Abstract
This study proposed a novel multi-source deep learning framework for predicting the tensile properties of superalloys containing microporosity defects by simultaneously incorporating both microstructure and defect features. A comprehensive multi-source dataset was constructed using multi-source microstructure and microporosity defect images obtained from tensile specimens extracted from cast plates with diverse microstructural and defect characteristics. The target mechanical properties include ultimate tensile strength (UTS), yield strength (YS), and elongation (EL). Compared to models trained using only microstructure or defect images, the proposed multi-source framework achieved superior prediction accuracy, with R2 values exceeding 0.93 for all three properties. In addition, the mean absolute error (MAE) decreased with an increasing number of microstructural image channels, indicating that the incorporation of multi-source microstructural features significantly enhances model performance. Furthermore, an explainable AI methodology was applied to reveal the underlying mechanisms by which microstructural and defect features govern tensile behavior. This framework presents a data-driven methodology for uncovering microstructure-defect-property relationships, providing a potential pathway for accurate mechanical property prediction of defect-containing superalloys. Its modular architecture can also be readily applied to other alloys.
Similar content being viewed by others
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
Chen, Y. Q., Francis, E., Robson, J., Preuss, M. & Haigh, S. J. Compositional variations for small-scale gamma prime (γ′) precipitates formed at different cooling rates in an advanced Ni-based superalloy. Acta Mater. 85, 199–206 (2015).
Krakow, R. et al. On the crystallography and composition of topologically close-packed phases in ATI 718Plus®. Acta Mater. 130, 271–280 (2017).
Liu, J. et al. In-plane and out-of-plane deformation at the sub-grain scale in polycrystalline materials assessed by confocal microscopy. Acta Mater. 169, 260–274 (2019).
Jaladurgam, N. R. et al. Microstructure-dependent deformation behaviour of a low γ′ volume fraction Ni-base superalloy studied by in-situ neutron diffraction. Acta Mater. 183, 182–195 (2020).
Ge, H. et al. Dislocation climbing dominated decomposition and fracture of carbides in a Ni-based superalloy. Acta Mater. 246, 118669 (2023).
Liu, Y. et al. Effects of microporosity and precipitates on the cracking behavior in polycrystalline superalloy Inconel 718. Mater. Charact. 132, 175–186 (2017).
Liu, Y. et al. Fracture mechanisms induced by microporosity and precipitates in isothermal fatigue of polycrystalline nickel based superalloy. Mater. Sci. Eng. A 736, 438–452 (2018).
Li, M., Coakley, J., Isheim, D., Tian, G. & Shollock, B. Influence of the initial cooling rate from γ′ supersolvus temperatures on microstructure and phase compositions in a nickel superalloy. J. Alloy. Compd. 732, 765–776 (2018).
Guo, X. et al. Solidification rate driven microstructural stability and its effect on the creep property of a polycrystalline nickel-based superalloy K465. Mater. Sci. Eng. A 770, 138530 (2020).
Nawrocki, J. et al. Effect of cooling rate on macro- and microstructure of thin-walled nickel superalloy precision castings. J. Manuf. Process 49, 153–161 (2020).
Tarasov, D., Tyagunov, A. & Milder, O. BRANN forecast of superalloy tensile strength versus Larson–Miller parameter relation and its approximation by a sigmoidal function. Math. Method Appl. Sci. 46, 16401–16414 (2023).
Yin, J. et al. Interpretable predicting creep rupture life of superalloys: enhanced by domain-specific knowledge. Adv. Sci. 11, 2307982 (2024).
Liu, Y. et al. Predicting creep rupture life of Ni-based single crystal superalloys using divide-and-conquer approach based machine learning. Acta Mater. 195, 454–467 (2020).
Wu, J., Xu, Z., Qiao, H., Zhao, J. & Huang, Z. Mechanical properties prediction of superalloy FGH4095 treated by laser shock processing based on machine learning. Mater. Lett. 297, 129970 (2021).
Xu, B. et al. Design of Ni-based turbine disc superalloys with improved yield strength using machine learning. J. Mater. Sci. 57, 10379–10394 (2022).
Farache, D. E. et al. Linking Stress-Rupture Properties to Processing Parameters of HAYNES® 718 Nickel Superalloy Using Machine Learning. In 10th International Symposium on Superalloy 718 and Derivatives, 383–398 (Springer, 2023).
Xie, R.-K. et al. Predicting multiaxial fatigue life of FGH96 superalloy based on machine learning models by considering failure process and loading paths. Int. J. Fatigue. 175, 107730 (2023).
Xu, B. et al. Data-driven design of Ni-based turbine disc superalloys to improve yield strength. J. Mater. Sci. Technol. 155, 175–191 (2023).
Yang, J., Liu, B. & Huang, H. Research on composition-process-property prediction of die casting Al alloys via combining feature creation and attention mechanisms. J. Mater. Res. Technol. 28, 335–346 (2024).
Herriott, C. & Spear, A. D. Predicting microstructure-dependent mechanical properties in additively manufactured metals with machine- and deep-learning methods. Comput. Mater. Sci. 175, 109599 (2020).
Kusano, M. et al. Tensile properties prediction by multiple linear regression analysis for selective laser melted and post heat-treated Ti-6Al-4V with microstructural quantification. Mater. Sci. Eng. A 787, 139549 (2020).
Porro, M., Zhang, B., Parmar, A. & Shin, Y. C. Data-driven modeling of mechanical properties for 17-4 PH stainless steel built by additive manufacturing. Integrating Mater. Manuf. Innov. 11, 241–255 (2022).
Parmar, A., Kumar, R. S., Atwater, M. & Shin, Y. C. Prediction of mechanical properties of additively manufactured 316L steel based on microstructural features using data-driven approaches. J. Mater. Eng. Perform. 34, 26739–26750 (2025).
Pei, X. et al. Robustness of machine learning to color, size change, normalization, and image enhancement on micrograph datasets with large sample differences. Mater. Des. 232, 112086 (2023).
Kondo, R., Yamakawa, S., Masuoka, Y., Tajima, S. & Asahi, R. Microstructure recognition using convolutional neural networks for prediction of ionic conductivity in ceramics. Acta Mater. 141, 29–38 (2017).
Lubbers, N., Lookman, T. & Barros, K. Inferring low-dimensional microstructure representations using convolutional neural networks. Phys. Rev. E 96, 052111 (2017).
Cang, R., Li, H., Yao, H., Jiao, Y. & Ren, Y. Improving direct physical properties prediction of heterogeneous materials from imaging data via convolutional neural network and a morphology-aware generative model. Comput. Mater. Sci. 150, 212–221 (2018).
Wu, J., Yin, X. & Xiao, H. Seeing permeability from images: fast prediction with convolutional neural networks. Sci. Bull. 63, 1215–1222 (2018).
Zheng, X., Zheng, P., Zheng, L., Zhang, Y. & Zhang, R.-Z. Multi-channel convolutional neural networks for materials properties prediction. Comput. Mater. Sci. 173, 109436 (2020).
Nam, C. Convolutional neural network guided prediction of saturation magnetic flux density of Fe-based metallic glasses. Comput. Mater. Sci. 225, 112198 (2023).
Murakami, Y., Furushima, R., Shiga, K., Miyajima, T. & Omura, N. Mechanical property prediction of aluminium alloys with varied silicon content using deep learning. Acta Mater. 286, 120683 (2025).
Gollapalli, A. & Singh, A. K. Design of novel interpretable deep learning framework for microstructure–property relationships in nickel and cobalt based superalloys. Comput. Mater. Sci. 253, 113854 (2025).
Shishir, M. I. R., Elapolu, M. S. R. & Tabarraei, A. A deep learning model for predicting mechanical properties of polycrystalline graphene. Comput. Mater. Sci. 218, 111924 (2023).
Lee, J.-W., Park, W. B., Lee, J. H., Singh, S. P. & Sohn, K.-S. A deep-learning technique for phase identification in multiphase inorganic compounds using synthetic XRD powder patterns. Nat. Commun. 11, 86 (2020).
Feng, S. et al. A general and transferable deep learning framework for predicting phase formation in materials. Npj Comput. Mater. 7, 10 (2021).
Liu, K. et al. Quantifying the influences of carbides and porosities on the fatigue crack evolution of a Ni-based single-crystal superalloy using X-ray tomography. Acta Met. Sin. Engl. 35, 133–145 (2022).
Wu, Y. et al. Slip and fracture behavior of δ-Ni3Nb plates in a polycrystalline nickel-based superalloy during fatigue. Scr. Mater. 171, 36–41 (2019).
Liu, Y. et al. Crack formation and microstructure-sensitive propagation in low cycle fatigue of a polycrystalline nickel-based superalloy with different heat treatments. Int. J. Fatigue. 108, 79–89 (2018).
Khatavkar, N., Swetlana, S. & Singh, A. K. Accelerated prediction of Vickers hardness of Co- and Ni-based superalloys from microstructure and composition using advanced image processing techniques and machine learning. Acta Mater. 196, 295–303 (2020).
Wang, X. et al. Machine learning method for estimating the defect-related mechanical properties of additive manufactured alloys. Eng. Fract. Mech. 291, 109559 (2023).
Xu, J., Li, L., Liu, X., Li, H. & Feng, Q. Quantitative models of high temperature creep microstructure-property correlation of a nickel-based single crystal superalloy with physical and statistical features. J. Mater. Res. Technol. 19, 2301–2313 (2022).
Wen, H. et al. Machine learning-assisted constitutive modeling of a novel powder metallurgy superalloy. Int. J. Mech. Sci. 260, 108654 (2023).
Wu, R., Zeng, L., Fan, J., Peng, Z. & Zhao, Y. Composition, heat treatment, microstructure and loading condition based machine learning prediction of creep life of superalloys. Mech. Mater. 187, 104819 (2023).
Duan, Z., Chen, W., Pei, X., Hou, H. & Zhao, Y. A multimodal data-driven design of low pressure die casting gating system for aluminum alloy cabin. J. Mater. Res. Technol. 27, 2723–2736 (2023).
Zhang, Z. et al. Multimodal deep-learning framework for accurate prediction of wettability evolution of laser-textured surfaces. ACS Appl. Mater. Interfaces. 15, 10261–10272 (2023).
Ma, D. et al. Online porosity prediction in laser welding of aluminum alloys based on a multi-fidelity deep learning framework. J. Intell. Manuf. 35, 55–73 (2024).
Zhu, L. et al. Prediction of ultimate tensile strength of Al-Si alloys based on multimodal fusion learning. MGE Adv. 2, e26 (2024).
Ren, D., Wang, C., Wei, X., Lai, Q. & Xu, W. Building a quantitative composition-microstructure-property relationship of dual-phase steels via multimodal data mining. Acta Mater. 252, 118954 (2023).
Liu, K. et al. Quantifying the effects of carbides and pores on fatigue damages of Ni-based single crystal superalloys at elevated temperature using X-Ray CT scans. J. Nondestr. Eval. 41, 47 (2022).
Dong, X. et al. Microstructure of carbides at grain boundaries in nickel based superalloys. J. Mater. Sci. Technol. 28, 1031–1038 (2012).
Sui, S. et al. The influence of Laves phases on the room temperature tensile properties of Inconel 718 fabricated by powder feeding laser additive manufacturing. Acta Mater. 164, 413–427 (2019).
Zhu, Q., Wang, C., Qin, H., Chen, G. & Zhang, P. Effect of the grain size on the microtensile deformation and fracture behaviors of a nickel-based superalloy via EBSD and in-situ synchrotron radiation X-ray tomography. Mater. Charact. 156, 109875 (2019).
Selvaraju, R. R. et al. Grad-CAM: visual explanations from deep networks via gradient-based localization. Int J. Comput Vis. 128, 336–359 (2020).
Wu, J. J. et al. Effect of wall thickness on micropores and stress-rupture properties of a single-crystal nickel-based superalloy. Mater. Sci. Eng. A 872, 144941 (2023).
Du, B. et al. Tensile, creep behavior and microstructure evolution of an as-cast Ni-based K417G polycrystalline superalloy. J. Mater. Sci. Technol. 34, 1805–1816 (2018).
Zhang, Y., Lan, L. & Zhao, Y. Effect of precipitated phases on the mechanical properties and fracture mechanisms of Inconel 718 alloy. Mater. Sci. Eng. A 864, 144598 (2023).
Li, J. et al. Enhancing tensile properties of wrought Ni-based superalloy ATI 718Plus at elevated temperature via morphology control of η phase. Mater. Charact. 169, 110547 (2020).
Taylor, G. I. The mechanism of plastic deformation of crystals. Part I.—Theoretical. Proc. R. Soc. Lond. Ser. A Contain. Pap. A Math. Phys. Character 145, 362–387 (1934).
Parish, C. M. Cluster analysis of combined EDS and EBSD data to solve ambiguous phase identifications. Microsc. Microanal. 28, 371–382 (2022).
Taylor, L. & Nitschke, G. Improving deep learning with generic data augmentation. In 2018 IEEE Symposium Series on Computational Intelligence (SSCI), 1542–1547 (IEEE, 2018).
Bernett, J. et al. Guiding questions to avoid data leakage in biological machine learning applications. Nat. Methods 21, 1444–1453 (2024).
Simonyan, K. & Zisserman, A. Very deep convolutional networks for large-scale image recognition. In 3rd International Conference on Learning Representations, 1–14 (ICLR 2015).
He, K., Zhang, X., Ren, S., Sun, J. Deep residual learning for image recognition. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770–778 (IEEE, 2016).
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A. & Chen, L. C. MobileNetV2: inverted residuals and linear bottlenecks. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 4510–4520 (IEEE, 2018).
Tan, M. & Le, Q. EfficientNet: rethinking model scaling for convolutional neural networks. In Proceedings of the 36th International Conference on Machine Learning (eds Kamalika, C. & Ruslan, S.) 6105–6114 (PMLR, 2019).
Huang, G., Liu, Z., Weinberger, K. Q. Densely Connected Convolutional Networks. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2261–2269 (IEEE, 2017).
Kingma, D. P. & Ba, J. Adam: a method for stochastic optimization. In Proceedings of the 3rd International Conference on Learning Representations, 13 (ICLR 2015).
Paszke, A. et al. Pytorch: an imperative style, high-performance deep learning library. In Proceedings of the 33rd International Conference on Neural Information Processing Systems, 32 (ACM, 2019)
Selvaraju, R.R. et al. Grad-CAM: visual explanations from deep networks via gradient-based localization. In 2017 IEEE International Conference on Computer Vision (ICCV), 618–626 (IEEE, 2017).
Acknowledgements
This study was financially supported by the following funding sources: the National Science and Technology Major Project of China (J2019-VI-0004-0117), the National Natural Science Foundation of China (51971142, 52031012, 52090042, and 52201045). Wei Xiong was involved in this work during his spare time without funding support. Wei Xiong contributed to this research in his spare time without funding.
Author information
Authors and Affiliations
Contributions
H.P.Y. conceptualized the project, developed, trained the multi-source frameworks and drafted the manuscript. H.P.Y., C.Y.D., Y.T.L. conducted the experiments and performed data analysis. M.D.K., J.W., W.X., B.D.S. supervised the overall project. M.D.K., Y.H.L., J.W., B.D.S. provided funding. All authors reviewed the manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Yu, H., Ding, C., Kang, M. et al. Mechanical property prediction of superalloys with microporosity defects using a multi-source deep learning framework. npj Comput Mater (2026). https://doi.org/10.1038/s41524-026-02055-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41524-026-02055-4
