Structural Optimization of 3D Braided Tungsten–Copper Composites via Active Learning and Finite Element Simulation

Abstract

Three-dimensional (3D) braided tungsten–copper (W–Cu) composites are attractive for combining high strength with good electrical performance, yet the high-dimensional design space makes manual optimization impractical. This work integrates high-fidelity finite element modeling (FEM) with an active learning (AL) scheme driven by a Gaussian process regression (GPR) surrogate and an uncertainty-aware acquisition policy. The surrogate provides accurate predictions together with quantified uncertainty, enabling data-efficient exploration of candidate braiding parameters while satisfying a prescribed electrical conductivity requirement. The framework rapidly identifies an optimal braided architecture with a peak yield strength of 1149 MPa, approximately a 30% improvement over the best design in the initial set. SHAP-based interpretability further reveals physically meaningful guidelines, with the in-plane spacing coefficient of the Z-direction fibers emerging as the dominant factor. These results demonstrate a fast interpretable route for optimizing metallic braided architectures and highlight the broader utility of FEM-coupled active learning for materials design.

Keywords

active learning / finite element modeling / representative volume element / three-dimensional braided / tungsten-copper composites

References

[1] B. Zhang, K. Yang, Z. Huang, and J. Wang, “Recent Advances in W-Cu Composites: A Review on the Fabrication, Application, Property, Densification, and Strengthening Mechanism,” Advanced Engineering Materials 26, no. 1 (2024): 2301204, https://doi.org/10.1002/adem.202301204.

[2] F. Yao, W. Chen, Y. Yang, et al., “Engineering Performance of Tungsten Network Reinforced Copper Matrix Composites Synthesized by Selective Laser Melting and Infiltration,” Science and Technology of Advanced Materials 25, no. 1 (2024): 2309888, https://doi.org/10.1080/14686996.2024.2309888.

[3] L.-M. Luo, X.-P. Ding, W.-Z. Xu, C. Y. Wang, Y. Q. Qin, and Y. C. Wu, “Strengthening Mechanisms and Research Progress in the W/Cu Interfaces,” International Journal of Refractory Metals and Hard Materials 125 (2024): 106900, https://doi.org/10.1016/j.ijrmhm.2024.106900.

[4] T. Han, C. Hou, Z. Zhao, et al., “Simultaneous Enhancement of Strength and Conductivity via Self-Assembled Lamellar Architecture,” Nature Communications 15, no. 1 (2024): 1863, https://doi.org/10.1038/s41467-024-46029-w.

[5] A. Röthlisberger, S. Häberli, F. Krogh, H. Galinski, D. C. Dunand, and R. Spolenak, “Ice-Templated W–Cu Composites With High Anisotropy,” Scientific Reports 9, no. 1 (2019): 476, https://doi.org/10.1038/s41598-018-36604-9.

[6] Y. Sun, T. Han, C. Hou, et al., “Multilayered W–Cu Composites With Enhanced Strength, Electrical Conductivity and Wear Resistance,” Composites Part B: Engineering 283 (2024): 111641, https://doi.org/10.1016/j.compositesb.2024.111641.

[7] D. Zhang, X. Zheng, J. Zhou, et al., “Effect of Braiding Architectures on the Mechanical and Failure Behavior of 3D Braided Composites: Experimental Investigation,” Polymers 14, no. 9 (2022): 1916, https://doi.org/10.3390/polym14091916.

[8] X. Wang, H. Li, Y. Zhang, Y. Guan, S. Yan, and J. Zhai, “Compressive Failure Characteristics of 3D Four-Directional Braided Composites With Prefabricated Holes,” Materials 17, no. 15 (2024): 3821, https://doi.org/10.3390/ma17153821.

[9] Y. Liu, Y. Hou, T. Sapanathan, R. Nie, L. Meng, and Y. Xu, “A Multiscale Strategy for Exploring the Mechanical Behavior of 3D Braided Composite Thin-Walled Cylinders,” Thin-Walled Structures 198 (2024): 111705, https://doi.org/10.1016/j.tws.2024.111705.

[10] Y. Xue, Z. Li, G. Li, et al., “Modeling of Multi-Scale 3-D Conductive Network and Electrical Conductivity of Carbon Fiber Braided Composites,” Composites Part A: Applied Science and Manufacturing 187 (2024): 108479, https://doi.org/10.1016/j.compositesa.2024.108479.

[11] Z. Liu, Y. Xue, B. Sun, B. Gu, and M. Hu, “Microstructural Modeling on Electrical Conductivities of 3-D Carbon/Epoxy Woven Composites Along Different Directions,” Composite Structures 329 (2024): 117764, https://doi.org/10.1016/j.compstruct.2023.117764.

[12] X. Li, Y. Xue, B. Gu, and B. Sun, “Electrical Conductivities and Temperature Distributions of Carbon Fiber 3D Woven Composites With Different Electric Field Direction,” Thin-Walled Structures 200 (2024): 111910, https://doi.org/10.1016/j.tws.2024.111910.

[13] Y. Xue, S. Huang, B. Sun, and B. Gu, “Electro-Thermal Coupling Behavior and Temperature Distribution of 3-D Braided Composite Under Direct Current,” Composites Science and Technology 216 (2021): 109043, https://doi.org/10.1016/j.compscitech.2021.109043.

[14] M. Hussain, B. Dong, Z. Qiu, U. Garbe, Z. Pan, and H. Li, “A Review on the Additive Manufacturing of W–Cu Composites,” Metals 15, no. 2 (2025): 197, https://doi.org/10.3390/met15020197.

[15] Y. Wang, K. Wang, and C. Zhang, “Applications of Artificial Intelligence/Machine Learning to High-Performance Composites,” Composites Part B: Engineering 285 (2024): 111740, https://doi.org/10.1016/j.compositesb.2024.111740.

[16] G. Shi, Y. Ouyang, Y. Ren, et al., “Multi-Scale Finite Element Analysis Integrated With Machine Learning for Efficient Prediction of Thermal Conductivity in 3D Orthogonal Woven Composites,” Composites Part B: Engineering 308 (2026): 113005, https://doi.org/10.1016/j.compositesb.2025.113005.

[17] E. Ghane, M. Fagerström, and M. Mirkhalaf, “Multi-Fidelity Data Fusion for Inelastic Woven Composites: Combining Recurrent Neural Networks With Transfer Learning,” Composites Science and Technology 267 (2025): 111163, https://doi.org/10.1016/j.compscitech.2025.111163.

[18] M. M. Hiremath, N. Doshi, T. Bernthaler, et al., “3D Damage Evolution and Microstructural-Based Machine Learning Model for Stiffness Prediction in Woven Composite Under Cyclic Loads,” International Journal of Fatigue 197 (2025): 108913, https://doi.org/10.1016/j.ijfatigue.2025.108913.

[19] Y. Zhang, H. Li, L. Ge, L. Zheng, Z. Tang, and F. Zhao, “A Deep Learning Model Optimized by Bayesian Optimization With Hyperband for Fast Prediction of the Elastic Properties of 3D Tubular Braided Composites at Different Temperatures,” Polymer Composites 46, no. 4 (2025): 3296–3307, https://doi.org/10.1002/pc.29171.

[20] J. Zheng, K. Qian, and D. Zhang, “Reverse Reconstruction of Geometry Modeling and Numerical Verification of 2.5D Woven Composites Based on Deep Learning,” Composite Structures 329 (2024): 117801, https://doi.org/10.1016/j.compstruct.2023.117801.

[21] X. Song, J. Zhou, D. Zhang, et al., “Trans-Scale Analysis of 3D Braided Composites With Voids Based on Micro-CT Imaging and Unsupervised Machine Learning,” Composites Science and Technology 249 (2024): 110494, https://doi.org/10.1016/j.compscitech.2024.110494.

[22] M.-K. Hsu, B.-Y. Huang, and C.-H. Yu, “Ai-driven Optimization of Woven Composite via Integrated Deep and Reinforcement Learning,” Materials and Design 259 (2025): 114798, https://doi.org/10.1016/j.matdes.2025.114798.

[23] M. El Fallaki Idrissi, A. Pasquale, F. Meraghni, F. Praud, and F. Chinesta, “Advanced Meta-Modeling Framework Combining Machine Learning and Model Order Reduction Towards Real-Time Virtual Testing of Woven Composite Laminates in Nonlinear Regime,” Composites Science and Technology 262 (2025): 111055, https://doi.org/10.1016/j.compscitech.2025.111055.

[24] K. Zhu, W. Cao, C. Ran, and B. Gu, “A Novel Automatic Crack Classification Algorithm of 3-D Woven Composites Based on Deep-Learning U-Net Model,” Engineering Fracture Mechanics 289 (2023): 109488, https://doi.org/10.1016/j.engfracmech.2023.109488.

[25] X. Liu, X.-Y. Zhou, B. Liu, and C. Gao, “Multiscale Modeling of Woven Composites by Deep Learning Neural Networks and Its Application in Design Optimization,” Composite Structures 324 (2023): 117553, https://doi.org/10.1016/j.compstruct.2023.117553.

[26] S. Almarri, M. Lloyd, E. Darnbrough, and D. Armstrong, “Quantifying Delamination Energy in Tungsten on Silicon Thin Films Through Nanoindentation and Nanoscratch,” Materials & Design 253 (2025): 113873, https://doi.org/10.1016/j.matdes.2025.113873.

[27] D. J. Walters, D. J. Luscher, and J. D. Yeager, “Considering Computational Speed vs. Accuracy: Choosing Appropriate Mesoscale RVE Boundary Conditions,” Computer Methods in Applied Mechanics and Engineering 374 (2021): 113572, https://doi.org/10.1016/j.cma.2020.113572.

[28] R. Tibshirani, “Regression Shrinkage and Selection via the Lasso,” Journal of the Royal Statistical Society: Series B 58, no. 1 (2018): 267–288, https://doi.org/10.1111/j.2517-6161.1996.tb02080.x.

[29] C. E. Rasmussen and C. K. I. Williams, Gaussian Processes for Machine Learning (MIT Press, 2005), https://gaussianprocess.org/gpml/chapters/RW.pdf.

[30] L. Breiman, “Random Forests,” Machine Learning 45, no. 1 (2001): 5–32, https://doi.org/10.1023/a:1010933404324.

[31] B. Schölkopf, A. Smola, and K.-R. Müller, “Nonlinear Component Analysis as a Kernel Eigenvalue Problem,” Neural Computation 10, no. 5 (1998): 1299–1319, https://doi.org/10.1162/089976698300017467.

[32] F. Pedregosa, G. Varoquaux, A. Gramfort, et al., “Scikit-Learn: Machine Learning in Python,” Journal of Machine Learning Research 12 (2011): 2825–2830, https://www.jmlr.org/papers/volume12/pedregosa11a/pedregosa11a.pdf.

[33] T. Head, M. Kumar, H. Nahrstaedt, et al., Scikit-Optimize: Bayesian Optimization in Python (Zenodo, 2020), https://zenodo.org/records/4014775.

[34] S. M. Lundberg, G. Erion, H. Chen, et al., “From Local Explanations to Global Understanding With Explainable AI for Trees,” Nature Machine Intelligence 2, no. 1 (2020): 56–67, https://doi.org/10.1038/s42256-019-0138-9.

[35] T. Lookman, P. V. Balachandran, D. Xue, and R. Yuan, “Active Learning in Materials Science With Emphasis on Adaptive Sampling Using Uncertainties for Targeted Design,” npj Computational Materials 5, no. 1 (2019): 21, https://doi.org/10.1038/s41524-019-0153-8.

[36] N. Srinivas, A. Krause, S. M. Kakade, and M. W. Seeger, “Information-Theoretic Regret Bounds for Gaussian Process Optimization in the Bandit Setting,” IEEE Transactions on Information Theory 58, no. 5 (2012): 3250–3265, https://doi.org/10.1109/tit.2011.2182033.

[37] A. S. Raihan, H. Khosravi, S. Das, et al., “Accelerating Material Discovery With a Threshold-Driven Hybrid Acquisition Policy-Based Bayesian Optimization,” Manufacturing Letters 42 (2024): 116–121, https://doi.org/10.1016/j.mfglet.2024.09.157.

[38] S. Wager, T. Hastie, and B. Efron, “Confidence Intervals for Random Forests: The Jackknife and the Infinitesimal Jackknife,” Journal of Machine Learning Research 15, no. 1 (2014): 1625–1651.

[39] J. Carrete, H. Montes-Campos, R. Wanzenböck, E. Heid, and G. K. H. Madsen, “Deep Ensembles vs. Committees for Uncertainty Estimation in Neural-Network Force Fields: Comparison and Application to Active Learning,” Journal of Chemical Physics 158, no. 20 (2023): 204801, https://doi.org/10.1063/5.0146905.

[40] H. Gietl, A. V. Müller, J. W. Coenen, et al., “Textile Preforms for Tungsten Fibre-Reinforced Composites,” Journal of Composite Materials 52, no. 28 (2018): 3875–3884, https://doi.org/10.1177/0021998318771149.

[41] C. Hou, X. Song, F. Tang, et al., “W–Cu Composites With Submicron- and Nanostructures: Progress and Challenges,” NPG Asia Materials 11, no. 1 (2019): 74, https://doi.org/10.1038/s41427-019-0179-x.

[42] T. Han, C. Hou, Z. Zhao, et al., “W–Cu Composites With Excellent Comprehensive Properties,” Composites Part B: Engineering 233 (2022): 109664, https://doi.org/10.1016/j.compositesb.2022.109664.

[43] N. Dehdari Ebrahimi and Y. S. Ju, “Thermal Conductivity of Sintered Copper Samples Prepared Using 3D Printing-Compatible Polymer Composite Filaments,” Additive Manufacturing 24 (2018): 479–485, https://doi.org/10.1016/j.addma.2018.10.025.

[44] N. Castaneda, B. Wisner, J. Cuadra, S. Amini, and A. Kontsos, “Investigation of the Z-Binder Role in Progressive Damage of 3D Woven Composites,” Composites Part A: Applied Science and Manufacturing 98 (2017): 76–89, https://doi.org/10.1016/j.compositesa.2016.11.022.

[45] M. N. Saleh, A. Yudhanto, G. Lubineau, and C. Soutis, “The Effect of z-Binding Yarns on the Electrical Properties of 3D Woven Composites,” Composite Structures 182 (2017): 606–616, https://doi.org/10.1016/j.compstruct.2017.09.081.

Rights & permissions

2026 The Author(s). Materials Genome Engineering Advances published by Wiley-VCH GmbH on behalf of University of Science and Technology Beijing.