1887

Abstract

Summary

Geological CO2 storage (GCS) has been commonly recognized as an effective approach to reduce greenhouse gas emissions. The trapping mechanisms have been widely studied in the literature, where solubility and residual trapping are considered as safer for short-term entrapment. However, density-driven upward CO2 can hamper those mechanisms since the existence of free-phase CO2 would lead to the risk of CO2 leakage. One way to mitigate such risks is to adopt optimization strategies that attempt to adjust the well controls (e.g., injection rate) to maximize the immobilized CO2 (or minimize the free-phase CO2). However, model-based optimizations are computationally demanding, especially when time-consuming forward simulations such as those used for GCS are used. Surrogate models provide an attractive fit-for-purpose alternative for generating the required simulation responses at a reduced computational cost.

Recently, deep learning-based surrogate models have been proposed to speed up the optimization procedure. A major limitation of these models is their generalizability or extrapolation power, that is, their ability to predict beyond the training data, which is likely to happen during the optimization iterations. We propose an active learning strategy to address this issue by adapting the training data to the optimization iterations to improve the local accuracy of the model. Active learning is popular when unlabeled data is abundant but labeling (running simulation in our application) is expensive. One of the main advantages of active learning is that instead of frontloading the computation, it selectively and dynamically adds new data points to the training process, thereby adapting the prediction accuracy and distributing the computational budget efficiently. We apply active learning-based optimization with an artificial neural network (ANN) proxy model to maximize the immobilized CO2 during GCS. For local gradient-based optimization, active learning provides an efficient approach to adaptively sample new training data around the optimization path and update the ANN-based surrogate model to maintain its local accuracy.

Compared to the traditional off-line training approach, active learning results in improved model accuracy and computational efficiency. Active learning is a general framework that can be used in other subsurface flow applications to reduce computation and to improve the consistency between surrogate models and their corresponding full-scale simulation models.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202244051
2022-09-05
2026-03-09

  • From This Site

    /content/papers/10.3997/2214-4609.202244051
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. [1]B.Metz, O.Davidson, H. C.De Coninck, M.Loos and L.Meyer, IPCC special report on carbon dioxide capture and storage, Cambridge: Cambridge University Press, 2005.
     [Google Scholar]
  2. [2]H.Shamshiri and B.Jafarpour, “Controlled CO2 injection into heterogeneous geologic formations for improved solubility and residual trapping,” Water Resources Research, vol. 48, 2012.
     [Google Scholar]
  3. [3]F.Zheng, A.Jahandideh, B.Jha and B.Jafarpour, “Geologic CO2 storage optimization under geomechanical risk using coupled-physics models,” International Journal of Greenhouse Gas Control, vol. 110, p. 103385, 2021.
     [Google Scholar]
  4. [4]M.Tang, X.Ju and L.J.Durlofsky, “Deep-learning-based coupled flow-geomechanics surrogate model for CO2 sequestration,” International Journal of Greenhouse Gas Control, vol. 118, p. 103692, 2022.
     [Google Scholar]
  5. [5]S.Razavi, B. A.Tolson and D.H.Burn, “Review of surrogate modeling in water resources,” Water Resources Research, vol. 48, 2012.
     [Google Scholar]
  6. [6]G. G.Wang and S.Shan, “Review of metamodeling techniques in support of engineering design optimization,” 2007.
     [Google Scholar]
  7. [7]B.Settles, “Active learning literature survey,” 2009.
     [Google Scholar]
  8. [8]S.Katoch, S. S.Chauhan and V.Kumar, “A review on genetic algorithm: past, present, and future,” Multimedia Tools and Applications, vol. 80, pp. 8091–8126, 2021.
     [Google Scholar]
  9. [9]P. J. M.Van Laarhoven and E. H.L.Aarts, “Simulated annealing,” in Simulated annealing: Theory and applications, Springer, 1987, pp. 7–15.
     [Google Scholar]
  10. [10]F.Forouzanfar, E.Della Rossa, R.Russo and A.C.Reynolds, “Life-cycle production optimization of an oil field with an adjoint-based gradient approach,” Journal of Petroleum Science and Engineering, vol. 112, pp. 351–358, 2013.
     [Google Scholar]
  11. [11]J. F. B.M.Kraaijevanger, P. J.P.Egberts, J. R.Valstar and H.W.Buurman, “Optimal waterflood design using the adjoint method,” in SPE Reservoir Simulation Symposium, 2007.
     [Google Scholar]
  12. [12]Y.Chen, D. S.Oliver and D.Zhang, “Efficient ensemble-based closed-loop production optimization,” Spe Journal, vol. 14, pp. 634–645, 2009.
     [Google Scholar]
  13. [13]S. T.Do and A.C.Reynolds, “Theoretical connections between optimization algorithms based on an approximate gradient,” Computational Geosciences, vol. 17, pp. 959–973, 2013.
     [Google Scholar]
  14. [14]R.M.Fonseca, S.S.Kahrobaei, L. J. T.Van Gastel, O.Leeuwenburgh and J.D.Jansen, “Quantification of the impact of ensemble size on the quality of an ensemble gradient using principles of hypothesis testing,” in SPE Reservoir Simulation Symposium, 2015.
     [Google Scholar]
  15. [15]B.Yegnanarayana, Artificial neural networks, PHI Learning Pvt. Ltd., 2009.
     [Google Scholar]
  16. [16]Z.Liu and A.C.Reynolds, “Gradient-enhanced support vector regression for robust life-cycle production optimization with nonlinear-state constraints,” SPE Journal, vol. 26, pp. 1590–1613, 2021.
     [Google Scholar]
  17. [17]B.Shahriari, K.Swersky, Z.Wang, R. P.Adams and N.De Freitas, “Taking the human out of the loop: A review of Bayesian optimization,” Proceedings of the IEEE, vol. 104, pp. 148–175, 2015.
     [Google Scholar]
  18. [18]S.Kitayama, M.Arakawa and K.Yamazaki, “Sequential approximate optimization using radial basis function network for engineering optimization,” Optimization and Engineering, vol. 12, pp. 535–557, 2011.
     [Google Scholar]
  19. [19]A.R.Conn, N. I. M.Gould and P.L.Toint, Trust region methods, SIAM, 2000.
     [Google Scholar]
  20. [20]J.Yu and B.Jafarpour, “Active Learning for Well Control Optimization with Surrogate Models,” SPE Journal, pp. 1–21, 2022.
     [Google Scholar]
  21. [21]M.Asadollahi and G.Naevdal, “Waterflooding optimization using gradient based methods,” in SPE/EAGE Reservoir Characterization & Simulation Conference, 2009.
     [Google Scholar]
/content/papers/10.3997/2214-4609.202244051
Loading
Active Learning for Efficient Optimization of Geological CO2 Storage with Surrogate Models
Conference Proceedings 2022, 1 (2022); https://doi.org/10.3997/2214-4609.202244051
/content/papers/10.3997/2214-4609.202244051
/content/papers/10.3997/2214-4609.202244051
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