Journal Home > Volume 28 , Issue 4

The accuracy and efficiency of three-dimensional (3D) surface forming, which directly affects the cycle and quality of production, is important in manufacturing. In practice, given the uncertainty of metal plate springback, an error exists between the actual plate and the target surface, which creates a nonlinear mapping from computer aided design models to bending surfaces. Technicians need to reconfigure parameters and process a surface multiple times to delicately control springback, which greatly wastes human and material resources. This study aims to address the springback control problem to improve the efficiency and accuracy of sheet metal forming. A basic computation approach is proposed based on the DeepFit model to calculate the springback value in 3D surface bending. To address the sample data shortage problem, we put forward an advanced approach by combining a deep learning model with case-based reasoning (CBR). Next, a multi-model fused bending parameter generation framework is devised to implement the advanced springback computation approach through surface data preprocessing, CBR-based model matching, convolution neural network-based machining surface generation, and bending parameter generation with a series of model transformations. Moreover, the proposed approaches and the framework are verified by considering saddle surface processing as an example. Overall, this study provides a new idea to improve the accuracy and efficiency of surface processing.


menu
Abstract
Full text
Outline
About this article

Three Dimensional Metal-Surface Processing Parameter Generation Through Machine Learning-Based Nonlinear Mapping

Show Author's information Min Zhu1Yanjun Dong1Bingqing Shen1Haiyan Yu2Lihong Jiang1Hongming Cai1( )
School of Software, Shanghai Jiao Tong University, Shanghai 200240, China
School of Mechanical Engineering, Donghua University, Shanghai 200240, China

Abstract

The accuracy and efficiency of three-dimensional (3D) surface forming, which directly affects the cycle and quality of production, is important in manufacturing. In practice, given the uncertainty of metal plate springback, an error exists between the actual plate and the target surface, which creates a nonlinear mapping from computer aided design models to bending surfaces. Technicians need to reconfigure parameters and process a surface multiple times to delicately control springback, which greatly wastes human and material resources. This study aims to address the springback control problem to improve the efficiency and accuracy of sheet metal forming. A basic computation approach is proposed based on the DeepFit model to calculate the springback value in 3D surface bending. To address the sample data shortage problem, we put forward an advanced approach by combining a deep learning model with case-based reasoning (CBR). Next, a multi-model fused bending parameter generation framework is devised to implement the advanced springback computation approach through surface data preprocessing, CBR-based model matching, convolution neural network-based machining surface generation, and bending parameter generation with a series of model transformations. Moreover, the proposed approaches and the framework are verified by considering saddle surface processing as an example. Overall, this study provides a new idea to improve the accuracy and efficiency of surface processing.

Keywords:

3D surface, point-cloud, machine learning, case-based reasoning, industrial software
Received: 22 March 2022 Revised: 27 June 2022 Accepted: 01 July 2022 Published: 06 January 2023 Issue date: August 2023
References(26)
[1]
Q. F. Zhang, Compensation algorithm for springback in multi-point forming and its validation by numerical simulation and experimental methods, (in Chinese), PhD dissertation, Jilin University, Changchun, China, 2014.
[2]
X. D. Zhang, S. L. Liu, Y. Liu, X. F. Hu, and C. Gao, Review on development trend of launch and recovery technology for USV, (in Chinese), Chin. J. Ship Res., vol. 13, no. 6, pp. 50–57, 2018.
[3]
Y. X. Feng, Z. F. Zhang, G. D. Tian, Z. H. Lv, S. X. Tian, and H. F. Jia, Data-driven accurate design of variable blank holder force in sheet forming under interval uncertainty using sequential approximate multi-objective optimization, Future Generat. Comp. Syst., vol. 86, pp. 1242–1250, 2018.
[4]
Y. C. Gao, Z. X. Zhang, Y. X. Feng, M. Savchenko, I. Hagiwara, and H. Zheng, Flexible mesh morphing in sustainable design using data mining and mesh subdivision, Future Generat. Comp. Syst., vol. 108, pp. 987–994, 2020.
[5]
Z. J. Hou and L. G. Gu, Triangulation based on 3D reconstruction technique, Appl. Mech. Mater., vol. 121–126, pp. 4249–4253, 2011.
[6]
W. Wang, T. Y. Su, H. X. Liu, X. F. Li, Z. Jia, L. Zhou, Z. L. Song, and M. Ding, Surface reconstruction from unoriented point clouds by a new triangle selection strategy, Comput. Graph., vol. 84, pp. 144–159, 2019.
[7]
X. J. Qin, Z. T. Hu, H. B. Zheng, and M. Y. Zhang, Surface reconstruction from unorganized point clouds based on edge growing, Adv. Manuf., vol. 7, no. 3, pp. 343–352, 2019.
[8]
A. N. Ravari and H. D. Taghirad, Reconstruction of B-spline curves and surfaces by adaptive group testing, Comput. Aided Des., vol. 74, pp. 32–44, 2016.
[9]
W. J. Wang, L. L. Lai, J. Chen, and Q. Y. Wu, CAM-based non-local attention network for weakly supervised fire detection, Serv. Oriented Comput. Appl., vol. 16, no. 2, pp. 133–142, 2022.
[10]
W. Zhou, J. Y. Jia, C. X. Huang, and Y. Q. Cheng, Web3D learning framework for 3D shape retrieval based on hybrid convolutional neural networks, Tsinghua Science and Technology, vol. 25, no. 1, pp. 93–102, 2020.
[11]
M. Gadelha, R. Wang, and S. Maji, Multiresolution tree networks for 3D point cloud processing, in Proc. 15th European Conf. Computer Vision, Munich, Germany, 2018, pp. 105–122.
[12]
J. J. Wu, C. K. Zhang, T. F. Xue, W. T. Freeman, and J. B. Tenenbaum, Learning a probabilistic latent space of object shapes via 3D generative-adversarial modeling, in Proc. 30th Int. Conf. Neural Information Processing Systems, Barcelona, Spain, 2016, pp. 82–90.
[13]
C. R. Qi, H. Su, K. C. Mo, and L. J. Guibas, PointNet: Deep learning on point sets for 3D classification and segmentation, presented at the 2017 IEEE Conf. Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 2017, pp. 77–85.
[14]
C. R. Qi, L. Yi, H. Su, and L. J. Guibas, PointNet++: Deep hierarchical feature learning on point sets in a metric space, in Proc. 31st Int. Conf. Neural Information Processing Systems, Long Beach, CA, USA, 2017, pp. 5105–5114.
[15]
A. Paigwar, O. Erkent, C. Wolf, and C. Laugier, Attentional PointNet for 3D-object detection in point clouds, presented at the 2019 IEEE/CVF Conf. Computer Vision and Pattern Recognition Workshops (CVPRW), Long Beach, CA, USA, 2019, pp. 1297–1306.
[16]
W. H. Zhou, J. G. Lu, and W. L. Yue, A new semantic segmentation method of point cloud based on PointNet and VoxelNet, presented at the 2019 Chinese Control and Decision Conf. (CCDC), Nanchang, China, 2019, pp. 803–808.
[17]
Y. Wang, Y. B. Sun, Z. W. Liu, S. E. Sarma, M. M. Bronstein, and J. M. Solomon, Dynamic graph CNN for learning on point clouds, ACM Trans. Graph., vol. 38, no. 5, p. 146, 2019.
[18]
P. Yuan, C. F. Wang, Y. Hu, J. X. Li, C. Y. Zhang, W. Huang, and F. Yu, Development of large plate bending machine for shipbuilding with three-dimensional numerical control, (in Chinese), Shipbuilding of China, vol. 55, no. 2, pp. 122–131, 2014.
[19]
D. Saini, S. Kumar, and T. R. Gulati, NURBS-based geometric inverse reconstruction of free-form shapes, J. King Saud Univ.-Comput. Inform. Sci., vol. 29, no. 1, pp. 116–133, 2017.
[20]
Y. Ben-Shabat and S. Gould, DeepFit: 3D surface fitting via neural network weighted least squares, in Proc. 16th European Conf. Computer Vision, Glasgow, UK, 2020, pp. 20–34.
[21]
D. W. Wei, H. S. Ning, F. F. Shi, Y. L. Wan, J. B. Xu, S. K. Yang, and L. Zhu, Dataflow management in the Internet of Things: Sensing, control, and security, Tsinghua Science and Technology, vol. 26, no. 6, pp. 918–930, 2021.
[22]
J. C. Wang, Y. J. Xie, R. C. Yang, J. C. Liu, and H. Zhou, Numerical analysis and process planning of cold bending and press forming of curved plate for offshore platform, Ship Eng., vol. 43, no. 4, pp. 94–101, 2021.
[23]
Y. D. Chen, Numerical simulation study on the springback control of multi-point forming for curved hull plate, (in Chinese), Master dissertation, Jilin University, Changchun, China, 2015.
[24]
Y. L. Fang, M. Xia, P. L. Ji, X. Zhou, M. Zeng, and X. G. Liu, Deep spherical panoramic representation for 3D shape recognition, (in Chinese), J. Comput. Aided Des. Comput. Graph., vol. 29, no. 9, pp. 1689–1695, 2017.
[25]
T. Q. Li, Research on springback prediction method of profile roll forming based on machine learning, (in Chinese), Master dissertation, Yanshan University, Qinhuangdao, China, 2019.
[26]
Q. Y. Wei, Studies on the prediction of spring-back in air V-bending of sheet-metal, (in Chinese), Master dissertation, Jilin University, Changchun, China 2021.
Publication history
Copyright
Acknowledgements
Rights and permissions

Publication history

Received: 22 March 2022
Revised: 27 June 2022
Accepted: 01 July 2022
Published: 06 January 2023
Issue date: August 2023

Copyright

© The author(s) 2023.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 61972243).

Rights and permissions

The articles published in this open access journal are distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/).

Return