자원환경지질 (Economic and Environmental Geology)

  • 제53권4호
  • /
  • Pages.479-489
  • /
  • 2020
  • /
  • 1225-7281(pISSN)
  • /
  • 2288-7962(eISSN)
대한자원환경지질학회 (The Korean Society of Economic and Environmental Geology)
권호 표지
DOI QR코드

DOI QR Code

머신러닝을 사용한 단층 탐지 기술 연구 동향 분석

Research Trend Analysis for Fault Detection Methods Using Machine Learning

  • 배우람 (부경대학교 에너지자원공학과) ;
  • 하완수 (부경대학교 에너지자원공학과)
  • Bae, Wooram (Department of Energy Resources Engineering, Pukyong National University) ;
  • Ha, Wansoo (Department of Energy Resources Engineering, Pukyong National University)
  • 투고 : 2020.03.20
  • 심사 : 2020.06.29
  • 발행 : 2020.08.28
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

단층은 근원암에서 형성된 석유 가스 등의 탄화수소가 이동하는 통로이자 탄화수소를 가두는 덮개암의 역할을 할 수 있는 지질구조로, 탄화수소가 축적된 저류층을 찾기 위한 탄성파 탐사의 주요 대상 중 하나이다. 하지만 기존의 유사성, 응집성, 분산, 기울기, 단층가능성 등 탄성파 자료의 측면 방향 불연속성을 활용하는 단층 감지 방법들은 전문지식을 갖춘 해석자가 많은 계산 비용과 시간을 투자해야 한다는 문제가 있다. 따라서 많은 연구자들이 단층 해석에 필요한 계산 비용과 시간을 절약하기 위한 다양한 연구를 진행하고 있고, 최근에는 머신러닝 기술을 활용한 연구들이 활발히 수행되고 있다. 단층 해석에는 다양한 머신러닝 기술들 중 서포트백터머신, 다층퍼셉트론, 심층 신경망, 합성곱 신경망 등의 알고리즘이 사용되고 있다. 특히 합성곱 신경망을 활용한 연구는 독자적인 구조의 모델을 사용한 연구뿐만 아니라, 이미지 처리 분야에서 성능이 검증된 모델을 활용한 연구 및 단층의 위치와 주향, 경사 등의 정보를 함께 해석하는 연구도 활발히 진행되고 있다. 이 논문에서는 이러한 연구들을 조사하고 분석하여, 현재까지 단층 위치 및 단층 정보 해석에 가장 효과적인 기술은 영상 처리 분야에서 검증된 U-Net 구조를 바탕으로 한 합성곱 신경망인 것을 확인했다. 이러한 합성곱 신경망에 전이학습 및 데이터 증식 기법을 접목하면 앞으로 더욱 효과적인 단층 감지 및 정보 해석이 가능할 것으로 기대된다.

A fault is a geological structure that can be a migration path or a cap rock of hydrocarbon such as oil and gas, formed from source rock. The fault is one of the main targets of seismic exploration to find reservoirs in which hydrocarbon have accumulated. However, conventional fault detection methods using lateral discontinuity in seismic data such as semblance, coherence, variance, gradient magnitude and fault likelihood, have problem that professional interpreters have to invest lots of time and computational costs. Therefore, many researchers are conducting various studies to save computational costs and time for fault interpretation, and machine learning technologies attracted attention recently. Among various machine learning technologies, many researchers are conducting fault interpretation studies using the support vector machine, multi-layer perceptron, deep neural networks and convolutional neural networks algorithms. Especially, researchers use not only their own convolution networks but also proven networks in image processing to predict fault locations and fault information such as strike and dip. In this paper, by investigating and analyzing these studies, we found that the convolutional neural networks based on the U-Net from image processing is the most effective one for fault detection and interpretation. Further studies can expect better results from fault detection and interpretation using the convolutional neural networks along with transfer learning and data augmentation.

키워드

참고문헌

  1. Aqrawi, A.A. and Boe, T.H. (2011). Improved fault segmentation using a dip guided and modified 3D Sobel filter. In SEG Technical Program Expanded Abstracts 2011 (pp. 999-1003). Society of Exploration Geophysicists.
  2. Araya-Polo, M., Dahlke, T., Frogner, C., Zhang, C., Poggio, T. and Hohl, D. (2017). Automated fault detection without seismic processing. The Leading Edge, v.36(3), p.208-214. https://doi.org/10.1190/tle36030208.1
  3. Chang, D.K., Yang, W.Y., Yong, X.S., Li, H.S., Wang, Y.H. and Chen, D.W. (2019, December). Semantic segmentation network for 3D seismic fault system detection. In SEG 2019 Workshop: Fractured Reservoir & Unconventional Resources Forum: Prospects and Challenges in the Era of Big Data, Lanzhou, China, 1-3 September 2019 (pp. 113-116). Society of Exploration Geophysicists.
  4. Chang, D., Yang, W., Yong, X. and Yang, Q. (2018, December). Seismic fault detection using deep learning technology. In International Geophysical Conference, Beijing, China, 24-27 April 2018 (pp. 1770-1773). Society of Exploration Geophysicists and Chinese Petroleum Society.
  5. Choi, Y., Seol, S.J., Byun, J. and Kim, Y. (2019). Vertical resolution enhancement of seismic data with convolutional U-net. In SEG Technical Program Expanded Abstracts 2019 (pp. 2388-2392). Society of Exploration Geophysicists.
  6. Cunha, A., Pochet, A., Lopes, H. and Gattass, M. (2020). Seismic fault detection in real data using transfer learning from a convolutional neural network pretrained with synthetic seismic data. Computers and Geosciences, .135, 104344. https://doi.org/10.1016/j.cageo.2019.104344
  7. Di, H. (2018). Developing a seismic pattern interpretation network (SpiNet) for automated seismic interpretation. arXiv preprint arXiv:1810.08517.
  8. Di, H., Shafiq, M.A. and AlRegib, G. (2017). Seismic-fault detection based on multiattribute support vector machine analysis. In SEG Technical Program Expanded Abstracts 2017 (pp. 2039-2044). Society of Exploration Geophysicists.
  9. Di, H., Shafiq, M.A., Wang, Z. and AlRegib, G. (2019). Improving seismic fault detection by super-attributebased classification. Interpretation, v.7(3), SE251-SE267. https://doi.org/10.1190/INT-2018-0188.1
  10. Fehler, M. and Larner, K. (2008). SEG advanced modeling (SEAM): Phase I first year update. The Leading Edge, v.27(8), p.1006-1007. https://doi.org/10.1190/1.2967551
  11. Geron, A. (2017). Hands-On Machine Learning with Scikit-Learn and TensorFlow. O'Reilly Media, California, United States
  12. Gersztenkorn, A. and Marfurt, K.J. (1999). Eigenstructurebased coherence computations as an aid to 3-D structural and stratigraphic mapping. Geophysics, v.64(5), p.1468-1479. https://doi.org/10.1190/1.1444651
  13. Guo, B., Liu, L. and Luo, Y. (2018, December). Automatic seismic fault detection with convolutional neural network. In International Geophysical Conference, Beijing, China, 24-27 April 2018 (pp. 1786-1789). Society of Exploration Geophysicists and Chinese Petroleum Society.
  14. Hale, D. (2009). Structure-oriented smoothing and semblance. CWP report, 635(635).
  15. Hale, D. (2013). Methods to compute fault images, extract fault surfaces, and estimate fault throws from 3D seismic images. Geophysics, v.78(2), O33-O43. https://doi.org/10.1190/geo2012-0331.1
  16. Huang, L., Dong, X. and Clee, T.E. (2017). A scalable deep learning platform for identifying geologic features from seismic attributes. The Leading Edge, v.36(3), p.249-256. https://doi.org/10.1190/tle36030249.1
  17. Hwang, H.S., Lee, S.K., Lee, T.S. and Sung, N.H. (2000). Minimisation Technique for Seismic Noise Using a Neural Network. Geophysics and Geophysical Exploration, v.3(3), p.83-87.
  18. Karimi, P., Fomel, S., Wood, L. and Dunlap, D. (2015). Predictive coherence: Interpretation, 3. SAE1-SAE7, http://dx. doi. org/10.1190/INT-2015-0030.1.
  19. Kim, T.Y. and Yoon, W.J. (1999). Seismic Traveltime Tomography using Neural Network. Geophysics and Geophysical Exploration, v.2(4), p.167-173.
  20. Lee, H. and Shin, C.H. (2019a). Investigation of Advanced Seismic Interpretation Using Machine Learning Technology. Proceedings of Fall Meeting, The Korean Institute of Gas, p.126-126.
  21. Lee, H. and Shin, C.H. (2019b). Investigation of Quality Improvement Techniques of Seismic Data Using Machine Learning Technology. Proceedings of Fall Meeting, The Korean Institute of Gas, p.124-124.
  22. Lee, H., Mo, C.H., Park S.S. and Shin, C.H. (2018). Methods to Improve the Quality of Seismic Data Using Machine Learning Techniques. Proceedings of Fall Meeting, The Korean Institute of Gas, p.151-151.
  23. Li, F. and Lu, W. (2014). Coherence attribute at different spectral scales. Interpretation, v.2(1), SA99-SA106. https://doi.org/10.1190/INT-2013-0089.1
  24. Li, S., Yang, C., Sun, H. and Zhang, H. (2019). Seismic fault detection using an encoder-decoder convolutional neural network with a small training set. Journal of Geophysics and Engineering, v.16(1), p.175-189. https://doi.org/10.1093/jge/gxy015
  25. Marfurt, K.J., Kirlin, R.L., Farmer, S.L. and Bahorich, M.S. (1998). 3-D seismic attributes using a semblancebased coherency algorithm. Geophysics, v.63(4), p.1150-1165. https://doi.org/10.1190/1.1444415
  26. Marfurt, K.J., Sudhaker, V., Gersztenkorn, A., Crawford, K.D. and Nissen, S.E. (1999). Coherency calculations in the presence of structural dip. Geophysics, v.64(1), p.104-111. https://doi.org/10.1190/1.1444508
  27. Park, J., Yoon, D., Seol, S.J. and Byun, J. (2019). Reconstruction of seismic field data with convolutional UNet considering the optimal training input data. In SEG Technical Program Expanded Abstracts 2019 (pp. 4650-4654). Society of Exploration Geophysicists.
  28. Pochet, A., Diniz, P.H., Lopes, H. and Gattass, M. (2018). Seismic fault detection using convolutional neural networks trained on synthetic poststacked amplitude maps. IEEE Geoscience and Remote Sensing Letters, v.16(3), p.352-356. https://doi.org/10.1109/lgrs.2018.2875836
  29. Randen, T., Pedersen, S.I. and Sonneland, L. (2001). Automatic extraction of fault surfaces from threedimensional seismic data. In SEG Technical Program Expanded Abstracts 2001 (pp. 551-554). Society of Exploration Geophysicists.
  30. Ronneberger, O., Fischer, P. and Brox, T. (2015, October). U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention (pp. 234-241). Springer, Cham.
  31. Van Bemmel, P.P. and Pepper, R.E. (2000). Seismic signal processing method and apparatus for generating a cube of variance values., U.S. Patent No. 6,151,555.
  32. Wu, X. (2017). Directional structure-tensor-based coherence to detect seismic faults and channels. Geophysics, v.82(2), p.A13-A17. https://doi.org/10.1190/geo2016-0473.1
  33. Wu, X. and Hale, D. (2016). 3D seismic image processing for faults. Geophysics, v.81(2), IM1-IM11. https://doi.org/10.1190/geo2015-0380.1
  34. Wu, X., Liang, L., Shi, Y. and Fomel, S. (2019a). FaultSeg3D: Using synthetic data sets to train an end-to-end convolutional neural network for 3D seismic fault segmentation. Geophysics, v.84(3), IM35-IM45. https://doi.org/10.1190/geo2018-0646.1
  35. Wu, X., Liang, L., Shi, Y., Geng, Z. and Fomel, S. (2019b). Multitask learning for local seismic image processing: fault detection, structure-oriented smoothing with edge-preserving, and seismic normal estimation by using a single convolutional neural network. Geophysical Journal International, v.219(3), p.2097-2109. https://doi.org/10.1093/gji/ggz418
  36. Wu, X., Shi, Y., Fomel, S. and Liang, L. (2018). Convolutional neural networks for fault interpretation in seismic images. In SEG Technical Program Expanded Abstracts 2018 (pp. 1946-1950). Society of Exploration Geophysicists.
  37. Wu, X., Shi, Y., Fomel, S., Liang, L., Zhang, Q. and Yusifov, A.Z. (2019c). FaultNet3D: predicting fault probabilities, strikes, and dips with a single convolutional neural network. IEEE Transactions on Geoscience and Remote Sensing, v.57(11), p.9138-9155. https://doi.org/10.1109/tgrs.2019.2925003
  38. Xiong, W., Ji, X., Ma, Y., Wang, Y., AlBinHassan, N.M., Ali, M.N. and Luo, Y. (2018). Seismic fault detection with convolutional neural network. Geophysics, v.83(5), O97-O103. https://doi.org/10.1190/geo2017-0666.1
  39. Zhao, T. (2019). 3D convolutional neural networks for efficient fault detection and orientation estimation. In SEG Technical Program Expanded Abstracts 2019 (pp. 2418-2422). Society of Exploration Geophysicists.
  40. Zhao, T. and Mukhopadhyay, P. (2018). A fault detection workflow using deep learning and image processing. In SEG Technical Program Expanded Abstracts 2018 (pp. 1966-1970). Society of Exploration Geophysicists.
  41. Zheng, Z.H., Kavousi, P. and Di, H.B. (2014). Multiattributes and neural network-based fault detection in 3D seismic interpretation. In Advanced Materials Research (Vol. 838, pp. 1497-1502). Trans Tech Publications Ltd. https://doi.org/10.4028/www.scientific.net/AMR.838-841.1497
  42. Zhou, R., Cai, Y., Yu, F. and Hu, G. (2019). Seismic fault detection with iterative deep learning. In SEG Technical Program Expanded Abstracts 2019 (pp. 2503-2507). Society of Exploration Geophysicists.