastic behaviour using artificial neural networks-genetic algorithm method[J]. Materials & Design，2011，32(1)：48-53.
[15] KAZAN R，FIRAT M，TIRYAKI A E. Prediction of springback in wipe-bending process of sheet metal using neural network[J]. Materials & Design，2009，30(2)：418-423.
[16] YIN F，MAO H，HUA L，et al. Back propagation neural network modeling for warpage prediction and optimization of plastic products during injection molding[J]. Materials & Design，2011，32(4)：1844-1850.
[17] JAMLI M R，ARIFFIN A K，WAHAB D A. Integration of feed forward neural network and finite element in the draw-bend springback prediction[J]. Expert Systems with Applications，2014，41(8)：3662-3670.
[18] 林祥烽. 铝型材三维拉弯构件成形精度与BP神经网络回弹模型的研究[D]. 长春：吉林大学，2019. LIN Xiangfeng. Research on forming accuracy and BP neural network rebound model of aluminum profile Sanvira bending member[D]. Changchun：Jilin University，2019.
[19] KAZAN R，FIRAT M，TIRYAKI A E. Prediction of springback in wipe-bending process of sheet metal using neural network[J]. Materials & Design，2009，30(2)：418-423.
[20] LI L，ZHANG Z，XU B. Prediction of spherical sheet springback based on a sparrow-search-algorithm- optimized BP neural network[J]. Metals，2022，12：1377.
[21] LIU W，LIU Q，RUAN F，et al. Springback prediction for sheet metal forming based on GA-ANN technology[J]. Journal of Materials Processing Tech.，2007，187：227-231.
[22] XU B，LI L，WANG Z，et al. Prediction of springback in local bending of hull plates using an optimized backpropagation neural network[J]. Copernicus GmbH，2021，12(2)：777-789.
[23] INAMDAR M V，DATE P P，DESAI U B. Studies on the prediction of springback in air vee bending of metallic sheets using an artificial neural network[J]. Journal of Materials Processing Technology，2000，108(1)：45-54.
[24] MOHSEN A，MOHAMMAD R S，AMIR M K. An improved particle swarm optimization based on neural network for surface roughness optimization in face milling of 6061-t6 aluminum[J]．International Journal of Applied Engineering Research，2010，5(19)：3191-3201.
[25] SONG S J，PARK H J，SHIN Y K，et al. Flaw characterization in tubes by inversion of eddy current signals using neural networks trained by finite element model-based synthetic data[M]// Review of Progress in Quantitative Nondestructive Evaluation. Boston：Springer，1999，18(A)：881-888.
[26] 郝玲玲，朱永利. 改进萤火虫算法与小波神经网络相结合的变压器故障诊断[J]. 科学技术与工程，2019，19(31)：156-161. HAO Lingling，ZHU Yongli. Transformer fault diagnosis based on improved firefly algorithm combined with wavelet neural network[J]. Science Technology and Engineering，2019，19(31)：156-161.
[27] 湛利华，许晓龙，贾树峰，等. 7055铝合金加筋板蠕变时效成形有限元仿真[J]. 中国有色金属学报，2013，23(8)：2104-2109. ZHAN Lihua，XU Xiaolong，JIA Shufeng，et al. Finite element simulation of creep aging forming of 7055 aluminum alloy stiffened plate[J]. Chinese Journal of Nonferrous Metals，2013，23(8)：2104-2109.

基于SPSO-BP神经网络的带筋壁板蠕变成形高精度快速回弹预测

High Precision and Rapid Springback Prediction for Creep Forming of Ribbed Panel Based on SPSO-BP Neural Network

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	曹建国, 王雷雷, 张兆祥, 曹媛, 王犇, 高博. 基于粒子群算法的核电锆合金冷轧板形控制VCR辊形研究[J]. 机械工程学报, 2025, 61(8): 75-84.
[2]	邵东, 尹思维, 李亮, 王翔宇, 魏凌涛, 周道林. 基于神经网络的ESC线控制动压力估计与控制[J]. 机械工程学报, 2025, 61(4): 219-228.
[3]	解甜, 彭宣, 张梦哲, 王传礼, 徐壮. 基于GMA宏微精密驱动器的磁滞非线性模型与参数辨识[J]. 机械工程学报, 2025, 61(17): 245-254.
[4]	王泽根, 宫岛, 周劲松, 刘广宇. 基于车轮踏面凹磨演化控制的动车组运行稳定性优化研究[J]. 机械工程学报, 2025, 61(16): 283-292.
[5]	曾达幸, 许林淼, 范朝辉, 刘亚, 侯雨雷, 卢文娟. 仿中医按摩执行器按摩力分析及优化设计[J]. 机械工程学报, 2025, 61(15): 247-260.
[6]	戚宇彤, 孙楚研, 李淑慧. 7075铝合金超低温变形行为与回弹预测研究[J]. 机械工程学报, 2025, 61(14): 99-107.
[7]	孙浩, 原野, 张琦, 张小贝. 基于终端斯坦纳树的高效多分支线束自动布局方法[J]. 机械工程学报, 2025, 61(14): 410-421.
[8]	来鑫, 翁嘉辉, 杨一鹏, 孙宇飞, 周龙, 郑岳久, 韩雪冰. 基于BPNN-EKF-GD-RF算法的锂离子电池组荷电状态估计方法[J]. 机械工程学报, 2025, 61(12): 251-265.
[9]	杨晨, 李涛, 王明宇, 杨东东, 闫萌. 基于不完全信息补全的产品全生命周期评价方法[J]. 机械工程学报, 2025, 61(11): 311-325.
[10]	王俊玖, 刘金玉, 侯秀娟, 齐振国, 李志敏, 刘涛. 列车车体柔性装配精度预测的力学-数据混合建模方法[J]. 机械工程学报, 2024, 60(6): 177-186.
[11]	郑学斌, 韩龙帅, 李学涛, 鄂宏伟, 吴向东, 万敏. DP780双相钢U弯回弹预测影响因素研究[J]. 机械工程学报, 2024, 60(6): 197-206.
[12]	刘承尚, 陈锐, 李芳, 王萌, 徐戊矫. 基于改进遗传算法的超高强钢硬化模型参数确定及折弯回弹预测[J]. 机械工程学报, 2024, 60(12): 240-249.
[13]	卢琪, 苏凯新, 张继旺, 闫涛, 杨冰, 张浩楠. 基于BP神经网络和遗传算法的简单链型悬挂接触网结构优化[J]. 机械工程学报, 2024, 60(12): 313-320.
[14]	王汝贵, 董奕辰, 陈辉庆. 一种变胞式工业机器人时间最优轨迹规划[J]. 机械工程学报, 2023, 59(5): 100-111.
[15]	董海江, 郑婷婷, 于进涛, 付正家, 刘秀成, 邢智翔, 闫正祥, 何存富. 汽车B柱加强板软硬部过渡特性的磁巴克豪森噪声检测方法[J]. 机械工程学报, 2023, 59(4): 10-17.