High-Order Statistical Generative Model: Reconstruction of Anisotropic Porous Materials

doi:10.16552/j.cnki.issn1001-1625.2025.0770

Abstract

Abstract: Porous materials exhibit typical heterogeneous characteristics, with their macroscopic properties fundamentally governed by microstructural features. Therefore, establishing an accurate microstructure characterization system and realizing high-fidelity numerical modeling were the core scientific issues to reveal the structure-activity relationship of materials. This study proposes a deep learning-based reconstruction method for multiphase anisotropic porous microstructure. By integrating a deep convolutional generative adversarial network model with a slice sampling strategy, and decomposing the complex multiphase reconstruction task into a series of two-phase reconstruction tasks based on the One-hot encoding principle, the proposed method efficiently reconstruct multiphase anisotropic porous materials. In addition, a three-point statistical correlation function algorithm is introduced as a high-order statistical controller for the reconstruction model, addressing the limitations of traditional second-order statistical descriptors and enabling precise quantification of high-order microstructural features. This work provides valuable theoretical insights and practical guidance for performance prediction and microstructure optimization of porous materials.

Key words: porous material, multiphase, microstructure, reconstruction, high-order statistic, deep convolutional generative adversarial network model

CLC Number:

TB12

XU Nuo, WANG Fengjuan, WU Haotian, CHEN Jie, JIANG Jinyang, XU Wenxiang. High-Order Statistical Generative Model: Reconstruction of Anisotropic Porous Materials[J]. BULLETIN OF THE CHINESE CERAMIC SOCIETY, 2025, 44(11): 4274-4282.

References

[1] TAHMASEBI P. A state-of-the-art review of experimental and computational studies of granular materials: properties, advances, challenges, and future directions[J]. Progress in Materials Science, 2023, 138: 101157.
[2] SKOLNICK M, TORQUATO S. Simulated diffusion spreadability for characterizing the structure and transport properties of two-phase materials[J]. Acta Materialia, 2023, 250: 118857.
[3] HUA R Y, WU Y, JIANG J Y, et al. Mesomechanical insights into ITZ volume fraction around realistic-morphological aggregates and elastic properties of concrete[J]. Construction and Building Materials, 2025, 489: 142428.
[4] SAHIMI M, TAHMASEBI P. Reconstruction, optimization, and design of heterogeneous materials and media: basic principles, computational algorithms, and applications[J]. Physics Reports, 2021, 939: 1-82.
[5] YEONG C L Y, TORQUATO S. Reconstructing random media[J]. Physical Review E, 1998, 57(1): 495-506.
[6] BHADURI A, GUPTA A, OLIVIER A, et al. An efficient optimization based microstructure reconstruction approach with multiple loss functions[J]. Computational Materials Science, 2021, 199: 110709.
[7] CHEN D D, TENG Q Z, HE X H, et al. Stable-phase method for hierarchical annealing in the reconstruction of porous media images[J]. Physical Review E, 2014, 89: 013305.
[8] STREBELLE S. Conditional simulation of complex geological structures using multiple-point statistics[J]. Mathematical Geology, 2002, 34(1): 1-21.
[9] ALEXANDER S K, FIEGUTH P, IOANNIDIS M A, et al. Hierarchical annealing for synthesis of binary images[J]. Mathematical Geosciences, 2009, 41(4): 357-378.
[10] TAHMASEBI P, SAHIMI M. Reconstruction of three-dimensional porous media using a single thin section[J]. Physical Review E, 2012, 85(6): 066709.
[11] TAHMASEBI P, SAHIMI M. Cross-correlation function for accurate reconstruction of heterogeneous media[J]. Physical Review Letters, 2013, 110(7): 078002.
[12] LI Y, HAN G H, JIAN P P. 3D reconstruction of unlimited-size real-world porous media by combining a BicycleGAN-based multimodal dictionary and super-dimension reconstruction[J]. Geoenergy Science and Engineering, 2023, 228: 212005.
[13] LI Y, TENG Q Z, HE X H, et al. Super-dimension-based three-dimensional nonstationary porous medium reconstruction from single two-dimensional image[J]. Journal of Petroleum Science and Engineering, 2019, 174: 968-983.
[14] LI Y, TENG Q Z, HE X H, et al. Dictionary optimization and constraint neighbor embedding-based dictionary mapping for superdimension reconstruction of porous media[J]. Physical Review E, 2019, 99(6): 062134.
[15] CHAN S, ELSHEIKH A H. Exemplar-based synthesis of geology using kernel discrepancies and generative neural networks[J]. ArXiv Preprint ArXiv:1809.07748, 2018.
[16] XU N, CHEN S H. Establishing the interfacial microstructure-behavior relations in composites via stochastic morphology reconstruction and deep learning[J]. Acta Materialia, 2023, 244: 118582.
[17] FENG J X, HE X H, TENG Q Z, et al. Reconstruction of porous media from extremely limited information using conditional generative adversarial networks[J]. Physical Review E, 2019, 100(3): 033308.
[18] JIAO Y, STILLINGER F H, TORQUATO S. A superior descriptor of random textures and its predictive capacity[J]. Proceedings of the National Academy of Sciences of the United States of America, 2009, 106(42): 17634-17639.
[19] CHEN D D, XU Z, WANG X R, et al. Fast reconstruction of multiphase microstructures based on statistical descriptors[J]. Physical Review E, 2022, 105(5): 055301.
[20] TAHMASEBI P, SAHIMI M. Cross-correlation function for accurate reconstruction of heterogeneous media[J]. Physical Review Letters, 2013, 110(7): 078002.
[21] GAO M L, HE X H, TENG Q Z, et al. Reconstruction of three-dimensional porous media from a single two-dimensional image using three-step sampling[J]. Physical Review E, 2015, 91: 013308.
[22] XIA Z X, TENG Q Z, WU X H, et al. Three-dimensional reconstruction of porous media using super-dimension-based adjacent block-matching algorithm[J]. Physical Review E, 2021, 104(4): 045308.
[23] ABDOLAHNEJAD M, LIU P X. Deep learning for face image synthesis and semantic manipulations: a review and future perspectives[J]. Artificial Intelligence Review, 2020, 53(8): 5847-5880.
[24] CHAN S, ELSHEIKH A H. Parametric generation of conditional geological realizations using generative neural networks[J]. Computational Geosciences, 2019, 23(5): 925-952.
[25] MOSSER L, DUBRULE O, BLUNT M J. Conditioning of generative adversarial networks for pore and reservoir scale models[C]//80th EAGE Conference and Exhibition 2018. Copenhagen, Denmark. EAGE Publications BV, 2018: (1): 1-5.
[26] FENG J X, TENG Q Z, LI B, et al. An end-to-end three-dimensional reconstruction framework of porous media from a single two-dimensional image based on deep learning[J]. Computer Methods in Applied Mechanics and Engineering, 2020, 368: 113043.
[27] XIA P F, BAI H L, ZHANG T. Multi-scale reconstruction of porous media based on progressively growing generative adversarial networks[J]. Stochastic Environmental Research and Risk Assessment, 2022, 36(11): 3685-3705.
[28] CAO D P, HOU Z Y, LIU Q, et al. Reconstruction of three-dimension digital rock guided by prior information with a combination of InfoGAN and style-based GAN[J]. Journal of Petroleum Science and Engineering, 2022, 208: 109590.
[29] XU W X, WU H T, SUN Q, et al. Multi-criteria guided deep learning for 3D microstructure reconstruction of cementitious materials from a single 2D image: application to transport properties[J]. Construction and Building Materials, 2024, 447: 138164.
[30] HLUSHKOU D, TALLAREK U. Analysis of microstructure-effective diffusivity relationships for the interparticle pore space in physically reconstructed packed beds[J]. Journal of Chromatography A, 2018, 1581/1582: 173-179.
[31] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, ET AL. Generative adversarial networks[C]. Advances in Neural Information Processing Systems, 2014, 3: 2672-2680.
[32] RADFORD A, METZ L, CHINTALA S. Unsupervised representation learning with deep convolutional generative adversarial networks[J]. arXiv preprint arXiv:1511.06434, 2015.
[33] GULRAJANI I, AHMED F, ARJOVSKY M, et al. Improved training of wasserstein GANs[C]. Advances in Neural Information Processing Systems, 2017, 30.
[34] GAYON-LOMBARDO A, MOSSER L, BRANDON N P, et al. Pores for thought: generative adversarial networks for stochastic reconstruction of 3D multi-phase electrode microstructures with periodic boundaries[J]. NPJ Computational Materials, 2020, 6: 82.
[35] SEIBERT P, RAßLOFF A, KALINA K, et al. Microstructure characterization and reconstruction in Python: mcrpy[J]. Integrating Materials and Manufacturing Innovation, 2022, 11(3): 450-466.
[36] HASLACH H W Jr. Random heterogeneous materials: microstructure and macroscopic properties[J]. Applied Mechanics Reviews, 2002, 55(4): B62-B63.
[37] CHAN S, ELSHEIKH A H. Parametric generation of conditional geological realizations using generative neural networks[J]. Computational Geosciences, 2019, 23(5): 925-952.
[38] COVENEY P V, MAILLET J B, WILSON J L, et al. Lattice-gas simulations of ternary amphiphilic fluid flow in porous media[J]. International Journal of Modern Physics C, 1998, 9(8): 1479-1490.