Retrieving S-wave velocity from surface wave multimode dispersion curves with DispINet

Lei Fu; Lei Pan; Qingbo Ma; Sheng Dong; Xiaofei Chen

doi:10.1016/j.jappgeo.2021.104430

Retrieving S-wave velocity from surface wave multimode dispersion curves with DispINet

Lei Fu, Lei Pan, Qingbo Ma, Sheng Dong, Xiaofei Chen

Research output: Contribution to journal › Article › peer-review

9 Scopus citations

Abstract

We present a data-driven dispersion curve inversion network (DispINet) that directly maps the multimode dispersion curves to the S-wave velocity model. DispINet is trained with synthetic samples that consist of multimode dispersion curve data and the corresponding S-wave velocity models. The S-wave velocity models are generated according to prior information in a specific study area. Multimode dispersion curves are calculated by the generalized reflection-transmission coefficient method for each model. The well-trained DispINet could be used to predict the S-wave velocity model for other dispersion curve data never seen by DispINet. Testing data and Qademah field data results show that DispINet has the ability to retrieve a reasonable underground S-wave velocity structure in real time.

Original language	English (US)
Pages (from-to)	104430
Journal	Journal of Applied Geophysics
Volume	193
DOIs	https://doi.org/10.1016/j.jappgeo.2021.104430
State	Published - Jul 28 2021
Externally published	Yes

ASJC Scopus subject areas

Geophysics

Access to Document

10.1016/j.jappgeo.2021.104430

Cite this

@article{9caf0a1525aa422d91e3b0ecebea2a75,

title = "Retrieving S-wave velocity from surface wave multimode dispersion curves with DispINet",

abstract = "We present a data-driven dispersion curve inversion network (DispINet) that directly maps the multimode dispersion curves to the S-wave velocity model. DispINet is trained with synthetic samples that consist of multimode dispersion curve data and the corresponding S-wave velocity models. The S-wave velocity models are generated according to prior information in a specific study area. Multimode dispersion curves are calculated by the generalized reflection-transmission coefficient method for each model. The well-trained DispINet could be used to predict the S-wave velocity model for other dispersion curve data never seen by DispINet. Testing data and Qademah field data results show that DispINet has the ability to retrieve a reasonable underground S-wave velocity structure in real time.",

author = "Lei Fu and Lei Pan and Qingbo Ma and Sheng Dong and Xiaofei Chen",

note = "KAUST Repository Item: Exported on 2021-08-23 Acknowledgements: This research was partially supported by the National Natural Science Foundation of China (Grant Nos. 41974044, U1901602, and 41790465). We thank the subsurface imaging and fluid modeling (CSIM) group at KAUST for providing the Qademah field data. This publication acknowledges KAUST support, but has no KAUST affiliated authors.",

year = "2021",

month = jul,

day = "28",

doi = "10.1016/j.jappgeo.2021.104430",

language = "English (US)",

volume = "193",

pages = "104430",

journal = "Journal of Applied Geophysics",

issn = "0926-9851",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Retrieving S-wave velocity from surface wave multimode dispersion curves with DispINet

AU - Fu, Lei

AU - Pan, Lei

AU - Ma, Qingbo

AU - Dong, Sheng

AU - Chen, Xiaofei

N1 - KAUST Repository Item: Exported on 2021-08-23 Acknowledgements: This research was partially supported by the National Natural Science Foundation of China (Grant Nos. 41974044, U1901602, and 41790465). We thank the subsurface imaging and fluid modeling (CSIM) group at KAUST for providing the Qademah field data. This publication acknowledges KAUST support, but has no KAUST affiliated authors.

PY - 2021/7/28

Y1 - 2021/7/28

N2 - We present a data-driven dispersion curve inversion network (DispINet) that directly maps the multimode dispersion curves to the S-wave velocity model. DispINet is trained with synthetic samples that consist of multimode dispersion curve data and the corresponding S-wave velocity models. The S-wave velocity models are generated according to prior information in a specific study area. Multimode dispersion curves are calculated by the generalized reflection-transmission coefficient method for each model. The well-trained DispINet could be used to predict the S-wave velocity model for other dispersion curve data never seen by DispINet. Testing data and Qademah field data results show that DispINet has the ability to retrieve a reasonable underground S-wave velocity structure in real time.

AB - We present a data-driven dispersion curve inversion network (DispINet) that directly maps the multimode dispersion curves to the S-wave velocity model. DispINet is trained with synthetic samples that consist of multimode dispersion curve data and the corresponding S-wave velocity models. The S-wave velocity models are generated according to prior information in a specific study area. Multimode dispersion curves are calculated by the generalized reflection-transmission coefficient method for each model. The well-trained DispINet could be used to predict the S-wave velocity model for other dispersion curve data never seen by DispINet. Testing data and Qademah field data results show that DispINet has the ability to retrieve a reasonable underground S-wave velocity structure in real time.

UR - http://hdl.handle.net/10754/670716

UR - https://linkinghub.elsevier.com/retrieve/pii/S0926985121001774

UR - http://www.scopus.com/inward/record.url?scp=85112486993&partnerID=8YFLogxK

U2 - 10.1016/j.jappgeo.2021.104430

DO - 10.1016/j.jappgeo.2021.104430

M3 - Article

SN - 0926-9851

VL - 193

SP - 104430

JO - Journal of Applied Geophysics

JF - Journal of Applied Geophysics

ER -

Retrieving S-wave velocity from surface wave multimode dispersion curves with DispINet

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this