1887
Volume 39 Number 7
  • ISSN: 0263-5046
  • E-ISSN: 1365-2397

Abstract

Abstract

Quantitative workflows utilizing real-time data to constrain uncertainty have the potential to significantly improve geosteering. Fast updates based on real-time data are particularly important when drilling in complex reservoirs with high uncertainties in pre-drill models. However, practical assimilation of real-time data requires effective geological modelling and mathematically robust parameterization. We propose a generative adversarial deep neural network (GAN), which is trained to reproduce geologically consistent 2D sections of fluvial successions. Offline training produces a fast GAN-based approximation of complex geology parameterized as a 60-dimensional model vector with standard Gaussian distribution of each component. Probabilistic forecasts are generated using an ensemble of equiprobable model vector realizations. A forward-modelling sequence, including a GAN, converts the initial (prior) ensemble of realizations into EM log predictions. An ensemble smoother minimizes statistical misfits between predictions and real-time data, yielding an update of model vectors and reduced uncertainty around the well. Updates can then be translated to probabilistic predictions of facies and resistivities. This paper demonstrates a workflow for geosteering in an outcrop-based synthetic fluvial succession.

In our example, the method reduces uncertainty and correctly predicts most of the major geological features up to 500 m ahead of drill-bit.

The condensed summary is also submitted for presentation at the 3rd EAGE/SPE Geosteering Workshop to be held 2–4 November 2021, online.

Loading

Article metrics loading...

/content/journals/10.3997/1365-2397.fb2021051
2021-07-01
2021-09-19
Loading full text...

Full text loading...

References

  1. Alyaev, S., Suter, E., Bratvold, R.B., Hong, A., Luo, X. and Fossum, K.
    [2019]. A decision support system for multi-target geosteering.Journal of Petroleum Science and Engineering, 183, 106381, 1–14.
    [Google Scholar]
  2. Alyaev, S., Shahriari, M., Pardo, D., Omella, A.J., Larsen, D.S., Jahani, N. and Suter, E.
    [2021]. Modeling extra-deep EM logs using a deep neural network.Geophysics (Just-Accepted Articles).
    [Google Scholar]
  3. Antonsen, F., Teixeira De Oliveira, M.E., Petersen, S.A., Metcalfe, R.W., Hermanrud, K., Constable, M.V. and Nickel, M.
    [2018]. Geosteering in complex mature fields through integration of 3d multi-scale LWD-data, geomodels, surface and time-lapse seismic. In SPWLA 59th Annual Logging Symposium. OnePetro, 1–16.
    [Google Scholar]
  4. Arjovsky, M., Chintala, S. and Bottou, L.
    [2017]. Wasserstein generative adversarial networks. In: International conference on machine learning. PMLR, 214–223.
    [Google Scholar]
  5. Chan, S. and Elsheikh, A. H.
    [2019]. Parametric generation of conditional geological realizations using generative neural networks.Computational Geosciences, 23(5), 925–952.
    [Google Scholar]
  6. Chen, Y. and Oliver, D. S.
    [2013]. Levenberg–Marquardt forms of the iterative ensemble smoother for efficient history matching and uncertainty quantification.Computational Geosciences, 17(4), 689–703.
    [Google Scholar]
  7. Chen, Y., Lorentzen, R. J. and Vefring, E.H.
    [2015]. Optimization of well trajectory under uncertainty for proactive geosteering.SPE Journal, 20(02), 368–383.
    [Google Scholar]
  8. Chen, Y., Luo, X. and Vefring, E.H.
    [2014]. On Bias Correction for Parameter Estimation Problems with Applications to Model Updating for Geosteering. In: ECMOR XIV-14th European Conference on the Mathematics of Oil Recovery. EAGE, 1–19.
    [Google Scholar]
  9. Cole, R. D. and Cumella, S.P.
    [2005]. Sand-body architecture in the lower Williams Fork Formation (Upper Cretaceous), Coal Canyon, Colorado, with comparison to the Piceance Basin subsurface.The Mountain Geologis, 42, 85–107.
    [Google Scholar]
  10. Ellison, A.I.
    [2004]. Numerical modeling of heterogeneity within a fluvial point-bar deposit using outcrop and LIDAR data: Williams Fork Formation, Piceance Basin, Colorado.MSc thesis, University of Colorado. 1–225.
    [Google Scholar]
  11. Fossum, K., Alyaev, S., Tveranger, J. and Elsheikh, A.
    [2021]. Deep learning for prediction of complex geology ahead of drilling.Preprint accepted to LNCS, arXiv:2104.02550, 1–14.
    [Google Scholar]
  12. Goodfellow, I.J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville A. and Bengio, Y.
    [2014]. Generative adversarial networks.arXiv preprint arXiv:1406.2661, 1–9.
    [Google Scholar]
  13. Hermanrud, K., Antonsen, F., Teixeira De Oliveira, M.E., Petersen, S.A. and Constable, M.
    [2019]. Future Geosteering and Well Placement Solutions from an Operator Perspective. In: SPE Annual Caspian Technical Conference. SPE, 1–12.
    [Google Scholar]
  14. Larsen, D.S., Boesing, D., Hartmann, A., Martakov, S., Mumtaz, A. Skillings, J. and Vianna, A.
    [2018]. Boundary detection ahead of the bit - A sensitivity study of extra deep azimuthal resistivity. In Offshore Technology Conference Asia. OTC: 1–11.
    [Google Scholar]
  15. Li, H. and Misra, S.
    [2017]. Prediction of subsurface NMR T2 distributions in a shale petroleum system using variational autoencoder-based neural networks.IEEE Geoscience and Remote Sensing Letters, 14(12), 2395–2397.
    [Google Scholar]
  16. Manouchehri, S., Pham, N., Hellem, T. A. and Roden, R.
    [2020]. A multi-disciplinary approach to establish a workflow for the application of machine learning for detailed reservoir description–Wisting case study.First Break, 38(7), 81–88.
    [Google Scholar]
  17. Panjaitan, H.
    [2006]. Sand-body dimensions in outcrop and subsurface, lower Williams Fork Formation, Piceance Basin, Colorado.MS thesis, Colorado School of Mines, Golden, Colorado. 1–170.
    [Google Scholar]
  18. Pranter, M.J., Cole, R.D., Panjaitan, H. and Sommer, N.K.
    [2009]. Sandstone-body dimensions in a lower coastal-plain depositional setting: lower Williams Fork Formation, Coal Canyon, Piceance Basin, Colorado.AAPG bulletin, 93(10), 1379–1401.
    [Google Scholar]
  19. Pranter, M.J., and Sommer, N.K.
    [2011]. Static connectivity of fluvial sandstones in a lower coastal-plain setting: An example from the Upper Cretaceous lower Williams Fork Formation, Piceance Basin, Colorado.AAPG bulletin, 95(6), 899–923.
    [Google Scholar]
  20. Pranter, M.J., Hewlett, A.C., Cole, R.D., Wang, H. and Gilman, J.
    [2014]. Fluvial architecture and connectivity of the Williams Fork Formation: use of outcrop analogues for stratigraphic characterization and reservoir modelling.Geological Society, London, Special Publications, 387(1), 57–83.
    [Google Scholar]
  21. Radford, A., Metz, L. and Chintala, S.
    [2016]. Unsupervised representation learning with deep convolutional generative adversarial networks. In: 4th International Conference on Learning Representations. ICLR, 1–14.
    [Google Scholar]
  22. Trampush, S.M., Hajek, E.A., Straub, K.M. and Chamberlin, E.P.
    [2017]. Identifying autogenic sedimentation in fluvial-deltaic stratigraphy: Evaluating the effect of outcrop-quality data on the compensation statistic.Journal of Geophysical Research: Earth Surface, 122(1), 91–113.
    [Google Scholar]
http://instance.metastore.ingenta.com/content/journals/10.3997/1365-2397.fb2021051
Loading
/content/journals/10.3997/1365-2397.fb2021051
Loading

Data & Media loading...

  • Article Type: Research Article
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