1887

Abstract

Summary

Predicting the spatial distribution of facies and collocated acoustic impedance (IP) in the subsurface from fullstack seismic data is fundamental for assessing mineral and energy natural resources potential. In recent years, deep generative models (DGMs) such as variational autoencoders (VAEs) and generative adversarial networks (GANs) were proposed as powerful methods to reproduce complex facies patterns, honoring prior geological data. Variational Bayesian inference using inverse autoregressive flows (IAF) can be performed to infer the solution to a geophysical inversion problem from the encoded latent space of such pre-trained DGMs. The method proposed was successfully tested on synthetic case studies of ground penetrating radar data inversion, although not accounting for the spatial uncertainty affecting the facies-dependent property, from which the geophysical data is calculated. The influence of such uncertainty can significantly influence the accuracy inversion. In this work we propose specific VAE and GAN architectures to simultaneously predict facies and IP while accounting for both their spatial uncertainties. We then test the inversion with IAF on seismic inversion problems, demonstrating the methods capacity in reproducing the statistics of the training images and to solve the seismic inversion problem. We also evaluate advantages and limitations of both networks by comparing the results obtained.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202335077
2023-11-27
2026-04-10

  • From This Site

    /content/papers/10.3997/2214-4609.202335077
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Blei, D. M., Kucukelbir, A., and McAuliffe, J. D. [2017]. Variational Inference: A Review for Statisticians.Journal of the American Statistical Association, 112(518), 859–877. https://doi.org/10.1080/01621459.2017.1285773
     [Google Scholar]
  2. Chan, S., and Elsheikh, A. H. [2019]. Parametric generation of conditional geological realizations using generative neural networks.Computational Geosciences, 23(5), 925–952. https://doi.org/10.1007/s10596-019-09850-7
     [Google Scholar]
  3. De Figueiredo, L. P., Grana, D., Roisenberg, M., and Rodrigues, B. B. [2019]. Gaussian mixture Markov chain Monte Carlo method for linear seismic inversion.GEOPHYSICS, 84(3), R463–R476. https://doi.org/10.1190/geo2018-0529.1
     [Google Scholar]
  4. Dupont, E., Zhang, T., Tilke, P., Liang, L., and Bailey, W. [2018, July 5]. Generating Realistic Geology Conditioned on Physical Measurements with Generative Adversarial Networks. Retrieved from http://arxiv.org/abs/1802.03065
     [Google Scholar]
  5. González, E. F., Mukerji, T., and Mavko, G. [2008]. Seismic inversion combining rock physics and multiple-point geostatistics.Geophysics, 73(1), R11–R21. https://doi.org/10.1190/1.2803748
     [Google Scholar]
  6. 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:1406.2661 [Cs, Stat]. Retrieved from http://arxiv.org/abs/1406.2661
     [Google Scholar]
  7. Grana, D., Mukerji, T., and Doyen, P. [2021]. Seismic reservoir modeling: theory, examples, and algorithms.Hoboken, NJ Chichester: Wiley Blackwell.
     [Google Scholar]
  8. Hoffman, M., Sountsov, P., Dillon, J. V., Langmore, I., Tran, D., and Vasudevan, S. [2019, March 8]. NeuTra-lizing Bad Geometry in Hamiltonian Monte Carlo Using Neural Transport. Retrieved from http://arxiv.org/abs/1903.03704
     [Google Scholar]
  9. Kingma, D. P., and Welling, M. [2013]. Auto-Encoding Variational Bayes.https://doi.org/10.48550/ARXIV.1312.6114
     [Google Scholar]
  10. Levy, S., Laloy, E., and Linde, N. [2022]. Variational Bayesian inference with complex geostatistical priors using inverse autoregressive flows.Computers & Geosciences, 105263. https://doi.org/10.1016/j.cageo.2022.105263
     [Google Scholar]
  11. Linde, N., Renard, P., Mukerji, T., and Caers, J. [2015]. Geological realism in hydrogeological and geophysical inverse modeling: A review.Advances in Water Resources, 86, 86–101. https://doi.org/10.1016/j.advwatres.2015.09.019
     [Google Scholar]
  12. Lopez-Alvis, J., Laloy, E., Nguyen, F., and Hermans, T. [2021]. Deep generative models in inversion: The impact of the generator’s nonlinearity and development of a new approach based on a variational autoencoder.Computers & Geosciences, 152, 104762. https://doi.org/10.1016/j.cageo.2021.104762
     [Google Scholar]
  13. Nunes, R., Soares, A., Azevedo, L., and Pereira, P. [2017]. Geostatistical Seismic Inversion with Direct Sequential Simulation and Co-simulation with Multi-local Distribution Functions.Mathematical Geosciences, 49(5), 583–601. https://doi.org/10.1007/s11004-016-9651-0
     [Google Scholar]
  14. Tarantola, A. [2005]. Inverse problem theory and methods for model parameter estimation. Philadelphia, PA: Society for Industrial and Applied Mathematics.
     [Google Scholar]
/content/papers/10.3997/2214-4609.202335077
Loading
Deep Generative Networks for Multivariate Fullstack Seismic Inversion with Using Inverse Autoregressive Flows
Conference Proceedings 2023, 1 (2023); https://doi.org/10.3997/2214-4609.202335077
/content/papers/10.3997/2214-4609.202335077
/content/papers/10.3997/2214-4609.202335077
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