Standard

An artificial neural network approach for the inversion of surface wave dispersion curves. / Yablokov, Alexandr V.; Serdyukov, Alexandr S.; Loginov, Georgy N. et al.

In: Geophysical Prospecting, Vol. 69, No. 7, 09.2021, p. 1405-1432.

Research output: Contribution to journalArticlepeer-review

Harvard

Yablokov, AV, Serdyukov, AS, Loginov, GN & Baranov, VD 2021, 'An artificial neural network approach for the inversion of surface wave dispersion curves', Geophysical Prospecting, vol. 69, no. 7, pp. 1405-1432. https://doi.org/10.1111/1365-2478.13107

APA

Yablokov, A. V., Serdyukov, A. S., Loginov, G. N., & Baranov, V. D. (2021). An artificial neural network approach for the inversion of surface wave dispersion curves. Geophysical Prospecting, 69(7), 1405-1432. https://doi.org/10.1111/1365-2478.13107

Vancouver

Yablokov AV, Serdyukov AS, Loginov GN, Baranov VD. An artificial neural network approach for the inversion of surface wave dispersion curves. Geophysical Prospecting. 2021 Sept;69(7):1405-1432. doi: 10.1111/1365-2478.13107

Author

Yablokov, Alexandr V. ; Serdyukov, Alexandr S. ; Loginov, Georgy N. et al. / An artificial neural network approach for the inversion of surface wave dispersion curves. In: Geophysical Prospecting. 2021 ; Vol. 69, No. 7. pp. 1405-1432.

BibTeX

@article{95daf267b2424172a8fd76a66ac1ccd8,
title = "An artificial neural network approach for the inversion of surface wave dispersion curves",
abstract = "We describe a new algorithm for the inversion of one-dimensional shear-wave velocity profiles from dispersion curves of the fundamental mode of Rayleigh surface waves. The novelties of our approach are that the layer velocities and thicknesses are set as unknowns, and an artificial neural network is proposed to solve the inverse problem. We suggest that training data should be calculated for a set of random synthetic velocity layered models, while layer thicknesses and velocities should be set to fixed intervals, with ranges estimated based on the systematic application of empirical relations between Rayleigh and S-wave velocities to the dispersion data. Our main challenge is a total overhaul of the artificial neural network, which includes selecting the optimal artificial neural network architecture and parameters by performing a large number of numerical experiments. Our synthetic results show that the accuracy of the proposed approach outperforms that of the Monte Carlo approach. We illustrate our proposed method with West Siberia data processing obtained from an area of approximately 800 (Formula presented.). From a user perspective, the main strength of our method is the computationally efficient processing of large amounts of dispersion data, which make it well suited for four-dimensional near-surface monitoring.",
keywords = "Inversion, Surface wave",
author = "Yablokov, {Alexandr V.} and Serdyukov, {Alexandr S.} and Loginov, {Georgy N.} and Baranov, {Valery D.}",
note = "Funding Information: The algorithmic and methodical parts of the reported study were funded by the Russian Science Foundation (RSF), project number 20‐77‐10023. The practical part of the reported study (field data processing) was funded by the Russian Foundation for Basic Research (RFBR), project number 19‐35‐90055. We would like to thank the associate editor of this paper, Clement Kostov, the reviewer Zheng‐Dong Zhang, and another anonymous reviewer for their comments and corrections. Publisher Copyright: {\textcopyright} 2021 European Association of Geoscientists & Engineers",
year = "2021",
month = sep,
doi = "10.1111/1365-2478.13107",
language = "English",
volume = "69",
pages = "1405--1432",
journal = "Geophysical Prospecting",
issn = "0016-8025",
publisher = "Wiley-Blackwell",
number = "7",

}

RIS

TY - JOUR

T1 - An artificial neural network approach for the inversion of surface wave dispersion curves

AU - Yablokov, Alexandr V.

AU - Serdyukov, Alexandr S.

AU - Loginov, Georgy N.

AU - Baranov, Valery D.

N1 - Funding Information: The algorithmic and methodical parts of the reported study were funded by the Russian Science Foundation (RSF), project number 20‐77‐10023. The practical part of the reported study (field data processing) was funded by the Russian Foundation for Basic Research (RFBR), project number 19‐35‐90055. We would like to thank the associate editor of this paper, Clement Kostov, the reviewer Zheng‐Dong Zhang, and another anonymous reviewer for their comments and corrections. Publisher Copyright: © 2021 European Association of Geoscientists & Engineers

PY - 2021/9

Y1 - 2021/9

N2 - We describe a new algorithm for the inversion of one-dimensional shear-wave velocity profiles from dispersion curves of the fundamental mode of Rayleigh surface waves. The novelties of our approach are that the layer velocities and thicknesses are set as unknowns, and an artificial neural network is proposed to solve the inverse problem. We suggest that training data should be calculated for a set of random synthetic velocity layered models, while layer thicknesses and velocities should be set to fixed intervals, with ranges estimated based on the systematic application of empirical relations between Rayleigh and S-wave velocities to the dispersion data. Our main challenge is a total overhaul of the artificial neural network, which includes selecting the optimal artificial neural network architecture and parameters by performing a large number of numerical experiments. Our synthetic results show that the accuracy of the proposed approach outperforms that of the Monte Carlo approach. We illustrate our proposed method with West Siberia data processing obtained from an area of approximately 800 (Formula presented.). From a user perspective, the main strength of our method is the computationally efficient processing of large amounts of dispersion data, which make it well suited for four-dimensional near-surface monitoring.

AB - We describe a new algorithm for the inversion of one-dimensional shear-wave velocity profiles from dispersion curves of the fundamental mode of Rayleigh surface waves. The novelties of our approach are that the layer velocities and thicknesses are set as unknowns, and an artificial neural network is proposed to solve the inverse problem. We suggest that training data should be calculated for a set of random synthetic velocity layered models, while layer thicknesses and velocities should be set to fixed intervals, with ranges estimated based on the systematic application of empirical relations between Rayleigh and S-wave velocities to the dispersion data. Our main challenge is a total overhaul of the artificial neural network, which includes selecting the optimal artificial neural network architecture and parameters by performing a large number of numerical experiments. Our synthetic results show that the accuracy of the proposed approach outperforms that of the Monte Carlo approach. We illustrate our proposed method with West Siberia data processing obtained from an area of approximately 800 (Formula presented.). From a user perspective, the main strength of our method is the computationally efficient processing of large amounts of dispersion data, which make it well suited for four-dimensional near-surface monitoring.

KW - Inversion

KW - Surface wave

UR - http://www.scopus.com/inward/record.url?scp=85107587875&partnerID=8YFLogxK

U2 - 10.1111/1365-2478.13107

DO - 10.1111/1365-2478.13107

M3 - Article

AN - SCOPUS:85107587875

VL - 69

SP - 1405

EP - 1432

JO - Geophysical Prospecting

JF - Geophysical Prospecting

SN - 0016-8025

IS - 7

ER -

ID: 29280479