Прогнозирование пористости стандартных образцов керна на основе микротомографии, разметки с учителем и трансформера со сдвигаемыми окнами

Недавние достижения в области машинного обуче­ния позволили автоматически анализировать изображе­ния микротомографии (микро-КТ), способствуя более эффективной идентификации свойств горных пород. Целью данного исследования является прогнозирование экспериментально измеренной открытой пористости пород-коллекторов с использованием изображений микро-КТ стандартных образцов керна. Был собран на­бор данных из 136 образцов керна, включая 49 образцов песчаника и 87 образцов карбоната. Открытая пористость была экспериментально определена с использованием газового волюметра. Образцы керна (30 ± 1 мм в высо­ту и диаметр) были отсканированы с помощью микро­КТ с разрешением 34,6-38,0 мкм, что дало 16-битные стеки изображений. Набор данных состоял из 100 232 изображений (64 119 карбоната и 36 113 песчаника). Для маркировки изображений мы ввели контролируе­мый метод под названием «Сегментация неразрешенных пор с помощью экспериментального эталона» (SUPER), который сегментирует темные воксели для соответствия экспериментально измеренной открытой пористости, адаптируясь к характеристикам каждого образца. Были обучены три модели трансформера со сдвигаемыми окнами (Swin): универсальная модель и специализи­рованные модели для песчаника и карбоната. Модели использовали трансферное обучение с весами ImageNet, за которым последовала тонкая настройка. Тестирование подтвердило, что специализированные модели превзошли универсальную модель. Это подчеркивает, что обучение ансамбля моделей, адаптированных к определенным типам пород, приводит к лучшей производительности, чем одна общая модель для прогнозирования пористости. Основная проблема возникла с песчаниками, особенно мелкозернистыми типами, где мелкие поры сливались из- за ограничений разрешения. Последующая работа должна быть направлена на улучшение разрешения изображений и непосредственное введение детализированных изо­бражений в модель. Метод имеет потенциал применения для полноразмерного керна и ранней оценки пористости в неэкстрагированных стандартных образцах, включая хрупкие коллекторы с нефтью или битумами.

1. Alqahtani N., Alzubaidi F., Armstrong R.T., Swietojanski P. & Mostaghimi P. (2020). Machine learning for predicting properties of porous media from 2d X-ray images. Journal of Petroleum Science and Engineering, 184, 106514. https://doi.org/10.1016/j.petrol.2019.106514

2. Alqahtani N.J., Niu Y., Wang Y.Da, Chung T., Lanetc Z., Zhuravljov A., Armstrong R.T. & Mostaghimi P. (2022). Super-Resolved Segmentation of X-ray Images of Carbonate Rocks Using Deep Learning. Transport in Porous Media, 143(2), pp. 497–525. https://doi.org/10.1007/s11242-022-01781-9

3. Alyafei N., Raeini A.Q., Paluszny A., & Blunt M.J. (2015). A Sensitivity Study of the Effect of Image Resolution on Predicted Petrophysical Properties. Transport in Porous Media, 110(1), pp. 157–169. https://doi.org/10.1007/s11242-015-0563-0

4. Alzahrani M.K., Shapoval A., Chen Z. & Rahman S.S. (2023). PoreGNN: A graph neural network-based framework for predicting flow properties of porous media from micro-CT images. Advances in Geo-Energy Research, 10(1), pp. 39–55. https://doi.org/10.46690/ager.2023.10.05

5. Ambartzumian R.V. (1990). Cavalieri principle and other prerequisities. In Factorization Calculus and Geometric Probability. Cambridge University Press, 286 p. https://doi.org/10.1017/cbo9781139086561.002

6. Andrä H., Combaret N., Dvorkin J., Glatt E., Han J., Kabel M., Keehm Y., Krzikalla F., Lee M., Madonna C., Marsh M., Mukerji T., Saenger E.H., Sain R., Saxena N., Ricker S., Wiegmann A., & Zhan X. (2013). Digital rock physics benchmarks—Part I: Imaging and segmentation. Computers & Geosciences, 50, pp. 25–32. https://doi.org/10.1016/j.cageo.2012.09.005

7. Blunt M.J., Bijeljic B., Dong H., Gharbi O., Iglauer S., Mostaghimi P., Paluszny A. & Pentland C. (2013). Pore-scale imaging and modelling. Advances in Water Resources, 51, pp. 197–216. https://doi.org/10.1016/j.advwatres.2012.03.003

8. Brenne E.O., Dahl V.A., & Jørgensen P.S. (2021). A physical model for microstructural characterization and segmentation of 3D tomography data. Materials Characterization, 171, 110796. https://doi.org/10.1016/j.matchar.2020.110796

9. Carmignato S., Dewulf W. & Leach R. (2018). Industrial X-Ray Computed Tomography. Springer International Publishing, 369 p. https://doi.org/10.1007/978-3-319-59573-3

10. Chawshin K., Berg C.F., Varagnolo D., & Lopez O. (2021). Lithology classification of whole core CT scans using convolutional neural networks. SN Applied Sciences, 3(6), pp. 1–21. https://doi.org/10.1007/s42452-021-04656-8

11. Chawshin K., Berg C.F., Varagnolo D. & Lopez O. (2022). Automated porosity estimation using CT-scans of extracted core data. Computational Geosciences, 26(3), pp. 595–612. https://doi.org/10.1007/s10596-022-10143-9

12. dos Anjos C.E.M., Avila M.R.V, Vasconcelos A.G.P., Pereira Neta A.M., Medeiros L.C., Evsukoff A. G., Surmas R., & Landau L. (2021). Deep learning for lithological classification of carbonate rock micro-CT images. Computational Geosciences, 25(3), pp. 971–983. https://doi.org/10.1007/s10596-021-10033-6

13. Dullien F.A.L. (1992). Porous Media. Elsevier, 566 p. https://doi.org/10.1016/c2009-0-26184-8

14. Gärttner S., Alpak F. O., Meier A., Ray N. & Frank F. (2023). Estimating permeability of 3D micro-CT images by physics-informed CNNs based on DNS. Computational Geosciences, 27(2), pp. 245–262. https://doi.org/10.1007/s10596-022-10184-0

15. Gerke K.M., Korost D.V., Karsanina M.V., Korost S.R., Vasiliev R.V., Lavrukhin E.V. & Gafurova D.R. (2021). Modern approaches to pore space scale digital modeling of core structure and multiphase flow. Georesursy = Georesources, 23(2), pp. 197–213. https://doi.org/10.18599/GRS.2021.2.20

16. Gong H., Mu T., Li Q., Dai H., Li C., He Z., Wang W., Han F., Tuniyazi A., Li H., Lang X., Li Z., & Wang B. (2022). Swin-Transformer-Enabled YOLOv5 with Attention Mechanism for Small Object Detection on Satellite Images. Remote Sensing, 14(12), 2861. https://doi.org/10.3390/rs14122861

17. Graczyk K.M., Matyka M. (2020). Predicting porosity, permeability, and tortuosity of porous media from images by deep learning. Scientific Reports, 10(1), 21488. https://doi.org/10.1038/s41598-020-78415-x

18. Iklassov Z., Medvedev D., Nazarov O. & Razzokov S. (2022). AI for Porosity and Permeability Prediction from Geologic Core X-Ray Micro-Tomography. ArXiv preprint, pp. 1–7. https://doi.org/10.48550/arXiv.2205.13189

19. Jackson S.J., Niu Y., Manoorkar S., Mostaghimi P., & Armstrong R.T. (2021). Deep learning of multi-resolution X-Ray micro-CT images for multiscale modelling. ArXiv preprint, pp. 1–21. http://arxiv.org/abs/2111.01270

20. Jia Deng, Wei Dong, Socher R., Li-Jia Li, Kai Li, & Li Fei-Fei. (2009). ImageNet: A large-scale hierarchical image database. CVPR09, pp. 248–255. https://doi.org/10.1109/cvprw.2009.5206848

21. Kadyrov R., Glukhov M., Statsenko E. & Galliulin B. (2020). Enigma of ferruginous inclusions in Permian evaporites. Arabian Journal of Geosciences, 13(20), 1058. https://doi.org/10.1007/s12517-020-05995-3

22. Kadyrov R., Nurgaliev D., Saenger E.H., Balcewicz M., Minebaev R., Statsenko E., Glukhov M., Nizamova A., & Galiullin B. (2022). Digital rock physics: Defining the reservoir properties on drill cuttings. Journal of Petroleum Science and Engineering, 210, 110063. https://doi.org/10.1016/j.petrol.2021.110063

23. Kadyrov R., Starovoytov A. & Utemov E. (2020). Application of wavelet analysis of X-Ray computed tomography histogram for phase segmentation. IOP Conference Series: Earth and Environmental Science, 516(1), 012005. https://doi.org/10.1088/1755-1315/516/1/012005

24. Kadyrov R., Statsenko E. & Galiullin B. (2018). The porous space structure of domanik shales in the east of Russian plate. International Multidisciplinary Scientific GeoConference Surveying Geology and Mining Ecology Management, SGEM, 18(1.4), pp. 907–914. https://doi.org/10.5593/sgem2018/1.4/S06.118

25. Kadyrov R., Statsenko E. & Nguyen T.H. (2024). Integrating μCT imaging of core plugs and transfer learning for automated reservoir rock characterization and tomofacies identification. Marine and Petroleum Geology, 168, 107014. https://doi.org/10.1016/j.marpetgeo.2024.107014

26. Kalule R., Abderrahmane H.A., Alameri W. & Sassi M. (2023). Stacked ensemble machine learning for porosity and absolute permeability prediction of carbonate rock plugs. Scientific Reports, 13(1), 9855. https://doi.org/10.1038/s41598-023-36096-2

27. Karimpouli S., Kadyrov R. (2022). Multistep Super Resolution Double-U-net (SRDUN) for enhancing the resolution of Berea sandstone images. Journal of Petroleum Science and Engineering, 216, 110833. https://doi.org/10.1016/j.petrol.2022.110833

28. Karimpouli, S., Kadyrov, R., Siegert, M., & Saenger, E. H. (2024). Applicability of 2D algorithms for 3D characterization in digital rocks physics: an example of a machine learning-based super resolution image generation. Acta Geophysica, 72(2), pp. 861–874. https://doi.org/10.1007/s11600-023-01149-7

29. Khan M.I., Khanal A. (2024). Machine Learning Assisted Prediction of Porosity and Related Properties Using Digital Rock Images. ACS Omega, 9(28), pp. 30205–30223. https://doi.org/10.1021/acsomega.3c10131

30. Kim S., Kim H. (2016). A new metric of absolute percentage error for intermittent demand forecasts. International Journal of Forecasting, 32(3), pp. 669–679. https://doi.org/10.1016/j.ijforecast.2015.12.003

31. Latief, F. D. E., Fauzi, U., Irayani Z. & Dougherty G. (2017). The effect of X-ray micro computed tomography image resolution on flow properties of porous rocks. Journal of Microscopy, 266(1), pp. 69–88. https://doi.org/10.1111/jmi.12521

32. Lavrukhin E.V., Gerke K.M., Romanenko K.A., Abrosimov K.N., & Karsanina M.V. (2021). Assessing the fidelity of neural network-based segmentation of soil XCT images based on pore-scale modelling of saturated flow properties. Soil and Tillage Research, 209, 104942. https://doi.org/10.1016/j.still.2021.104942

33. Lavrukhin E.V., Murygin D.A., Toropov K.V., Khlyupin A.N. & Gerke K.M. (2025). Development of Synthetic Tomography Methods for Porous Media. Mathematical Models and Computer Simulations, 17(1), pp. 34–45. https://doi.org/10.1134/S2070048224700704

34. Liu H., Ma C. & Zhu C. (2022). X-ray Micro CT Based Characterization of Pore-Throat Network for Marine Carbonates from South China Sea. Applied Sciences (Switzerland), 12(5), 2611. https://doi.org/10.3390/app12052611

35. Liu Z., Lin Y., Cao Y., Hu H., Wei Y., Zhang Z., Lin S. & Guo B. (2021). Swin Transformer: Hierarchical Vision Transformer using Shifted Windows. Proceedings of the IEEE International Conference on Computer Vision, pp. 9992–10002. https://doi.org/10.1109/ICCV48922.2021.00986

36. McPhee C., Reed J. & Zubizarreta I. (2015). Developments in Petroleum Science. Volume 64. Core Analysis: A BestPractice Guide (1st Edition). Elsevier Science Ltd, 852 p.

37. Mohammed A.S., Almutahhar M., Sattar K., Alhajeri A., Nazir A. & Ali U. (2023). Deep learning based porosity prediction for additively manufactured laser powder-bed fusion parts. Journal of Materials Research and Technology, 27, pp. 7330–7335. https://doi.org/10.1016/j.jmrt.2023.11.130

38. Naftaly M., Tikhomirov I., Hou P. & Markl D. (2020). Measuring Open Porosity of Porous Materials Using THz-TDS and an Index-Matching Medium. Sensors, 20(11), 3120. https://doi.org/10.3390/s20113120

39. Ortiz A.F., Herrera E.H. & Santos N. (2020). Porosity prediction from X-ray computed tomography logs (RHOB and PEF) using Artificial Neural Networks (ANN). Boletín de Geología, 42(3), pp. 141–149. https://doi.org/10.18273/revbol.v42n3-2020006

40. Paszke A., Gross S., Massa F., Lerer A., Bradbury J., Chanan G., Killeen T., Lin Z., Gimelshein N., Antiga L., Desmaison A., Köpf A., Yang E., DeVito Z., Raison M., Tejani A., Chilamkurthy S., Steiner B., Fang L., Bai J., Chintala S. (2019). PyTorch: An imperative style, high-performance deep learning library. In H. Wallach, H. Larochelle, A. Beygelzimer, F. Alché-Buc, E. Fox, & R. Garnett (Eds.). Advances in Neural Information Processing Systems, 32, pp. 1–12. https://doi.org/10.48550/arXiv.1912.01703

41. Qu H., Peng Y., Huang J., Pan Z. & Zhou F. (2023). Modelling of the impact of stress concentration on permeability in porous medium based on machine learning method. Geoenergy Science and Engineering, 224, 211655. https://doi.org/10.1016/j.geoen.2023.211655

42. Shah S.M., Gray F., Crawshaw J.P. & Boek E.S. (2016). Microcomputed tomography pore-scale study of flow in porous media: Effect of voxel resolution. Advances in Water Resources, 95, pp. 276–287. https://doi.org/10.1016/j.advwatres.2015.07.012

43. Shaik A.R., Al-Ratrout A.A., AlSumaiti A.M. & Jilani A.K. (2019). Rock Classification Based on Micro-CT Images using Machine Learning Techniques. D021S060R003. https://doi.org/10.2118/197651-MS

44. Shan L., Liu C., Liu Y., Kong W. & Hei X. (2022). Rock CT Image Super-Resolution Using Residual Dual-Channel Attention Generative Adversarial Network. Energies, 15(14), 5115. https://doi.org/10.3390/en15145115

45. Song R., Zheng L., Wang Y. & Liu J. (2020). Effects of Pore Structure on Sandstone Mechanical Properties Based on Micro-CT Reconstruction Model. Advances in Civil Engineering, (1), pp. 1–21. https://doi.org/10.1155/2020/9085045

46. Tembely M., AlSumaiti A.M. & Alameri W. (2020). A deep learning perspective on predicting permeability in porous media from network modeling to direct simulation. Computational Geosciences, 24(4), pp. 1541–1556. https://doi.org/10.1007/s10596-020-09963-4

47. Tiab D., Donaldson E.C. (2016). Petrophysics: Theory and Practice of Measuring Reservoir Rock and Fluid Transport Properties: Fourth Edition (Fourth Ed.). Elsevier, 894 p. https://doi.org/10.1016/C2014-0-03707-0

48. Van Geet M., Swennen R. & Wevers M. (2000). Quantitative analysis of reservoir rocks by microfocus X-ray computerised tomography. Sedimentary Geology, 132(1–2), pp. 25–36. https://doi.org/10.1016/S0037-0738(99)00127-X

49. Varfolomeev I., Yakimchuk I. & Safonov I. (2019). An application of deep neural networks for segmentation of microtomographic images of rock samples. Computers, 8(4), 72. https://doi.org/10.3390/computers8040072

50. Wang D., Yang R., Zhang Z., Liu H., Tan J., Li S., Yang X., Wang X., Tang K., Qiao Y. & Su P. (2023). P-Swin: Parallel Swin transformer multiscale semantic segmentation network for land cover classification. Computers and Geosciences, 175, 105340. https://doi.org/10.1016/j.cageo.2023.105340

51. Wang Y.Da, Armstrong R.T. & Mostaghimi P. (2019). Enhancing Resolution of Digital Rock Images with Super Resolution Convolutional Neural Networks. Journal of Petroleum Science and Engineering, 182, 106261. https://doi.org/10.1016/j.petrol.2019.106261

52. Wu H.Y., Kuo C.L., Lin C.Y. & Chu W.T. (2025). Deep learning in the advanced core sample porosity determination with XCT image. Earth Science Informatics, 18(1), 17. https://doi.org/10.1007/s12145-024-01602-1

53. Xu X., Feng Z., Cao C., Li M., Wu J., Wu Z., Shang Y. & Ye S. (2021). An improved swin transformer-based model for remote sensing object detection and instance segmentation. Remote Sensing, 13(23), 4779. https://doi.org/10.3390/rs13234779

54. Yang S. (2017). Fundamentals of Petrophysics. Springer Berlin Heidelberg, 509 p. https://doi.org/10.1007/978-3-662-55029-8

55. Zhang H., Yu H., Yuan X., Xu H., Micheal M., Zhang J., Shu H., Wang G. & Wu H. (2022). Permeability prediction of low-resolution porous media images using autoencoder-based convolutional neural network. Journal of Petroleum Science and Engineering, 208, 109589. https://doi.org/10.1016/j.petrol.2021.109589

56. Zhang L., Jing W., Yang Y., Yang H., Guo Y., Sun H., Zhao J. & Yao J. (2019). The Investigation of Permeability Calculation Using Digital Core Simulation Technology. Energies, 12(17), 3273. https://doi.org/10.3390/en12173273

57. Zhang P., Lee Y.Il, & Zhang J. (2019). A review of high-resolution X-ray computed tomography applied to petroleum geology and a case study. Micron, 124, 102702. https://doi.org/10.1016/j.micron.2019.102702

58. Zheng D., Zhong H., Camps-Valls G., Cao Z., Ma X., Mills B., Hu X., Hou M., & Ma C. (2024). Explainable deep learning for automatic rock classification. Computers & Geosciences, 184, 105511. https://doi.org/10.1016/j.cageo.2023.105511

59. Zollanvari A. (2023). Machine learning with Python: Theory and implementation. Machine Learning with Python: Theory and Implementation. Springer International Publishing, 452 p. https://doi.org/10.1007/978-3-031-33342-2