Опыт применения сверточных нейронных сетей к обратным задачам сейсмической разведки
Авторы
-
В. И. Голубев
-
М. И. Анисимов
Ключевые слова:
сейсмическая разведка
трещиноватые среды
математическое моделирование
сверточные нейронные сети
совместное машинное обучение
Аннотация
Работа посвящена исследованию возможности применения современных сверточных нейронных сетей для решения задач восстановления положения геологических включений и оценки скалярных параметров используемых моделей по данным сейсмической разведки. Для формирования обучающих и валидационных выборок используются синтетические сейсмограммы, рассчитанные явно-неявными сеточно-характеристическими схемами. В работе рассмотрены две архитектуры сетей для задач совместного машинного обучения, проведено сравнение результатов рассчитываемых оценок с моделями одиночного прогноза. Продемонстрировано значимое повышение качества прогноза.
Раздел
Методы и алгоритмы вычислительной математики и их приложения
Библиографические ссылки
- M. E. Badley, Practical Seismic Interpretation (IHRDC Publishers, Boston, 1985).
- B. Russel and D. Hampson, “Comparison of Poststack Seismic Inversion Methods,” SEG Tech. Program Expand. Abstr. 1991, 876-878 (1991).
doi 10.1190/1.1888870
- D. Kosloff, J. Sherwood, Z. Koren, et al., “Velocity and Interface Depth Determination by Tomography of Depth Migrated Gathers,” Geophysics 61 (8), 1511-1523 (1996).
doi 10.1190/1.1444076
- G. Roethe and A. Tarantola, “Use of Neural Networks for Inversion of Seismic Data,” SEG Tech. Program Expan. Abstr. 1991, 302-305 (1991).
doi 10.1190/1.1888938
- G. Röth and A. Tarantola, “Neural Networks and Inversion of Seismic Data,” J. Geophys. Res. Solid Earth 99 (B4), 6753-6768 (1994).
doi 10.1029/93JB01563
- E. P. Leite, A. C. Vidal, J. F. Bueno, and R. D. C. Duarte, “3D Acoustic Impedance and Porosity Mapping from Seismic Inversion and Neural Networks,” SEG Tech. Program Expand. Abstr. 2010, 2226-2230 (2010).
doi 10.1190/1.3513291
- D. Cao, P. An, and S. Liu, “Elastic-Parameters Inversion from EI Based on the Deep-Learning Method,” SEG Tech. Program Expand. Abstr. 2018, 640-644 (2018).
doi 10.1190/segam2018-2998479.1
- S. Eladj, M. Z. Doghmane, L. Aliouane, and S.-A. Ouadfeul, “Porosity Model Construction Based on ANN and Seismic Inversion: A Case Study of Saharan Field (Algeria),” in Advances in Geophysics, Tectonics and Petroleum Geosciences (Springer, Cham, 2022), pp. 241-243.
doi 10.1007/978-3-030-73026-0_55
- W. Lewis and D. Vigh, “Deep Learning Prior Models from Seismic Images for Full-Waveform Inversion,” SEG Tech. Program Expand. Abstr. 2017, 1512-1517 (2017).
doi 10.1190/segam2017-17627643.1
- B. Wu, D. Meng, L. Wang, et al., “Seismic Impedance Inversion Using Fully Convolutional Residual Network and Transfer Learning,” IEEE Geosci. Remote Sens. Lett. 17 (12), 2140-2144 (2020).
doi 10.1109/LGRS.2019.2963106
- V. C. Dodda, L. Kuruguntla, S. Razak, et al., “Seismic Lithology Interpretation Using Attention Based Convolutional Neural Networks,” 2023 3rd Int. Conf. on Intelligent Communication and Computational Techniques (2023), 1-5.
doi 10.1109/ICCT56969.2023.10075964
- Q. Sun, X. Wang, H. Ni, et al., “Fault Identification of U-Net Based on Enhanced Feature Fusion and Attention Mechanism,” Electronics 12 (12), Article Number 2562 (2023).
doi 10.3390/electronics12122562
- Y.-Q. Wang, Q. Wang, W.-K. Lu, et al., “Seismic Impedance Inversion Based on Cycle-Consistent Generative Adversarial Network,” Pet. Sci. 19 (1), 147-161 (2022).
doi 10.1016/j.petsci.2021.09.038
- V. Golubev and M. Anisimov, “Application of Convolutional Networks for Localization and Prediction of Scalar Parameters of Fractured Geological Inclusion,” Int. J. Appl. Mech. 16 (5), Article Number 2450064 (2024).
doi 10.1142/S1758825124500649
- Y. Zhang and Q. Yang, “An Overview of Multi-Task Learning,” Natl. Sci. Rev. 5 (1), 30-43 (2018).
doi 10.1093/nsr/nwx105
- M. Crawshaw, “Multi-Task Learning with Deep Neural Networks: A Survey,” arXiv: 2009.09796v1 [cs.LG] (2020).
doi 10.48550/arXiv.2009.09796
- G. S. Martin, R. Wiley, and K. J. Marfurt, “Marmousi2: An Elastic Upgrade for Marmousi,” Lead. Edge 25 (2), 156-166 (2006).
doi 10.1190/1.2172306
- I. S. Nikitin, N. G. Burago, V. I. Golubev, and A. D. Nikitin, “Continual Models of Layered and Block Media with Slippage and Delamination,” Procedia Struct. Integr. 23, 125-130 (2019).
doi 10.1016/j.prostr.2020.01.074
- I. S. Nikitin and V. I. Golubev, “Explicit-Implicit Schemes for Calculating the Dynamics of Layered Media with Nonlinear Conditions at Contact Boundaries,” J. Sib. Fed. Univ. Math. Phys. 14 (6), 768-778 (2021).
doi 10.17516/1997-1397-2021-14-6-768-778
- I. B. Petrov, V. I. Golubev, and E. K. Guseva, “Hybrid Grid-Characteristic Schemes for Arctic Seismic Problems,” Dokl. Akad. Nauk 501 (1), 67-73 (2021) [Dokl. Math. 104 (3), 374-379 (2021)].
doi 10.1134/S1064562421060132
- O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” in Lecture Notes in Computer Science (Springer, Cham, 2015), Vol. 9351, pp. 234-241.
doi 10.1007/978-3-319-24574-4_28
- I. Misra, A. Shrivastava, A. Gupta, and M. Hebert, “Cross-Stitch Networks for Multi-Task Learning,” in Proc. IEEE Conf. on Computer Vision and Pattern Recognition , Las Vegas, USA, 2016, pp. 3994-4003.
doi 10.1109/CVPR.2016.433