1887
  1. Home
  2. A-Z Publications
  3. Geophysical Prospecting
  4. Volume 72, Issue 3
  5. Article
Volume 72, Issue 3
  • E-ISSN: 1365-2478

Abstract

Abstract

The carbonate fault–karst reservoir is a special and significant reservoir in the Shunbei area, and the development of the cave has been controlled by strike‐slip faults. Due to the complex subsurface structures, fault–karst reservoir characterization is generally divided into fault and cave detection tasks. The potential spatial relationships between faults and caves might be neglected by using the separate detection scheme. The multi‐task learning network can perform multiple tasks simultaneously and exploit the potential features of training data by using a deep neural network. In this study, we built fault–karst models based on the geological background of the Shunbei area and synthesized fault–karst training data using three‐dimensional point spread function convolution. Then, we developed a multi‐task learning network to learn fault–karst features and detect faults and caves simultaneously. The test result demonstrates that the multi‐task learning network trained by synthetic fault–karst data can effectively identify the faults and caves in field seismic data. The comparisons of the multi‐task learning network, single‐task learning networks and conventional methods demonstrate the importance of spatial relationships between faults and caves and show the superiority of the multi‐task learning network. This technique could significantly assist in the exploration, development and well deployment for an ultra‐deep carbonate reservoir.

© 2024 European Association of Geoscientists & Engineers. © 2023 European Association of Geoscientists & Engineers.
Loading

Article metrics loading...

/content/journals/10.1111/1365-2478.13460
2024-02-21
2024-12-03

  • From This Site

    /content/journals/10.1111/1365-2478.13460
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Ao, Y., Lu, W., Jiang, B. & Monkam, P. (2020) Seismic structural curvature volume extraction with convolutional neural networks. IEEE Transactions on Geoscience and Remote Sensing, 59(9), 7370–7384.
     [Google Scholar]
  2. Baxter, J. (2000) A model of inductive bias learning. Journal of Artificial Intelligence Research, 12, 149–198.
     [Google Scholar]
  3. Cao, P., Chang, S., Zhu, Y., Shen, J., Qiao, Z., Shao, G. & Sun, X. (2021) New method for characterizing internal structure of fault‐karst reservoirs and analysis on acidizing fracturing effect: a case study in HLHT Oilfield, Tarim Basin, NW China. Lithosphere, 2021(Special 1), 6784641.
     [Google Scholar]
  4. Caruana, R. (1997) Multitask learning. Machine Learning, 28(1), 41–75.
     [Google Scholar]
  5. Caruana, R. (1993) Multitask learning: A knowledge‐based source of inductive bias. In: ICML'93: Proceedings of the tenth international conference on machine learning, Amherst MA, July 27–29, 1993. San Francisco, CA: Morgan Kaufmann, pp. 41–48.
     [Google Scholar]
  6. Chang, X., Wang, T.G., Qiming, L. & Guangxi, O. (2013) Charging of Ordovician reservoirs in the Halahatang depression (Tarim Basin, NW China) determined by oil geochemistry. Journal of Petroleum Geology, 36(4), 383–398.
     [Google Scholar]
  7. Chen, Y., Zhao, L., Pan, J., Li, C., Xu, M., Li, K., Zhang, F. & Geng, J. (2021) Deep carbonate reservoir characterisation using multi‐seismic attributes via machine learning with physical constraints. Journal of Geophysics and Engineering, 18(5), 761–775.
     [Google Scholar]
  8. Deng, S., Zhao, R., Kong, Q., Li, Y. & Li, B. (2022) Two distinct strike‐slip fault networks in the Shunbei area and its surroundings, Tarim Basin: hydrocarbon accumulation, distribution, and controlling factors. AAPG Bulletin, 106(1), 77–102.
     [Google Scholar]
  9. Ding, Y., Du, Q., Fu, L. & Jian, S. (2020) Reliability analysis of seismic attribute in the detection of fault‐karst. Interpretation, 8(4), SP61–SP70.
     [Google Scholar]
  10. Ding, Z., Wang, R., Chen, F., Yang, J., Zhu, Z., et al,. (2020) Origin, hydrocarbon accumulation and oil‐gas enrichment of fault‐karst carbonate reservoirs: a case study of Ordovician carbonate reservoirs in South Tahe area of Halahatang Oilfield, Tarim Basin. Petroleum Exploration and Development, 47(2), 306–317.
     [Google Scholar]
  11. Dorn, G., James, H., Dopkin, D. & Payne, B. (2005) Automatic fault extraction in 3D seismic interpretation. In: 67th EAGE conference & exhibition. Houten, the Netherlands: European Association of Geoscientists & Engineers, pp. cp‐1.
     [Google Scholar]
  12. Drozdzal, M., Vorontsov, E., Chartrand, G., Kadoury, S. & Pal, C. (2016) The importance of skip connections in biomedical image segmentation. In: Deep learning and data labeling for medical applications. Cham: Springer, pp. 179–187.
     [Google Scholar]
  13. Evgeniou, T. & Pontil, M. (2004) Regularized multi‐task learning. In: Proceedings of the tenth ACM SIGKDD international conference on knowledge discovery and data mining. New York: ACM, pp. 109–117.
     [Google Scholar]
  14. Jadon, S. (2020) A survey of loss functions for semantic segmentation. In: 2020 IEEE conference on computational intelligence in bioinformatics and computational biology (CIBCB). Piscataway, NJ: IEEE, pp. 1–7.
     [Google Scholar]
  15. Jensen, K., Lecomte, I., Gelius, L.J. & Kaschwich, T. (2021) Point‐spread function convolution to simulate prestack depth migrated images: a validation study. Geophysical Prospecting, 69(8‐9), 1571–1590.
     [Google Scholar]
  16. Jiao, F. (2017) Significance of oil and gas exploration in NE strike‐slip fault belts in Shuntuoguole area of Tarim Basin. Oil & Gas Geology, 38(5), 831–839.
     [Google Scholar]
  17. Jiao, F. (2018) Significance and prospect of ultra‐deep carbonate fault‐karst reservoirs in Shunbei Area, Tarim Basin. Oil & Gas Geology, 39(2), 207–216.
     [Google Scholar]
  18. Jing, J., Yan, Z., Zhang, Z., Gu, H. & Han, B. (2023) Fault detection using a convolutional neural network trained with point‐spread function‐convolution‐based samples. Geophysics, 88(1), IM1–IM14.
     [Google Scholar]
  19. LeCun, Y., Kavukcuoglu, K. & Farabet, C. (2010) Convolutional networks and applications in vision. In: Proceedings of 2010 IEEE international symposium on circuits and systems. Piscataway, NJ: IEEE, pp. 253–256.
     [Google Scholar]
  20. Li, Y., Song, J., Lu, W., Monkam, P. & Ao, Y. (2020) Multitask learning for super‐resolution of seismic velocity model. IEEE Transactions on Geoscience and Remote Sensing, 59(9), 8022–8033.
     [Google Scholar]
  21. Loshchilov, I. & Hutter, F. (2017) Decoupled weight decayregularization. arXiv [Preprint]. 7th International Conference on Learning Representations. [v3] 2019, ICLR 2019, New Orleans, LA, USA. Available from: https://doi.org/10.48550/arXiv.1711.05101 [Accessed 2022].
     [Google Scholar]
  22. Lu, X., Hu, W., Wang, Y., Li, X., Li, T., Lyu, Y., He, X. & Yang, D. (2015) Characteristics and development practice of fault‐karst carbonate reservoirs in Tahe area, Tarim Basin. Oil & Gas Geology, 36(3), 347–355.
     [Google Scholar]
  23. Lu, X., Wang, Y., Tian, F., Li, X., Yang, D., Li, T., Lv, Y. & He, X. (2017) New insights into the carbonate karstic fault system and reservoir formation in the Southern Tahe area of the Tarim Basin. Marine and Petroleum Geology, 86, 587–605.
     [Google Scholar]
  24. Ma, Y., Xunyu, C., Lu, Y., Zongjie, L., Huili, L., Shang, D. & Peirong, Z. (2022) Practice and theoretical and technical progress in exploration and development of Shunbei ultra‐deep carbonate oil and gas field, Tarim Basin, NW China. Petroleum Exploration and Development, 49(1), 1–20.
     [Google Scholar]
  25. Maurer, A. (2006) Bounds for linear multi‐task learning. The Journal of Machine Learning Research, 7, 117–139.
     [Google Scholar]
  26. Meng, J., Wang, S., Geng, W., Wang, P., Yang, L. & Sang, W. (2022) Multitask learning for pre‐stack three‐parameter inversion. In: 83rd EAGE annual conference & exhibition, volume 2022. Houten, the Netherlands: European Association of Geoscientists & Engineers, pp. 1–5.
     [Google Scholar]
  27. Nair, V. & Hinton, G.E. (2010) Rectified linear units improve restricted boltzmann machines. In: ICML'10: Proceedings of the 27th international conference on machine learning. Madison, WI: Omnipress, pp. 807–814.
     [Google Scholar]
  28. Qi, L. (2021) Structural characteristics and storage control function of the Shun I Fault Zone in the Shunbei region, Tarim Basin. Journal of Petroleum Science and Engineering, 203, 108653.
     [Google Scholar]
  29. Ronneberger, O., Fischer, P. & Brox, T. (2015) U‐net: convolutional networks for biomedical image segmentation. In: International conference on medical image computing and computer‐assisted intervention. Cham: Springer, pp. 234–241.
     [Google Scholar]
  30. Ruder, S. (2017) An overview of multi‐task learning in deep neural networks. arXiv [Preprint]. [v1] 2017. Available from: https://doi.org/10.48550/arXiv.1706.05098 [Accessed 2022].
  31. Ruder, S., Bingel, J., Augenstein, I. & Søgaard, A. (2019) Latent multi‐task architecture learning. Proceedings of the AAAI Conference on Artificial Intelligence, 33, 4822–4829.
     [Google Scholar]
  32. Wang, L., Yang, R., Li, D., Meng, L. & Xiao, Z. (2021) Seismic attributes for characterization and prediction of carbonate faulted karst reservoirs in the Tarim Basin, China. Interpretation, 9(3), T611–T622.
     [Google Scholar]
  33. Wang, Q., Wang, Y., Ao, Y. & Lu, W. (2021) Seismic inversion based on 2D‐CNN and multi‐task learning. In: 82nd EAGE annual conference & exhibition, volume 2021. Houten, the Netherlands: European Association of Geoscientists & Engineers, pp. 1–5.
     [Google Scholar]
  34. Wang, Y., Chen, J., Pang, X., Wang, G., Hu, T., Zhang, B., Huo, Z. & Chen, H. (2017) Hydrocarbon migration along fault intersection zone: a case study on Ordovician carbonate reservoirs in Tazhong area, Tarim Basin, NW China. Geological Journal, 52(5), 832–850.
     [Google Scholar]
  35. Wei, S., Caoyang, H., Xingwei, W. & Kelong, H. (2022) Carbonate karst cave recognition based on U‐Net segmentation. In: 83rd EAGE annual conference & exhibition, volume 2022. Houten, the Netherlands: European Association of Geoscientists & Engineers, pp. 1–5.
     [Google Scholar]
  36. Wei, Y., Li, Y.E., Yang, J., Zong, J., Fang, J. & Fu, H. (2020) Multi‐task learning based P/S wave separation and reverse time migration for VSP. In SEG technical program expanded abstracts 2020. Houston, TX: Society of Exploration Geophysicists, pp. 1671‐1675.
     [Google Scholar]
  37. White, R.E. (1991) Properties of instantaneous seismic attributes. The Leading Edge, 10(7), 26–32.
     [Google Scholar]
  38. Wu, X., Geng, Z., Shi, Y., Pham, N., Fomel, S. & Caumon, G. (2020) Building realistic structure models to train convolutional neural networks for seismic structural interpretation. Geophysics, 85(4), WA27–WA39.
     [Google Scholar]
  39. Wu, X., Liang, L., Shi, Y. & Fomel, S. (2019) Faultseg3d: using synthetic data sets to train an end‐to‐end convolutional neural network for 3D seismic fault segmentation. Geophysics, 84(3), IM35–IM45.
     [Google Scholar]
  40. Wu, X., Liang, L., Shi, Y., Geng, Z. & Fomel, S. (2019) 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, 219(3), 2097–2109.
     [Google Scholar]
  41. Wu, X., Yan, S., Qi, J. & Zeng, H. (2020) Deep learning for characterizing paleokarst collapse features in 3‐D seismic images. Journal of Geophysical Research: Solid Earth, 125(9), e2020JB019685.
     [Google Scholar]
  42. Xu, M., Zhao, L., Gao, S., Zhu, X. & Geng, J. (2022) Joint use of multiseismic information for lithofacies prediction via supervised convolutional neural networks. Geophysics, 87(5), M151–M162.
     [Google Scholar]
  43. Yan, X., Gu, H. & Yan, Y. (2022) Karst cave detection and prediction by using two fully convolutional neural networks. In SEG 2021 workshop: 4th International workshop on mathematical geophysics: traditional & learning, Virtual, 17–19 December 2021. Houston, TX: Society of Exploration Geophysicists, pp. 90–93.
     [Google Scholar]
  44. Yu, J., Li, Z. & Yang, L. (2016) Fault system impact on paleokarst distribution in the Ordovician Yingshan Formation in the Central Tarim Basin, Northwest China. Marine and Petroleum Geology, 71, 105–118.
     [Google Scholar]
  45. Yuan, S., Wang, J., Liu, T., Xie, T. & Wang, S. (2020) 6D phase‐difference attributes for wide‐azimuth seismic data interpretation. Geophysics, 85(6), IM37–IM49.
     [Google Scholar]
  46. Zhang, W., He, Z., Duan, T., Ma, Q., Li, M. & Zhao, H. (2023) Architecture characteristics and characterization methods of fault‐controlled karst reservoirs: a case study of the Shunbei 5 Fault Zone in the Tarim Basin, China. Interpretation, 11(1), SA47–SA62.
     [Google Scholar]
  47. Zhang, W., Taizhong, D., Meng, L., Huawei, Z., Shang, X. & Yan, W. (2021) Architecture characterization of Ordovician fault‐controlled paleokarst carbonate reservoirs in Tuoputai, Tahe Oilfield, Tarim Basin, NW China. Petroleum Exploration and Development, 48(2), 367–380.
     [Google Scholar]
  48. Zhang, Y. & Yang, Q. (2021) A survey on multi‐task learning. IEEE Transactions on Knowledge and Data Engineering, 34(12), 5586–5609.
     [Google Scholar]
  49. Zhao, L., Zou, C., Chen, Y., Shen, W., Wang, Y., Chen, H. & Geng, J. (2021) Fluid and lithofacies prediction based on integration of well‐log data and seismic inversion: a machine‐learning approach. Geophysics, 86(4), M151–M165.
     [Google Scholar]
  50. Zheng, X., Wu, B., Zhu, X. & Zhu, X. (2022) Multi‐task deep learning seismic impedance inversion optimization based on homoscedastic uncertainty. Applied Sciences, 12(3), 1200.
     [Google Scholar]
  51. Zhou, L., Gu, H., Ma, L., Xue, J. & Li, Z. (2017) Seismic response and prediction of Ordovician fractured vuggy carbonate reservoirs in the Tazhong Shunnan well a block. Interpretation, 5(1), SB45–SB55.
     [Google Scholar]
  52. Zhu, G., Milkov, A.V., Zhang, Z., Sun, C., Zhou, X., Chen, F., Han, J. & Zhu, Y. (2019) Formation and preservation of a giant petroleum accumulation in superdeep carbonate reservoirs in the Southern Halahatang oil field area, Tarim Basin, China. AAPG Bulletin, 103(7), 1703–1743.
     [Google Scholar]
/content/journals/10.1111/1365-2478.13460
Loading
Deep carbonate fault–karst reservoir characterization by multi‐task learning
Geophysical Prospecting 72, 1092 (2024); https://doi.org/10.1111/1365-2478.13460
/content/journals/10.1111/1365-2478.13460
/content/journals/10.1111/1365-2478.13460
Loading

Data & Media loading...

  • Article Type: Research Article
Keyword(s): deep learning; fault–karst reservoir; interpretation; multi‐task learning; reservoir geophysics; seismic

Most Cited This Month Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error