DOI: https://doi.org/10.26089/NumMet.v25r213

Optimization of the training dataset for NDM-net (Numerical Dispersion Mitigation neural network)

Authors

  • Elena A. Gondyul
  • Vadim V. Lisitsa
  • Kirill G. Gadylshin
  • Dmitry G. Vishnevsky

Keywords:

numerical dispersion
seismic modelling
deep learning

Abstract

In this paper we present a new approach to building a training dataset for the NDMnet (Numerical Dispersion Mitigation neural network) neural network, which suppresses numerical dispersion in modeling seismic wave fields. The NDM-net is trained to display the solution of the system of equations of the dynamic theory of elasticity, calculated on a coarse grid, into a solution modeled on a fine grid. However, in order to train an NDM-net, it is necessary to pre-calculate seismograms on a fine grid, which is a time-consuming procedure. To reduce the computational costs of the algorithm, an original approach is proposed that allows to reduce the learning time without loss of accuracy. It is proposed to consider a linear combination of three different parameters: the distance between sources, the similarity of seismograms and the similarity of velocity models as an effective metric for generating a training dataset. The weights of the linear combination are determined using a global sensitivity analysis.

Downloads

Published

2024-04-28

Issue

Vol. 25 (2024): Issue 2.

Section

Methods and algorithms of computational mathematics and their applications

Author Biographies

Elena A. Gondyul

Trofimuk Institute of Petroleum Geology and Geophysics SB RAS
• Engineer

Vadim V. Lisitsa

Trofimuk Institute of Petroleum Geology and Geophysics SB RAS
• Head of Laboratory

Kirill G. Gadylshin

Trofimuk Institute of Petroleum Geology and Geophysics SB RAS
• Senior Researcher

Dmitry G. Vishnevsky

Trofimuk Institute of Petroleum Geology and Geophysics SB RAS
• Researcher

References

  1. E. H. Saenger, N. Gold, and S. A. Shapiro, “Modeling the Propagation of Elastic Waves Using a Modified Finite-Difference Grid,” Wave Motion 31 (1), 77-92 (2000)
    doi 10.1016/S0165-2125(99)00023-2
  2. V. Lisitsa and D. Vishnevskiy, “Lebedev Scheme for the Numerical Simulation of Wave Propagation in 3D Anisotropic Elasticity,” Geophys. Prospect. 58 (4), 619-635 (2010).
    doi 10.1111/j.1365-2478.2009.00862.x
  3. V. Lisitsa, V. Tcheverda, and D. Vishnevsky, “Numerical Simulation of Seismic Waves in Models with Anisotropic Formations: Coupling Virieux and Lebedev Finite-Difference Schemes,” Comput. Geosci. 16 (4), 1135-1152 (2012).
    doi 10.1007/s10596-012-9308-0
  4. J. M. Carcione and F. Cavallini, “A Rheological Model for Anelastic Anisotropic Media with Applications to Seismic Wave Propagation,” Geophys. J. Int. 119 (1), 338-348 (1994).
    doi 10.1111/j.1365-246X.1994.tb00931.x
  5. D. M. Vishnevsky, V. V. Lisitsa, and G. V. Reshetova, “Numerical Simulation of Seismic Wave Propagation in Media with Viscoelastic Inclusions,” Numerical Methods and Programming (Vychislitel’nye Metody i Programmirovanie) 14 (1), 155-165 (2013).
    https://num-meth.ru/index.php/journal/article/view/593 . Cited March 29, 2024.
  6. J. M. Carcione, C. Morency, and J. E. Santos, “Computational Poroelasticity -- A Review,” Geophysics. 75 (5), 75A229-75A243 (2010).
    doi 10.1190/1.3474602
  7. Y. J. Masson and S. R. Pride, “Finite-Difference Modeling of Biot’s Poroelastic Equations across all Frequencies,” Geophysics 75 (2), N33-N41 (2010).
    doi 10.1190/1.3332589
  8. E. Caspari, M. Novikov, V. Lisitsa, et al., “Attenuation Mechanisms in Fractured Fluid-Saturated Porous Rocks: A Numerical Modelling Study,” Geophys. Prospect. 67 (4), 935-955 (2019).
    doi 10.1111/1365-2478.12667
  9. M. Verliac and J. L. Calvez, “Microseismic Monitoring for Reliable CO_2 Injection and Storage -- Geophysical Modeling Challenges and Opportunities,” The Lead. Edge 40 (6), 418-423 (2021).
    doi 10.1190/tle40060418.1
  10. M. Prasad, S. Glubokovskikh, T. Daley, et al., “CO_2 Messes with Rock Physics,” The Lead. Edge 40 (6), 424-432 (2021).
    doi 10.1190/tle40060424.1
  11. J. Virieux, S. Operto, H. Ben-Hadj-Ali, et al., “Seismic Wave Modeling for Seismic Imaging,” The Lead. Edge 28 (5), 538-544 (2009).
    doi 10.1190/1.3124928
  12. J. Virieux, H. Calandra, and R. É. Plessix, “A Review of the Spectral, Pseudo-Spectral, Finite-Difference and Finite-Element Modelling Techniques for Geophysical Imaging,” Geophys. Prospect. 59 (5), 794-813 (2011).
    doi 10.1111/j.1365-2478.2011.00967.x
  13. J. Virieux, “P-SV Wave Propagation in Heterogeneous Media: Velocity-Stress Finite-Difference Method,” Geophysics 51 (4), 889-901 (1986).
    doi 10.1190/1.1442147
  14. P. Moczo, J. Kristek, and M. Gális, The Finite-Difference Modelling of Earthquake Motions: Waves and Ruptures (Cambridge Univ. Press, Cambridge, 2014).
    doi 10.1017/CBO9781139236911
  15. V. M. Sadovskii and O. V. Sadovskaya, “A Numerical Algorithm for the Analysis of Viscoelastic Waves in the Kelvin-Voigt Medium,” Numerical Methods and Programming (Vychislitel’nye Metody i Programmirovanie) 15 (1), 98-108 (2014).
    https://en.num-meth.ru/index.php/journal/article/view/753 . Cited March 29, 2024.
  16. J. Zhang and D. J. Verschuur, “Elastic Wave Propagation in Heterogeneous Anisotropic Media Using the Lumped Finite-Element Method,” Geophysics 67 (2), 625-638 (2002).
    doi 10.1190/1.1468624
  17. J. Tromp, D. Komatitsch, and Q. Liu, “Spectral-Element and Adjoint Methods in Seismology,” Commun. Comput. Phys. 3 (1), 1-32 (2008).
    https://authors.library.caltech.edu/records/stavm-ptm61 . Cited March 29, 2024.
  18. Y. Zhang, J. Fisser, and M. Gerritsma, “A Hybrid Mimetic Spectral Element Method for Three-Dimensional Linear Elasticity Problems,” J. Comput. Phys. 433, Article Identifier 110179 (2021).
    doi 10.1016/j.jcp.2021.110179
  19. M. Dumbser and M. Käser, “An Arbitrary High-Order Discontinuous Galerkin Method for Elastic Waves on Unstructured Meshes - II. The Three-Dimensional Isotropic Case,” Geophys. J. Int. 167 (1), 319-336 (2006).
    doi 10.1111/j.1365-246X.2006.03120.x
  20. J. Puente, M. Käser, M. Dumbser, and H. Igel, “An Arbitrary High-Order Discontinuous Galerkin Method for Elastic Waves on Unstructured Meshes -- IV. Anisotropy,” Geophys. J. Int. 169 (3), 1210-1228 (2007).
    doi 10.1111/j.1365-246X.2007.03381.x
  21. V. Lisitsa, V. Tcheverda, and C. Botter, “Combination of the Discontinuous Galerkin Method with Finite Differences for Simulation of Seismic Wave Propagation,” J. Comput. Phys. 311, 142-157 (2016).
    doi 10.1016/j.jcp.2016.02.005
  22. B. Fornberg, “The Pseudospectral Method: Accurate Representation of Interfaces in Elastic Wave Calculations,” Geophysics 53 (5), 625-637 (1988).
    doi 10.1190/1.1442497
  23. A. Pleshkevich, D. Vishnevsky, V. Lisitsa, and V. Levchenko, “Parallel Algorithm for One-Way Wave Equation Based Migration for Seismic Imaging,” in Communications in Computer and Information Science (Springer, Cham, 2019), Vol. 965, pp. 125-135 (2019).
    doi 10.1007/978-3-030-05807-4_11
  24. V. V. Lisitsa, “Dispersion Analysis of the Discontinuous Galerkin Method as Applied to the Equations of the Dynamic Elasticity Theory,” Numerical Methods and Programming (Vychislitel’nye Metody i Programmirovanie). 16 (3), 387-406 (2015).
    doi 10.26089/NumMet.v16r338
  25. A. R. Levander, “Fourth-Order Finite-Difference P-SV Seismograms,” Geophysics. 53 (11), 1425-1436 (1988).
    doi 10.1190/1.1442422
  26. O. O’Reilly, T. Lundquist, E. M. Dunham, and J. Nordström, “Energy Stable and High-Order-Accurate Finite Difference Methods on Staggered Grids,” J. Comput. Phys. 346, 572-589 (2017).
    doi 10.1016/j.jcp.2017.06.030
  27. V. Lisitsa, “Dispersion Analysis of Discontinuous Galerkin Method on Triangular Mesh for Elastic Wave Equation,” Appl. Math. Model. 40 (7-8), 5077-5095 (2016).
    doi 10.1016/j.apm.2015.12.039
  28. Y. Liu and M. K. Sen, “Acoustic VTI Modeling with a Time-Space Domain Dispersion-Relation-Based Finite-Difference Scheme,” Geophysics 75 (3), A11-A17 (2010).
    doi 10.1190/1.3374477
  29. J.-B. Chen, “A 27-Point Scheme for a 3D Frequency-Domain Scalar Wave Equation Based on an Average-Derivative Method,” Geophys. Prospect. 62 (2), 258-277 (2014).
    doi 10.1111/1365-2478.12090
  30. Z. He, J. Zhang, and Z. Yao, “Determining the Optimal Coefficients of the Explicit Finite-Difference Scheme Using the Remez Exchange Algorithm,” Geophysics 84 (3), S137-S147 (2019).
    doi 10.1190/geo2018-0446.1
  31. E. F. M. Koene, J. O. A. Robertsson, F. Broggini, and F. Andersson, “Eliminating Time Dispersion from Seismic Wave Modeling,” Geophys. J. Int. 213 (1), 169-180 (2018).
    doi 10.1093/gji/ggx563
  32. R. Mittet, “Second-Order Time Integration of the Wave Equation with Dispersion Correction Procedures,” Geophysics 84 (4), T221-T235 (2019).
    doi 10.1190/geo2018-0770.1
  33. W. A. Mulder, “Temporal Dispersion Correction for Wave-Propagation Modelling with a Series Approach,” Geophys. Prospect. 72 (2), 301-314 (2024).
    doi 10.1111/1365-2478.13411
  34. K. Gadylshin, D. Vishnevsky, K. Gadylshina, and V. Lisitsa, “Numerical Dispersion Mitigation Neural Network for Seismic Modeling,” Geophysics 87 (3), T237-T249 (2022).
    doi 10.1190/geo2021-0242.1
  35. K. Gadylshina, V. Lisitsa, D. Vishnevsky, and K. Gadylshin, “An Artificial Neural Network that Reduces Numerical Variance for Postprocessing the Results of Seismic Modeling,” Geophys. Technol., No. 1, 99-109 (2022).
    doi 10.18303/2619-1563-2022-1-99
  36. E. Gondyul, V. Lisitsa, K. Gadylshin, and D. Vishnevsky, “Numerical Dispersion Mitigation Neural Network with the Model-Based Training Dataset Optimization,” in Lecture Notes in Computer Science (Springer, Cham, 2023), Vol. 14106, pp. 19-30.
    doi 10.1007/978-3-031-37111-0_2
  37. K. A. Gadylshina, D. M. Vishnevsky, K. G. Gadylshin, and V. V. Lisitsa, “Training Data Set Construction Based on the Hausdorff Metric for Numerical Dispersion Mitigation Neural Network in Seismic Modelling,” Numerical Methods and Programming (Vychislitel’nye Metody i Programmirovanie). 24 (2), 195-212 (2023).
    doi 10.26089/NumMet.v24r215
  38. K. Gadylshin, V. Lisitsa, D. Vishnevsky, and K. Gadylshina, “Hausdorff-Distance-Based Training Dataset Construction for Numerical Dispersion Mitigation Neural Network,” Comput. and Geosci. 180, Article Number 105438 (2023).
    doi 10.1016/j.cageo.2023.105438
  39. R. Wang, Y. Meng, W. Zhang, et al., “Remote Sensing Semantic Segregation for Water Information Extraction: Optimization of Samples via Training Error Performance,” IEEE Access 7, 13383-13395 (2019).
    doi 10.1109/ACCESS.2019.2894099
  40. S. Yan, L. Zhang, and D. Liu, “An Empirical Study on Optimization of Training Dataset in Harmfulness Prediction of Code Clone Using Ensemble Feature Selection Model,” in Proc. Int. Conf. on Information and Communication Technologies for Disaster Management, Sendai, Japan, December 04-07, 2018.Published by IEEE Press in
    https://ieeexplore.ieee.org/xpl/conhome/8634895/proceeding . Cited March 29, 2024.
    doi 10.1109/ICT-DM.2018.8636375
  41. D. Everaert and C. Potts, “GIO: Gradient Information Optimization for Training Dataset Selection,”
    doi 10.48550/arXiv.2306.11670
  42. J. Park, J. Choi, S. J. Seol, et al., “A Method for Adequate Selection of Training Data Sets to Reconstruct Seismic Data Using a Convolutional U-Net,” Geophysics 86 (5), V375-V388 (2021).
    doi 10.1190/geo2019-0708.1
  43. I. M. Sobol’, S. Tarantola, D. Gatelli, et al., “Estimating the Approximation Error when Fixing Unessential Factors in Global Sensitivity Analysis,” Reliab. Eng. Syst. Saf. 92 (7), 957-960 (2007).
    doi 10.1016/j.ress.2006.07.001
  44. J.-P. Berenger, “A Huygens Subgridding for the FDTD Method,” IEEE Trans. Antennas Propag. 54 (12), 3797-3804 (2006).
    doi 10.1109/TAP.2006.886519
  45. F. Collino and C. Tsogka, “Application of the Perfectly Matched Absorbing Layer Model to the Linear Elastodynamic Problem in Anisotropic Heterogeneous Media,” Geophysics 66 (1), 294-307 (2001).
    doi 10.1190/1.1444908
  46. V. Lisitsa, “Optimal Discretization of PML for Elasticity Problems,” Electron. Trans. Numer. Anal. 30, 258-277 (2008).
    http://eudml.org/doc/130538 . Cited March 29, 2024.
  47. V. Kostin, V. Lisitsa, G. Reshetova, and V. Tcheverda, “Local Time-Space Mesh Refinement for Simulation of Elastic Wave Propagation in Multi-Scale Media,” J. Comput. Phys. 281, 669-689 (2015).
    doi 10.1016/j.jcp.2014.10.047
  48. A. A. Samarskii, Theory of Difference Schemes (Nauka, Moscow, 1983; Marcel Dekker, New York, 2001).
  49. P. Moczo, J. Kristek, V. Varyčuk, et al., “3D Heterogeneous Staggered-Grid Finite-Difference Modeling of Seismic Motion with Volume Harmonic and Arithmetic Averaging of Elastic Moduli and Densities,” Bull. Seismol. Soc. Am. 92 (8), 3042-3066 (2002).
    doi 10.1785/0120010167
  50. D. Vishnevsky, V. Lisitsa, V. Tcheverda, and G. Reshetova, “Numerical Study of the Interface Errors of Finite-Difference Simulations of Seismic Waves,” Geophysics 79 (4), T219-T232 (2014).
    doi 10.1190/geo2013-0299.1
  51. V. Lisitsa, O. Podgornova, and V. Tcheverda, “On the Interface Error Analysis for Finite Difference Wave Simulation,” Comput. Geosci. 14 (4), 769-778 (2010).
    doi 10.1007/s10596-010-9187-1
  52. M. Ainsworth, “Discrete Dispersion Relation for hp-Version Finite Element Approximation at High Wave Number,” SIAM J. Numer. Anal. 42 (2), 553-575 (2004).
    doi 10.1137/S0036142903423460
  53. O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” in Lecture Notes in Computer Science (Springer, Cham, 2015), Vol. 9351, pp. 234-241.
    doi 10.1007/978-3-319-24574-4_28
  54. I. M. Sobol’, “Global Sensitivity Indices for Nonlinear Mathematical Models and Their Monte Carlo Estimates,” Math. Comput. Simul. 55 (1-3), 271-280 (2001).
    doi 10.1016/S0378-4754(00)00270-6
  55. A. Saltelli, M. Ratto, T. Andres, et al., “Sensitivity Analysis: from Theory to Practice,” in Global Sensitivity Analysis. The Primer (John Wiley, Chichester, 2008), pp. 237-275.
    doi 10.1002/9780470725184.ch6
  56. D. Yeung, I. Cloete, D. Shi, and W. W. Y. Ng, Sensitivity Analysis for Neural Networks (Springer, Berlin, 2010).
    doi 10.1007/978-3-642-02532-7
  57. U. Reuter and M. Liebscher, “Global Sensitivity Analysis in View of Nonlinear Structural Behavior,”
    https://www.researchgate.net/profile/Martin_Liebscher/publication/242318342_Global_sensitivity_analysis_in_view_of_nonlinear_structural_behavior/links/58e213c5aca272059ab08f35/Global-sensitivity-analysis-in-view-of-nonlinear-structural-behavior.pdf . Cited March 29, 2024.
  58. J. Morio, “Global and Local Sensitivity Analysis Methods for a Physical System,” Eur. J. Phys. 32 (6), 1577-1583 (2011).
    doi 10.1088/0143-0807/32/6/011
  59. A. Saltelli and S. Tarantola, “On the Relative Importance of Input Factors in Mathematical Models: Safety Assessment for Nuclear Waste Disposal,” J. Am. Stat. Assoc. 97 (459), 702-709 (2002).
    doi 10.1198/016214502388618447
  60. D. W. van der Merwe and A. P. Engelbrecht, “Data Clustering Using Particle Swarm Optimization,” in Proc. 2003 Congress on Evolutionary Computation, Canberra, Australia, December 08-12, 2003.Published by IEEE Press in Congress on EvolutionaryComputation, Vol. 1, pp. 215-220 (2003):
    https://ieeexplore.ieee.org/xpl/conhome/9096/proceeding?isnumber=28874&sortType=vol-only-seq&pageNumber=2 . Cited March 29, 2024.
    doi 10.1109/CEC.2003.1299577

License

Copyright (c) 2024 Е. А. Гондюл, В. В. Лисица, К. Г. Гадыльшин, D. M. Вишневский

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.