1887

Abstract

Summary

This work presents a novel method for bidirectional rock physics modeling and inversion through the use of coupled deep learning networks. Due to their frequent reliance on simplifying assumptions, traditional techniques typically encounter challenges including non-uniqueness and computational complexity. In order to guarantee consistency between forward and inverse mappings, we suggested a system that consists of two coupled neural networks through a shared latent space and trained simultaneously. In addition, our architecture leverages a hybrid loss function to balance forward, inverse, and cycle-consistency terms and rock physics related constrains.

Our method outperforms conventional methods such as stochastic methods in predicting porosity, shale volume, and calcite volume, and it achieves 96% accuracy in facies prediction in experimental results using carbonate reservoir data. Furthermore, the inclusion of a weight-sharing mechanism ensures parameter efficiency and promotes regularization, enhancing the robustness of the model. Lastly, this method offers a reliable, effective, and precise substitute for reservoir characterization.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.2025101292
2025-06-02
2026-02-15

  • From This Site

    /content/papers/10.3997/2214-4609.2025101292
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Ardizzone, L., Kruse, J., Wirkert, S., Rahner, D., Pellegrini, E. W., Klessen, R. S.,… and Köthe, U. [2018] Analyzing inverse problems with invertible neural networks. arXiv preprint arXiv:1808.04730.
     [Google Scholar]
  2. Bachrach, R. [2006] Joint estimation of porosity and saturation using stochastic rock-physics modelling. Geophysics71(5), O53–O63.
     [Google Scholar]
  3. Das, V. and Mukerji, T. [2020] Petrophysical properties prediction from prestack seismic data using convolutional neural networks. Geophysics85(5), N41–N55.
     [Google Scholar]
  4. Feng, G., Zeng, H.-H., Xu, X.-R., Tang, G.-Y. and Wang, Y.-X. [2023] Shear wave velocity prediction based on deep neural network and theoretical rock physics modelling. Frontiers in Earth Science10, 1025635.
     [Google Scholar]
  5. Grana, D. [2018] Joint facies and reservoir properties inversion. Geophysics83(3), M15–M24.
     [Google Scholar]
  6. Grana, D., Azevedo, L. and Liu, M. [2020] A comparison of deep machine learning and Monte Carlo methods for facies classification from seismic data. Geophysics85(4), WA41–WA52.
     [Google Scholar]
  7. Mavko, G., Mukerji, T. and Dvorkin, J. [2020] The rock physics handbook. Cambridge university press.
     [Google Scholar]
  8. Tarantola, A. [2005] Inverse problem theory and methods for model parameter estimation. SIAM.
     [Google Scholar]
  9. Zhu, J.-Y., Park, T., Isola, P. and Efros, A. A. [2017] Unpaired image-to-image translation using cycle-consistent adversarial networks. Proceedings of the IEEE international conference on computer vision, 2223–2232.
     [Google Scholar]
/content/papers/10.3997/2214-4609.2025101292
Loading
A Novel Approach for Forward and Inverse Rock Physics Modelling Using Coupled Deep Learning Networks
Conference Proceedings 2025, 1 (2025); https://doi.org/10.3997/2214-4609.2025101292
/content/papers/10.3997/2214-4609.2025101292
/content/papers/10.3997/2214-4609.2025101292
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