Modares Mechanical Engineering

Modares Mechanical Engineering

Stress Field Prediction Using Conditional Generative Adversarial Networks and Image Processing in a Perforated Plate under Static Loading

Document Type : Original Article

Authors
Behnam Anbarlooie
Melika Maleki Rad
Faculty of New Technologies and Aerospace Engineering, Shahid Beheshti University, Tehran, Iran
10.48311/mme.2025.96917.0
Abstract
The study of the mechanical properties and behavior of materials, as well as stress and strain fields, has been carried out using methods such as experiments, numerical methods, and precise mathematical solutions over the decades. In recent years, machine learning, and especially deep learning, have become one of the most commonly used methods in various engineering fields. One of its important applications is the prediction of material behavior in numerous structures. These methods have drawn significant attention due to their rapid execution, apposite accuracy, and implementation convenience, and are used as an alternative or supplementary tool for traditional analysis methods. Using the machine learning method, in case the problems are properly characterized, they can provide a much more powerful tool in a machine learning process compared to other tools. The purpose of this paper is to predict the stress field and maximum stress on a perforated plate under static loading using a deep learning method based on a conditional adversarial generative network (CGANs) and to quantify the results using an image processing method. Also, at the end, the numerical results obtained from this model are extracted and compared with the results attained from finite element analysis to evaluate the accuracy of the proposed model
Keywords
Stress Field, Machine Learning, Image Processing, Finite Element Method

Subjects

[1] I. J. Goodfellow, et al., "Generative adversarial nets," Advances in Neural Information Processing Systems, vol. 27, 2014. doi:10.48550/arXiv.1406.2646
[2] A. Aggarwal, M. Mittal, and G. Battineni, "Generative adversarial network: An overview of theory and applications," International Journal of Information Management Data Insights, vol. 1, no. 1, p. 100004, 2021. doi :10.1016/j.jjimei.2020.100004
[3] P. Isola, et al., "Image-to-image translation with conditional adversarial networks," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. doi:10.1109/CVPR.2017.632
[4] O. Ronneberger, P. Fischer, and T. Brox, "U-net: Convolutional networks for biomedical image segmentation," in International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2015. Springer. doi:10.1007/978-3-319-24574-4_28
[5] Z. Yang, C. H. Yu, and M. J. Buehler, "Deep learning model to predict complex stress and strain fields in hierarchical composites," Science Advances, vol. 7, no. 15, 2021.doi:10.1126/sciadv.abd7416
[6] C. Xu, et al., "Conditional Generative Adversarial Network Enabled Localized Stress Recovery of Periodic Composites," Computer Modeling in Engineering & Sciences (CMES), vol. 140, no. 1, 2024.doi:10.32604/cmes.2024.047327
[7] V. L. T. De Souza, et al., "A review on generative adversarial networks for image generation," Computers & Graphics, vol. 114, pp. 13–25, 2023. doi:10.1016/j.cag.2023.05.010
[8] Tensorflow, "Pix2Pix tutorial." Available: https://www.tensorflow.org/tutorials/generative/pix2pix. doi:10.5281/zenodo.4724125
[9] P. Ulmas and I. Liiv, "Segmentation of satellite imagery using U-net models for land cover classification," arXiv preprint arXiv:2003.02899, 2020.doi: 10.48550/arXiv.2003.02899
[10] J. Zhao, et al., "Firefighting Robot Extinguishment Decision‐Making Based on Visual Guidance: A Novel Attention and Scale U‐Net Model and Genetic Algorithm," Journal of Field Robotics, vol. 42, no. 4, pp. 1103–1124, 2025. doi: 10.1002/rob.22438
[11] G. Du, et al., "Medical image segmentation based on U-net: A review," Journal of Imaging Science & Technology, vol. 64, no. 2, 2020. doi:10.2352/J.ImagingSci.Technol.2020.64.2.020508
[12] Y. Gündüç, "Vit-gan: Image-to-image translation with vision transformers and conditional gans," arXiv preprint arXiv:2110.09305, 2021. doi: 10.48550/arXiv.2110.09305
 
 [13] Nie, Z., Jiang, H. and Kara, L.B., 2018. Deep learning for stress field prediction using convolutional neural networks. arXiv preprint arXiv:1808.08914.
[14] Zhao, Q., Sun, B., Zhao, W., Watanabe, T., Usui, T. and Takeda, H., 2024. Improved GAN-based deep learning approach for strain field prediction and failure analysis of precast bridge slab joints. Engineering Structures, 321, p.119023.doi:10.1016/j.engstruct.2024.119023
[15] Xiang, Y., Hou, J., Chen, X., Pidaparti, R., Song, K., Tang, K. and Wang, X., 2024. A GAN-based stepwise full-field mechanical prediction model for architected metamaterials. International Journal of Mechanical Sciences, 284, p.109771. doi:10.1016/j.ijmecsci.2024.109771 
[16] Sahu, R., Gupta, A., Mittal, D., Chatterjee, P. and Jha, S.K., 2025. Unsupervised Graph-GAN model for stress–strain field prediction in a composite. Journal of Materials Science, pp.1-20. doi:10.1007/s10853-025-10772-2 
[17] Liu, Y., Lin, Q., Pan, W., Yu, W., Ren, Y. and Zhao, Y., 2024. SR-M− GAN: A generative model for high-fidelity stress fields prediction of the composite bolted joints. Advanced Engineering Informatics, 61, p.102537. doi:10.1016/j.aei.2024.102537 
[18] Hui, X., Xu, Y., Niu, J. and Zhang, W., 2024. Rapid evaluation and prediction of cure-induced residual stress of composites based on cGAN deep learning model. Composite Structures, 330, p.117827.
doi:10.1016/j.compstruct.2023.117827 
[19] Amieghemen, G.E., Ramezani, M. and Sherif, M.M., 2025. Residual Pyramidal GAN (RP-GAN) for crack detection and prediction of crack growth in engineered cementitious composites. Measurement, 242, p.115769.doi:10.1016/j.measurement.2024.115769
[20] Jiang, L., Hu, Y., Li, H., Shao, X., Zhang, X., Kan, Q. and Kang, G., 2025. A cGAN-based fatigue life prediction of 316 austenitic stainless steel in high-temperature and high-pressure water environments. International Journal of Fatigue, 190, p.108633. doi:10.1016/j.ijfatigue.2024.108633
 
Modares Mechanical Engineering
Volume 25, Issue 12
December 2025
Pages 801-818