1887

Abstract

Summary

Multiphase flow is predicted based on repeated reservoir simulations to account for the geological uncertainties of the subsurface. We propose a deep-learning-based framework that maps geological parameters to dynamic outputs. Unlike other studies, the aim here is not to reach a general reservoir model but to train the models specifically for the reservoir and task at hand. This means using a smaller architecture, less data, and shorter time while the model learns only the necessary relationships to speed up a reservoir engineer’s work. This approach requires training a new model every time, which takes longer man using a single, general model. However, the advantage is mat the tailored model can be applied to any reservoir and task, from petroleum production to CO2 storage. The proposed method was tested on a carbon capture and sequestration project in Canada, resulting in a time reduction of 54% while maintaining an R2 score of 0.97 on the validation set. Additionally, a web application was developed as a user-friendly interface to allow reservoir engineers to utilise the framework.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202639098
2026-03-09
2026-02-08

  • From This Site

    /content/papers/10.3997/2214-4609.202639098
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Agarap, A.F. [2019] Deep Learning using Rectified Linear Units (ReLU). arXiv preprint arXiv:1803.08375.
     [Google Scholar]
  2. Aziz, K. and Settari, A. [1979] Petroleum Reservoir Simulation. Society of Petroleum Engineers, Richardson, TX.
     [Google Scholar]
  3. Chen, Z., Huan, G. and Ma, Y. [2006] Computational Methods for Multiphase Flows in Porous Media. SIAM, Philadelphia, PA.
     [Google Scholar]
  4. Doughty, C. [2009] Investigation of C02 Plume Behavior for a Large-Scale Pilot Test of Geologic Carbon Storage in a Saline Formation. Transport in Porous Media, 82, 49–76.
     [Google Scholar]
  5. Equinor [2023] webviz-config [software]. Available at: https://github.com/equinor/webviz-config.
     [Google Scholar]
  6. Harvey, S., Hopkins, J., Kuehl, H., O’Brien, S. and Mateeva, A. [2022] Quest CCS facility: Time-lapse seismic campaigns. International Journal of Greenhouse Gas Control, 117, 103665.
     [Google Scholar]
  7. He, K., Zhang, X., Ren, S. and Sun, J. [2016] Deep Residual Learning for Image Recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 770–778.
     [Google Scholar]
  8. Ioffe, S. and Szegedy, C. [2015] Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. In: Proceedings of the 32nd International Conference on Machine Learning. 448–456. ICML 2015.
     [Google Scholar]
  9. Jin, Z.L., Liu, Y. and Durlofsky, L.J. [2020] Deep-learning-based surrogate model for reservoir simulation with time-varying well controls. Journal of Petroleum Science and Engineering, 192, 107273.
     [Google Scholar]
  10. Kingma, D.P and Ba, J. [2015] Adam: A Method for Stochastic Optimization. In: 3rd International Conference on Learning Representations (ICLR), Conference Track Proceedings. San Diego, CA.
     [Google Scholar]
  11. Krizhevsky, A., Sutskever, I. and Hinton, G.E. [2012] ImageNet Classification with Deep Convolutional Neural Networks. In: Advances in Neural Information Processing Systems, 25. Curran Associates, Inc.
     [Google Scholar]
  12. Kula, E.F. [2025] WebAI: Lightweight web interface for physics-AI reservoir surrogates [software]. GitHub repository: https://github.com/edizferit/WebAI.
     [Google Scholar]
  13. Li, Z., Kovachki, N., Azizzadenesheli, K., Liu, B., Bhattacharya, K., Stuart, A. and Anandkumar, A. [2021] Fourier Neural Operator for Parametric Partial Differential Equations. arXiv:2010.08895. Preprint.
     [Google Scholar]
  14. Mo, S., Zhu, Y., Zabaras, N., Shi, X. and Wu, J. [2019] Deep Convolutional Encoder-Decoder Networks for Uncertainty Quantification of Dynamic Multiphase Flow in Heterogeneous Media. Water Resources Research, 55(1), 703–728.
     [Google Scholar]
  15. Noh, H., Hong, S. and Han, B. [2015] Learning Deconvolution Network for Semantic Segmentation. In: 2015 IEEE International Conference on Computer Vision (ICCV). IEEE, Santiago, Chile, 1520–1528.
     [Google Scholar]
  16. Plotly [2023] Dash: A Web Application Framework for Python [software]. Available at: https://dash.plotly.com.
     [Google Scholar]
  17. Shell Canada Ltd. [2012] Quest Carbon Capture and Storage Project: Annual Report. Tech. rep., Shell Canada Ltd., Calgary, AB. Available at Alberta Open Government: https://open.alberta.ca/publications/quest-carbon-capture-and-storage-project-report-2012.
     [Google Scholar]
  18. Wen, G., Hay, C. and Benson, S.M. [2021] CCSNet: A deep learning modeling suite for CO2 storage. Advances in Water Resources, 155, 104009.
     [Google Scholar]
/content/papers/10.3997/2214-4609.202639098
Loading
Hybrid Simulation-AI Surrogate Workflow for Rapid Uncertainty Quantification of Multiphase Flow
Conference Proceedings 2026, 1 (2026); https://doi.org/10.3997/2214-4609.202639098
/content/papers/10.3997/2214-4609.202639098
/content/papers/10.3997/2214-4609.202639098
Loading

Data & Media loading...

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