Оптимизация тренировочного набора данных для подавляющей численную дисперсию нейронной сети NDM-net
Авторы
-
Е. А. Гондюл
-
В. В. Лисица
-
К. Г. Гадыльшин
-
D. M. Вишневский
Ключевые слова:
численная дисперсия
сейсмическое моделирование
глубокое обучение
Аннотация
Предлагается оригинальный способ построения обучающего набора данных для нейронной сети NDM-net (Numerical Dispersion Mitigation neural network), подавляющей численную дисперсию при моделировании сейсмических волновых полей. NDM-net обучается отображать вычисленное на грубой сетке решение системы уравнений динамической теории упругости в рассчитанное на мелкой сетке. Данные сейсмограмм для обучения NDM-net предварительно рассчитываются на мелкой сетке, что является трудоемкой процедурой. Для снижения вычислительных затрат алгоритма время обучения необходимо сокращать без потери точности. В качестве эффективной метрики для генерации обучающего набора данных рассматривается линейная комбинация трех метрик: расстояния между источниками, меры сходства сейсмограмм и меры сходства скоростных моделей. Коэффициенты линейной комбинации определяются с помощью глобального анализа чувствительности.
Раздел
Методы и алгоритмы вычислительной математики и их приложения
Библиографические ссылки
- 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
- 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
- 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
- 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
- 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.
- J. M. Carcione, C. Morency, and J. E. Santos, “Computational Poroelasticity -- A Review,” Geophysics. 75 (5), 75A229-75A243 (2010).
doi 10.1190/1.3474602
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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.
- 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
- 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.
- 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
- 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
- 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
- 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
- B. Fornberg, “The Pseudospectral Method: Accurate Representation of Interfaces in Elastic Wave Calculations,” Geophysics 53 (5), 625-637 (1988).
doi 10.1190/1.1442497
- 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
- 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
- A. R. Levander, “Fourth-Order Finite-Difference P-SV Seismograms,” Geophysics. 53 (11), 1425-1436 (1988).
doi 10.1190/1.1442422
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- D. Everaert and C. Potts, “GIO: Gradient Information Optimization for Training Dataset Selection,”
doi 10.48550/arXiv.2306.11670
- 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
- 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
- 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
- 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
- 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.
- 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
- A. A. Samarskii, Theory of Difference Schemes (Nauka, Moscow, 1983; Marcel Dekker, New York, 2001).
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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
- 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.
- 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
- 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
- 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