Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
153 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

History Matching for Geological Carbon Storage using Data-Space Inversion with Spatio-Temporal Data Parameterization (2310.03228v1)

Published 5 Oct 2023 in cs.LG

Abstract: History matching based on monitoring data will enable uncertainty reduction, and thus improved aquifer management, in industrial-scale carbon storage operations. In traditional model-based data assimilation, geomodel parameters are modified to force agreement between flow simulation results and observations. In data-space inversion (DSI), history-matched quantities of interest, e.g., posterior pressure and saturation fields conditioned to observations, are inferred directly, without constructing posterior geomodels. This is accomplished efficiently using a set of O(1000) prior simulation results, data parameterization, and posterior sampling within a Bayesian setting. In this study, we develop and implement (in DSI) a deep-learning-based parameterization to represent spatio-temporal pressure and CO2 saturation fields at a set of time steps. The new parameterization uses an adversarial autoencoder (AAE) for dimension reduction and a convolutional long short-term memory (convLSTM) network to represent the spatial distribution and temporal evolution of the pressure and saturation fields. This parameterization is used with an ensemble smoother with multiple data assimilation (ESMDA) in the DSI framework to enable posterior predictions. A realistic 3D system characterized by prior geological realizations drawn from a range of geological scenarios is considered. A local grid refinement procedure is introduced to estimate the error covariance term that appears in the history matching formulation. Extensive history matching results are presented for various quantities, for multiple synthetic true models. Substantial uncertainty reduction in posterior pressure and saturation fields is achieved in all cases. The framework is applied to efficiently provide posterior predictions for a range of error covariance specifications. Such an assessment would be expensive using a model-based approach.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (44)
  1. Use of above-zone pressure data to locate and quantify leaks during carbon storage operations, International Journal of Greenhouse Gas Control 52 (2016) 32–43.
  2. Bayesian well-test 2D tomography inversion for CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT plume detection, International Journal of Greenhouse Gas Control 94 (2020) 102804.
  3. Estimation of plume distribution for carbon sequestration using parameter estimation with limited monitoring data, Water Resources Research 49 (2013) 4442–4464.
  4. Model-reduced adjoint-based inversion using deep-learning: Example of geological carbon sequestration modeling, Water Resources Research 58 (2022) e2021WR031041.
  5. Reducing uncertainty in geologic CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT sequestration risk assessment by assimilating monitoring data, International Journal of Greenhouse Gas Control 94 (2020) 102926.
  6. Dynamic characterization of geologic CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT storage aquifers from monitoring data with ensemble Kalman filter, International Journal of Greenhouse Gas Control 81 (2019) 199–215.
  7. Inference of rock flow and mechanical properties from injection-induced microseismic events during geologic CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT storage, International Journal of Greenhouse Gas Control 105 (2021) 103206.
  8. Deep-learning-based coupled flow-geomechanics surrogate model for CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT sequestration, International Journal of Greenhouse Gas Control 118:103692 (2022).
  9. U-FNO—An enhanced Fourier neural operator-based deep-learning model for multiphase flow, Advances in Water Resources 163 (2022) 104180.
  10. A robust deep learning workflow to predict multiphase flow behavior during geological CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT sequestration injection and post-injection periods, Journal of Hydrology 607 (2022) 127542.
  11. Fourier-MIONet: Fourier-enhanced multiple-input neural operators for multiphase modeling of geological carbon sequestration, arXiv preprint arXiv:2303.04778 (2023).
  12. Towards a predictor for CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT plume migration using deep neural networks, International Journal of Greenhouse Gas Control 105 (2021) 103223.
  13. W. Sun, L. J. Durlofsky, A new data-space inversion procedure for efficient uncertainty quantification in subsurface flow problems, Mathematical Geosciences 49 (2017) 679–715.
  14. Production forecasting and uncertainty quantification for naturally fractured reservoirs using a new data-space inversion procedure, Computational Geosciences 21 (2017) 1443–1458.
  15. A data-space inversion procedure for well control optimization and closed-loop reservoir management, Computational Geosciences 24 (2020) 361–379.
  16. Data-space inversion with ensemble smoother, Computational Geosciences 24 (2020) 1179–1200.
  17. Data-space inversion for rapid physics-informed direct forecasting in unconventional reservoirs, in: SPE Reservoir Simulation Conference, SPE, 2023, p. D010S001R013.
  18. W. Sun, L. J. Durlofsky, Data-space approaches for uncertainty quantification of CO2subscriptCO2\text{CO}_{2}CO start_POSTSUBSCRIPT 2 end_POSTSUBSCRIPT plume location in geological carbon storage, Advances in Water Resources 123 (2019) 234–255.
  19. Prediction-focused subsurface modeling: investigating the need for accuracy in flow-based inverse modeling, Mathematical Geosciences 47 (2015) 173–191.
  20. A. Satija, J. Caers, Direct forecasting of subsurface flow response from non-linear dynamic data by linear least-squares in canonical functional principal component space, Advances in Water Resources 77 (2015) 69–81.
  21. Direct forecasting of reservoir performance using production data without history matching, Computational Geosciences 21 (2017) 315–333.
  22. Direct prediction of spatially and temporally varying physical properties from time-lapse electrical resistance data, Water Resources Research 52 (2016) 7262–7283.
  23. A learning-based data-driven forecast approach for predicting future reservoir performance, Advances in Water Resources 118 (2018) 95–109.
  24. Training-image based geostatistical inversion using a spatial generative adversarial neural network, Water Resources Research 54 (2018) 381–406.
  25. Gradient-based deterministic inversion of geophysical data with generative adversarial networks: is it feasible?, Computers & Geosciences 133 (2019) 104333.
  26. Y. Liu, L. J. Durlofsky, 3D CNN-PCA: A deep-learning-based parameterization for complex geomodels, Computers & Geosciences 148 (2021) 104676.
  27. Characterization of the non-Gaussian hydraulic conductivity field via deep learning-based inversion of hydraulic-head and self-potential data, Journal of Hydrology 610 (2022) 127830.
  28. S. Jiang, L. J. Durlofsky, Data-space inversion using a recurrent autoencoder for time-series parameterization, Computational Geosciences 25 (2021) 411–432.
  29. Data-space inversion with a recurrent autoencoder for naturally fractured systems, Frontiers in Applied Mathematics and Statistics 7 (2021) 686754.
  30. Deep-learning-generalized data-space inversion and uncertainty quantification framework for accelerating geological CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT plume migration monitoring, Geoenergy Science and Engineering 224 (2023) 211627.
  31. Integrating deep learning-based data assimilation and hydrogeophysical data for improved monitoring of DNAPL source zones during remediation, Journal of Hydrology 601 (2021) 126655.
  32. Fusing stacked autoencoder and long short-term memory for regional multistep-ahead flood inundation forecasts, Journal of Hydrology 598 (2021) 126371.
  33. Implicit learning of convective organization explains precipitation stochasticity, Proceedings of the National Academy of Sciences 120 (2023) e2216158120.
  34. S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Computation 9 (1997) 1735–1780.
  35. Adversarial autoencoders, arXiv preprint arXiv:1511.05644 (2015).
  36. D. P. Kingma, J. Ba, Adam: a method for stochastic optimization, arXiv preprint arXiv:1412.6980 (2014).
  37. A. A. Emerick, A. C. Reynolds, Ensemble smoother with multiple data assimilation, Computers & Geosciences 55 (2013) 3–15.
  38. An integrated framework for optimal monitoring and history matching in CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT storage projects, Computational Geosciences (2023) 1–15.
  39. New trapping mechanism in carbon sequestration, Transport in Porous Media 82 (2010) 3–17.
  40. D. A. Cameron, L. J. Durlofsky, Optimization of well placement, CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT injection rates, and brine cycling for geological carbon sequestration, International Journal of Greenhouse Gas Control 10 (2012) 100–112.
  41. S. Jiang, L. J. Durlofsky, Treatment of model error in subsurface flow history matching using a data-space method, Journal of Hydrology 603 (2021) 127063.
  42. A. Y. Sun, J. P. Nicot, Inversion of pressure anomaly data for detecting leakage at geologic carbon sequestration sites, Advances in Water Resources 44 (2012) 20–29.
  43. Surrogate model for geological CO22{}_{2}start_FLOATSUBSCRIPT 2 end_FLOATSUBSCRIPT storage and its use in MCMC-based history matching, arXiv preprint arXiv:2308.06341 (2023).
  44. A. Jiang, B. Jafarpour, Deep convolutional autoencoders for robust flow model calibration under uncertainty in geologic continuity, Water Resources Research 57 (2021) e2021WR029754.
Citations (7)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now