All Issue

2024 Vol.27, Issue 3 Preview Page

Research Article

Seismic Data Processing Using BERT-Based Pretraining: Comparison of Shotgather Arrays

BERT 기반 사전학습을 이용한 탄성파 자료처리: 송신원 모음 배열 비교

Youngjae Shin1*
신 영재1*
1Department of Geology, Gyeongsang National University
1경상국립대학교 지질과학과
*Corresponding Author
31 August 2024. pp. 171-180
PDF XML Citation
Abstract

The processing of seismic data involves analyzing earthquake wave data to understand the internal structure and characteristics of the Earth, which requires high computational power. Recently, machine learning (ML) techniques have been introduced to address these challenges and have been utilized in various tasks such as noise reduction and velocity model construction. However, most studies have focused on specific seismic data processing tasks, limiting the full utilization of similar features and structures inherent in the datasets. In this study, we compared the efficacy of using receiver-wise time-series data (“receiver array”) and synchronized receiver signals (“time array”) from shotgathers for pretraining a Bidirectional Encoder Representations from Transformers (BERT) model. To this end, shotgather data generated from a synthetic model containing faults was used to perform noise reduction, velocity prediction, and fault detection tasks. In the task of random noise reduction, both the receiver and time arrays showed good performance. However, for tasks requiring the identification of spatial distributions, such as velocity estimation and fault detection, the results from the time array were superior.

Keywords
BERT
pretraining
seismic processing

탄성파 자료처리는 탄성파 자료를 분석하여 지구 내부 구조와 특성을 파악하는 기술로, 높은 컴퓨터 연산력이 요구된다. 이러한 도전 과제를 해결하기 위해 머신러닝 기술이 도입되었으며, 잡음 제거, 속도 모델 구축 등 다양한 작업에서 활용되고 있다. 그러나, 대부분의 연구는 특정 탄성파 처리 작업에 집중되어 있어 자료에 내재된 유사한 특징과 구조를 충분히 활용하지 못하는 한계가 있다. 본 연구에서는 BERT (Bidirectional Encoder Representations from Transformers) 기반의 사전학습을 위해 단일 송신원 모음에서 수신기별 시계열 자료(‘수신기 배열’)와 동일 시간에 기록된 수신기 신호(‘시간 배열’)를 입력 자료로 활용하는 방법을 비교하였다. 이를 위해 단층을 포함한 속도 모델에서 생성한 합성 송신원 모음 자료를 이용하여 잡음 제거, 속도 추정, 그리고 단층 확인 작업을 수행하였다. 임의 잡음 제거 작업에서는 수신기 및 시간 배열에서 모두 좋은 성능을 보였으나, 공간적인 분포 파악이 요구되는 속도 추정 및 단층 확인 작업에서는 시간 배열의 결과가 상대적으로 더 우수함을 확인하였다.

키워드
BERT
사전학습
탄성파 자료처리
References
1

Anjom, F. K., Vaccarino, F., and Socco, L. V., 2020, Machine learning for seismic exploration: Where are we and how far are we from the holy grail?, The Leading Edge, 39(10), 726-733. https://doi.org/10.1190/geo2023-0129.1

10.1190/geo2023-0129.1
2

Araya-Polo, M., Dahlke, T., Frogner, C., Zhang, C., Poggio, T., and Hohl, D., 2017, Automated fault detection without seismic processing, The Leading Edge, 36, 208-214. https://doi.org/10.1190/tle36030208.1

10.1190/tle36030208.1
3

Cova, D., Xie, P., and Trinh, P. T., 2020, Automated first break picking with constrained pooling networks, 90th Annual International Meeting, SEG, Expanded Abstracts, 1481-1485. https://doi.org/10.1190/segam2020-3427812.1

10.1190/segam2020-3427812.1
4

Devlin, J., Chang, M. W., Lee, K., and Toutanova, K., 2019, BERT: Pre-training of deep bidirectional transformers for language understanding, NAACL-HLT. https://doi.org/10.18653/v1/N19-1423

10.18653/v1/N19-1423
5

Graves, R. W., 1996, Simulating seismic wave propagation in 3-D elastic media using staggered-grid finite differences, Bulletin of the Seismological Society of America, 86(4), 1091-1106. https://doi.org/10.1785/BSSA0860041091

10.1785/BSSA0860041091
6

Harsuko, R. and Alkhalifah, T. A., 2022, StorSeismic: A new paradigm in deep learning for seismic processing, IEEE Transactions on Geoscience and Remote Sensing, 60, 1-15. https://doi.org/10.1109/TGRS.2022.3216660

10.1109/TGRS.2022.3216660
7

Hegna, S., Klüver, T., Lima, J., and Wisloff, J., 2019, Making the transition from discrete shot records to continuous seismic records and source wavefields, and its potential impact on survey efficiency and environmental footprint, Geophysical Prospecting, 67(6), 1472-1485. https:// doi.org/10.1111/1365-2478.12705

10.1111/1365-2478.12705
8

Helbig, K. and Thomsen, L., 2005, 75-plus years of anisotropy in exploration and reservoir seismics: a historical review of concepts and methods, Geophysics, 70, 9ND-23ND. https://doi.org/10.1190/1.2122407

10.1190/1.2122407
9

Kazei, V., Ovcharenko, O., and Alkhalifah, T., 2020, Velocity model building by deep learning: From general synthetics to field data application, 90th Annual International Meeting, SEG, Expanded Abstracts, 1561-1565. https://doi.org/10.1190/segam2020-3428324.1

10.1190/segam2020-3428324.1
10

Komatitsch, D. and Martin, R., 2007, An unsplit convolutional perfectly matched layer improved at grazing incidence for the seismic wave equation, Geophysics, 72(5), SM155-SM167. https://doi.org/10.1190/1.2757586

10.1190/1.2757586
11

Raissi, M., Perdikaris, P., and Karniadakis, G. E., 2019, Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations, Journal of Computational Physics, 378, 686-707. https://doi.org/10. 1016/j.jcp.2018.10.045

10.1016/j.jcp.2018.10.045
12

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I., 2017, Attention is all you need, Advances in Neural Information Processing Systems, 30. https://doi.org/10.48550/arXiv.1706.03762

13

Wang, E. and Nealon, J., 2019, Applying machine learning to 3D seismic image denoising and enhancement, Interpretation, 7(3), SE131-SE139. https://doi.org/10.1190/INT-2018-0224.1

10.1190/INT-2018-0224.1
14

Wu, H., Zhang, B., Li, F., and Liu, N., 2019a, Semiautomatic first-arrival picking of microseismic events by using the pixel-wise convolutional image segmentation method, Geophysics, 84(3), V143-V155. https://doi.org/10.1190/geo2018-0389.1

10.1190/geo2018-0389.1
15

Wu, H., Zhang, B., Lin, T., Li, F., and Liu, N., 2019b, White noise attenuation of seismic trace by integrating variational mode decomposition with convolutional neural network, Geophysics, 84(5), V307-V317. https://doi.org/10.1190/geo2018-0635.1

10.1190/geo2018-0635.1
16

Xu, Y., Li, J., and Chen, X., 2019, Physics informed neural networks for velocity inversion, 89th Annual International Meeting, SEG, Expanded Abstracts, 2584-2588. https://doi.org/10.1190/segam2019-3216823.1

10.1190/segam2019-3216823.1
17

Yang, F. and Ma, J., 2019, Deep-learning inversion: A next-generation seismic velocity model building method, Geophysics, 84(4), R583-R599. https://doi.org/10.1190/geo2018-0249.1

10.1190/geo2018-0249.1
18

Yilmaz, Ö., 2001, Seismic Data Analysis: Processing, Inversion, and Interpretation of Seismic Data, Society of Exploration Geophysicists. https:// doi.org/10.1190/1.9781560801580

10.1190/1.9781560801580
19

Yu, S., Ma, J., and Wang, W., 2019, Deep learning for denoising, Geophysics, 84(6), V333-V350. https://doi.org/10.1190/geo2018-0668.1

10.1190/geo2018-0668.1
Information
  • Publisher :Korean Society of Earth and Exploration Geophysicists
  • Publisher(Ko) :한국지구물리물리탐사학회
  • Journal Title :Geophysics and Geophysical Exploration
  • Journal Title(Ko) :지구물리와 물리탐사
  • Volume : 27
  • No :3
  • Pages :171-180
  • Received Date : 2024-07-19
  • Revised Date : 2024-08-13
  • Accepted Date : 2024-08-19
  • DOI :https://doi.org/10.7582/GGE.2024.27.3.171
Journal Informaiton Geophysics and Geophysical Exploration Geophysics and Geophysical Exploration
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • KCI 문헌 유사도 검사
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close