Low-Frequency Reconstruction for Full Waveform Inversion by Unsupervised Learning

IF 2.9 3区 地球科学 Q2 ASTRONOMY & ASTROPHYSICS Earth and Space Science Pub Date : 2024-11-03 DOI:10.1029/2024EA003565
Ningcheng Ciu, Tao Lei, Wei Zhang
{"title":"Low-Frequency Reconstruction for Full Waveform Inversion by Unsupervised Learning","authors":"Ningcheng Ciu,&nbsp;Tao Lei,&nbsp;Wei Zhang","doi":"10.1029/2024EA003565","DOIUrl":null,"url":null,"abstract":"<p>Obtaining reliable low-frequency seismic data is crucial for effectively reducing cycle-skipping in full waveform inversion. However, acquiring high signal-to-noise ratio low-frequency information from field data remains a challenge. An effective solution to mitigate cycle-skipping is to utilize low-frequency information synthesized by neural networks to obtain low-wavenumber initial models. Previous attempts to reconstruct synthetic low-frequency data using supervised learning methods have shown feasibility but were limited to training with synthetic data that required labeled information. In this study, we employed an unsupervised learning method, namely cycle-consistent adversarial networks (CycleGAN), to reconstruct large-scale-feature related low-frequency information based on the high-frequency input data. Unlike supervised learning, CycleGAN allows the use of field data as input to train the network, which is more closely aligned with practical applications. Nevertheless, this approach presents challenges in terms of training complexity and potential output stability. To overcome these challenges, we reconstructed an appropriate target data set that combines high, medium, and low-frequency components and incorporated additional loss functions to enhance the network's output performance. We conducted quantitative evaluations of the method's sensitivity to the target data set and its ability to handle low-quality input data through numerical testing. The final results from field data testing confirmed the feasibility and effectiveness of the proposed method.</p>","PeriodicalId":54286,"journal":{"name":"Earth and Space Science","volume":null,"pages":null},"PeriodicalIF":2.9000,"publicationDate":"2024-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1029/2024EA003565","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Earth and Space Science","FirstCategoryId":"89","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1029/2024EA003565","RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ASTRONOMY & ASTROPHYSICS","Score":null,"Total":0}
引用次数: 0

Abstract

Obtaining reliable low-frequency seismic data is crucial for effectively reducing cycle-skipping in full waveform inversion. However, acquiring high signal-to-noise ratio low-frequency information from field data remains a challenge. An effective solution to mitigate cycle-skipping is to utilize low-frequency information synthesized by neural networks to obtain low-wavenumber initial models. Previous attempts to reconstruct synthetic low-frequency data using supervised learning methods have shown feasibility but were limited to training with synthetic data that required labeled information. In this study, we employed an unsupervised learning method, namely cycle-consistent adversarial networks (CycleGAN), to reconstruct large-scale-feature related low-frequency information based on the high-frequency input data. Unlike supervised learning, CycleGAN allows the use of field data as input to train the network, which is more closely aligned with practical applications. Nevertheless, this approach presents challenges in terms of training complexity and potential output stability. To overcome these challenges, we reconstructed an appropriate target data set that combines high, medium, and low-frequency components and incorporated additional loss functions to enhance the network's output performance. We conducted quantitative evaluations of the method's sensitivity to the target data set and its ability to handle low-quality input data through numerical testing. The final results from field data testing confirmed the feasibility and effectiveness of the proposed method.

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
通过无监督学习实现全波形反演的低频重构
获取可靠的低频地震数据对于有效减少全波形反演中的周期跳跃至关重要。然而,从野外数据中获取高信噪比的低频信息仍是一项挑战。减少周期跳跃的有效解决方案是利用神经网络合成的低频信息来获取低波数初始模型。之前使用监督学习方法重建合成低频数据的尝试已显示出可行性,但仅限于使用需要标记信息的合成数据进行训练。在这项研究中,我们采用了一种无监督学习方法,即循环一致性对抗网络(CycleGAN),来根据高频输入数据重建与大尺度特征相关的低频信息。与监督学习不同,CycleGAN 允许使用现场数据作为训练网络的输入,这与实际应用更为贴近。不过,这种方法在训练复杂性和潜在输出稳定性方面存在挑战。为了克服这些挑战,我们重建了一个适当的目标数据集,将高、中、低频成分结合在一起,并加入了额外的损失函数,以提高网络的输出性能。我们通过数值测试对该方法对目标数据集的灵敏度及其处理低质量输入数据的能力进行了定量评估。实地数据测试的最终结果证实了建议方法的可行性和有效性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Earth and Space Science
Earth and Space Science Earth and Planetary Sciences-General Earth and Planetary Sciences
CiteScore
5.50
自引率
3.20%
发文量
285
审稿时长
19 weeks
期刊介绍: Marking AGU’s second new open access journal in the last 12 months, Earth and Space Science is the only journal that reflects the expansive range of science represented by AGU’s 62,000 members, including all of the Earth, planetary, and space sciences, and related fields in environmental science, geoengineering, space engineering, and biogeochemistry.
期刊最新文献
Prospects of Predicting the Polar Motion Based on the Results of the Second Earth Orientation Parameters Prediction Comparison Campaign Empirical Model of SSUSI-Derived Auroral Ionization Rates Low-Frequency Reconstruction for Full Waveform Inversion by Unsupervised Learning Reconstruction of Nearshore Surface Gravity Wave Heights From Distributed Acoustic Sensing Data The Self-Calibrating Tilt Accelerometer: A Method for Observing Tilt and Correcting Drift With a Triaxial Accelerometer
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1