Abstract
Geological facies modeling is a key component in exploration and characterization of subsurface reservoirs. While traditional geostatistical approaches are still commonly used nowadays, deep learning is gaining a lot of attention within geoscientific community for generating subsurface models, as a result of recent advance of computing powers and increasing availability of training data sets. This work presents a deep learning approach for geological facies modeling based on generative adversarial networks (GANs) combined with training-image-based simulation. In a typical application of learned networks, all neural network parameters are fixed after training, and the uncertainty in the trained model cannot be analyzed. To address this problem, a Bayesian GANs (BGANs) approach is proposed to create facies models. In this approach, a probability distribution is assigned to the neural parameters of the BGANs. Only neural parameters of the generator in BGANs are assigned with a probability function, and the ones in the discriminator are treated as fixed. Random samples are then drawn from the posterior distribution of neural parameters to simulate subsurface facies models. The proposed approach is applied to the two different geological depositional scenarios, fluvial channels and carbonate mounds, and the generated models reasonably capture the variability of the training/testing data. Meanwhile, the model uncertainty of learned networks is readily accessible. To fully sample the spatial distribution in the training image set, a large collection of samples of network parameters is required to be drawn from the posterior distribution, thus significantly increasing computational cost.
Similar content being viewed by others
References
Azevedo L, Paneiro G, Santos A, Soares A (2020) Generative adversarial network as a stochastic subsurface model reconstruction. Comput Geosci 24:1673–1692
Bisong E (2019) Building machine learning and deep learning models on Google cloud platform. Apress, New York
Blei DM, Kucukelbir A, McAuliffe JD (2017) Variational inference: a review for statisticians. J Am Stat Assoc 112(518):859–877
Buland A, Omre H (2003) Bayesian linearized AVO inversion. Geophysics 68(1):185–198
Caers J (2018) Bayesianism in the Geosciences. In: Daya Sagar B, Cheng Q, Agterberg F (eds) Handbook of mathematical geosciences. Springer, Cham
Chan S, Elsheikh AH (2019) Parametrization of stochastic inputs using generative adversarial networks with application in geology. arXiv preprint arXiv:1904.03677
Chen T, Fox EB, Guestrin C (2014) Stochastic gradient Hamiltonian Monte Carlo. arXiv preprint arXiv:1402.4102
Chilès JP, Delfiner P (1999) Geostatistics, modeling spatial uncertainty. Wiley press, New York
Cojan I, Fouché O, Lopéz S, Rivoirard J (2005) Process-based reservoir modelling in the example of meandering channel. In Leuangthong O, Deutsch CV (eds) Geostatistics Banff 2004. https://doi.org/10.1007/978-1-4020-3610-1_62
Conjard M, Grana D (2021) Ensemble-based seismic and production data assimilation using selection Kalman model. Math Geosci. https://doi.org/10.1007/s11004-021-09940-2
Cox M, Cox T (2008) Multidimensional scaling. In: Handbook of data visualization. Springer handbooks Comp. statistics. Springer, Berlin. https://doi.org/10.1007/978-3-540-33037-0_14
Cui Z, Zhang M, Cao Z, Cao C (2019) Image data augmentation for SAR sensor via generative adversarial nets. IEEE Access 7:42255–42268
Dana S, Srinivasan S, Karra S et al (2020) Towards real-time forecasting of natural gas production by harnessing graph theory for stochastic discrete fracture networks. J Petrol Sci Eng. https://doi.org/10.1016/j.petrol.2020.107791
Dupont E, Zhang T, Tilke P, Liang L, Bailey W (2018) Generating realistic geology conditioned on physical measurements with generative adversarial networks. arXiv preprint arXiv:1802.03065
Feng R, Balling N, Grana D, Dramsch JS, Hansen TM (2021a) Bayesian convolutional neural networks for seismic facies classification. IEEE Trans Geosci Remote Sens. https://doi.org/10.1109/TGRS.2020.3049012
Feng R, Grana D, Balling N (2021b) Variational inference in Bayesian neural network for well-log prediction. Geophysics 86(3):M91–M99
Gal Y, Hron J, Kendall, A (2017) Concrete dropout: advances in neural information processing systems, pp 3581–3590
Ghahramani Z (2015) Probabilistic machine learning and artificial intelligence. Nature 521:452–459
Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial networks. arXiv preprint arXiv:1406.2661
Grana D (2016) Bayesian linearized rock-physics inversion. Geophysics 81(6):D625–D641
Grana D, Mukerji T, Doyen D (2021) Seismic reservoir modeling: theory, examples, and algorithms. Wiley, New York
Hansen TM (2020) Efficient probabilistic inversion using the rejection sampler—exemplified on airborne EM data. Geophys J Int 224:543–557
Hansen TM, Vu LT, Mosegaard K, Cordua KS (2018) Multiple point statistical simulation using uncertain (soft) conditional data. Comput Geosci 114:1–10
Kendall A, Badrinarayanan V, Cipolla (2015) Bayesian SegNet: model uncertainty in deep convolutional encoder-decoder architectures for sense understanding. arXiv preprint arXiv:1511.02680
Lakshminarayanan B, Pritzel, A, Blundell C (2017) Simple and scalable predictive uncertainty estimation using deep ensembles: advances in neural information processing systems, pp 6402–6413
Laloy E, Hérault R, Jacques D, Linde N (2018) Training-image based geostatistical inversion using a spatial generative adversarial neural network. Water Resour Res 54:381–406
Laloy E, Linde N, Ruffino C, Hérault R, Gasso G, Jacques D (2019) Gradient-based deterministic inversion of geophysical data with generative adversarial networks: is it feasible? Comput Geosci. https://doi.org/10.1016/j.cageo.2019.104333
Liu Y, Sun W, Durlofsky LJ (2019) A deep-learning-based geological parameterization for history matching complex models. Math Geosci 51:725–766. https://doi.org/10.1007/s11004-019-09794-9
Loe MK, Grana D, Tjelmeland H (2021) Geophysics-based fluid-facies predictions using ensemble updating of binary state vectors. Math Geosci 53:325–347. https://doi.org/10.1007/s11004-021-09922-4
Mariethoz G, Caers J (2014) Multiple-point geostatistics: stochastic modeling with training images. Wiley press, New York
Michael HA, Li H, Boucher A, Sun T, Caers J, Gorelick SM (2010) Combining geologic-process models and geostatistics for conditional simulation of 3-D subsurface heterogeneity. Water Resour Res. https://doi.org/10.1029/2009WR008414
Mosegaard K, Tarantola A (1995) Monte Carlo sampling of solutions to inverse problems. J Geophys Res Solid Earth 100(B7):12431–12447. https://doi.org/10.1029/94JB03097
Mosegaard K, Sambridge M (2002) Monte Carlo analysis of inverse problems. Inverse Prob 18:R29–R54. https://doi.org/10.1088/0266-5611/18/3/201
Mosser L, Dubrule O, Blunt MJ (2017) Reconstruction of three-dimensional porous media using generative adversarial neural networks. Phys Rev E. https://doi.org/10.1103/PhysRevE.96.043309
Mosser L, Dubrule O, Blunt MJ (2019) Deepflow: history matching in the space of deep generative models. arXiv preprint arXiv:1905.05749
Mosser L, Dubrule O, Blunt MJ (2020) Stochastic seismic waveform inversion using generative adversarial networks as a geological prior. Math Geosci 52:53–79. https://doi.org/10.1007/s11004-019-09832-6
Nesvold E, Mukerji T (2019) Geomodeling using generative adversarial networks and a database of satel-lite imagery of modern river deltas. In: Petroleum geostatistics. European Association of Geoscientists and Engineers, vol 2019, No. 1, pp 1–5
Pyrcz MJ, Deutsch CV (2014) Geostatistical reservoir modeling. Oxford University Press, Oxford
Radford A, Metz L, Chintala S (2015) Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434
Rongier G, Rude CM, Herring T, Pankratius V (2020) An attempt at improving atmospheric corrections in InSAR using cycle-consistent adversarial networks. Earth. https://doi.org/10.31223/X5M594
Saatchi Y, Wilson AG (2017) Bayesian GAN. arXiv preprint arXiv:1705.09558
Sambridge M, Mosegaard K (2002) Monte Carlo methods in geophysical inverse problems. Rev Geophys. https://doi.org/10.1029/2000RG000089
Song S, Mukerji T, Hou J (2021a) GANSim: conditional facies simulation using an improved progressive growing of generative adversarial networks (GANs). Math Geosci. https://doi.org/10.1007/s11004-021-09934-0
Song S, Mukerji T, Hou J (2021b), Bridging the gap between geophysics and geology with generative adversarial networks. IEEE Trans Geosci Remote Sensing. https://doi.org/10.1109/TGRS.2021.3066975
Sra S, Nowozin S, Wright SJ (2011) Optimization for machine learning. The MIT Press, Cambridge
Strebelle S (2002) Conditional simulation of complex geological structures using multiple-point statistics. Math Geol 34:1–21. https://doi.org/10.1023/A:1014009426274
Strebelle S (2021) Multiple-point statistics simulation models: pretty pictures or decision-making tools? Math Geosci 53:267–278. https://doi.org/10.1007/s11004-020-09908-8
Sun AY (2018) Discovering state-parameter mappings in subsurface models using generative adversarial networks. Geophys Res Lett 45:11137–11146
Tahmasebi P, Kamrava S, Bai T, Sahimi M (2020) Machine learning in geo- and environmental sciences: from small to large scale. Adv Water Resour. https://doi.org/10.1016/j.advwatres.2020.103619
Tarantola A (2005) Inverse problem theory and methods for model parameter estimation. SIAM, Philadelphia
Turner R, Hung J, Frank E, Saatci Y, Yosinski J (2019) Metropolis-Hastings generative adversarial networks. arXiv preprint arXiv:1811.11357
Ushijima-Mwesigwa H, Hyman J, Hagberg A et al (2021) Multilevel graph partitioning for three-dimensional discrete fracture networks flow simulations. Math Geosci. https://doi.org/10.1007/s11004-021-09944-y
Wang Z, Bovik AC, Sheikh HR, Simoncelli EP (2004) Image quality assessment: from error visibility to structural similarity. IEEE Trains Image Process 13(4):600–612
You H, Cheng Y, Cheng T, Li C, Zhou P (2018) Bayesian cycle-consistent generative adversarial networks via marginalizing latent sampling. arXiv preprint arXiv:1811.07465
Zhang C, Song X, Azevedo L (2021) U-net generative adversarial network for subsurface facies modeling. Comput Geosci 25:553–573
Zhang TF, Tilke P, Dupont E, Zhu LC, Liang L, Bailey W (2019) Generating geologically realistic 3D reservoir facies models using deep learning of sedimentary architecture with generative adversarial networks. Pet Sci 16:541–549
Zhong Z, Sun AY, Ren B, Wang Y (2021) A deep-learning-based approach for reservoir production forecast under uncertainty. SPE J 26(3):1314–1340
Acknowledgements
This research is sponsored by the LOCRETA project. We also acknowledge the sponsors of SCERF.
Author information
Authors and Affiliations
Corresponding author
Appendix: Quality Evaluation Metrics
Appendix: Quality Evaluation Metrics
To assess the variability and similarity between the generated and testing data, the technique of multidimensional scaling (MDS) is used to reduce the high-dimensional problem to a low-dimensional data space while preserving similarity measure for comparison (Cox and Cox 2008). The MDS-transformed data points are unitless and can be plotted in the common Cartesian space. The main steps in the classical MDS algorithm are summarized as follows: (i) setup of the squared proximity matrix with Euclidean distance and double centering; (ii) determination of the first largest eigenvalues and corresponding eigenvectors of the double-centered matrix; (iii) calculation of the object coordinates in the new space (Cox and Cox 2008). The variability and similarity can be evaluated in the reduced dimension for the two model sets. For example, the projected data points can be sparsely distributed across the Cartesian space when diverse patterns exist in the original data or distributed in small clusters for multimodal distributions of patterns. The similarity of the data sets can be quantified by the range of these two clusters, and it should become indistinguishable when the two data sets are similar (Zhang et al. 2021).
For a quantitative measure of the variability within the generated models, the perception-based structural similarity index (SSIM) is introduced, on the basis of the definition given by Wang et al. (2004) and Sun (2018)
where \(y\) and \({y}^{*}\) represent two random images from the generated models, respectively; \(\mu \) is the mean and \({\sigma }^{2}\) is the variance; \({c}_{1}\) and \({c}_{2}\) are constant values to stabilize the denominator, and are taking the values of \({0.01}^{2}\) and \({0.03}^{2}\), respectively, for the gray images. The value range of SSIM is between −1 and 1, where 1 means that the two images are identical, and −1 is reached when the two images are completely different (Sun 2018). In image recognition, SSIM quantifies image degradation as structural information change where pixels that are spatially close with each other should have strong inter-dependencies. The perceptual phenomena, including the masking terms of luminance and contrast, are also accounted for in the SSIM to ensure that image distortions are detected as well (Wang et al. 2004). In general, compared to the mean squared error, SSIM is more indicative of perceived dissimilarity/similarity in the models.
Rights and permissions
About this article
Cite this article
Feng, R., Grana, D., Mukerji, T. et al. Application of Bayesian Generative Adversarial Networks to Geological Facies Modeling. Math Geosci 54, 831–855 (2022). https://doi.org/10.1007/s11004-022-09994-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11004-022-09994-w