Multi-task learning for seismic elastic parameter inversion with the lateral constraint of angle-gather difference

IF 6 1区 工程技术 Q2 ENERGY & FUELS Petroleum Science Pub Date : 2024-12-01 DOI:10.1016/j.petsci.2024.06.010
Pu Wang , Yi-An Cui , Lin Zhou , Jing-Ye Li , Xin-Peng Pan , Ya Sun , Jian-Xin Liu
{"title":"Multi-task learning for seismic elastic parameter inversion with the lateral constraint of angle-gather difference","authors":"Pu Wang ,&nbsp;Yi-An Cui ,&nbsp;Lin Zhou ,&nbsp;Jing-Ye Li ,&nbsp;Xin-Peng Pan ,&nbsp;Ya Sun ,&nbsp;Jian-Xin Liu","doi":"10.1016/j.petsci.2024.06.010","DOIUrl":null,"url":null,"abstract":"<div><div>Pre-stack seismic inversion is an effective way to investigate the characteristics of hydrocarbon-bearing reservoirs. Multi-parameter application is the key to identifying reservoir lithology and fluid in pre-stack inversion. However, multi-parameter inversion may bring coupling effects on the parameters and destabilize the inversion. In addition, the lateral recognition accuracy of geological structures receives great attention. To address these challenges, a multi-task learning network considering the angle-gather difference is proposed in this work. The deep learning network is usually assumed as a black box and it is unclear what it can learn. However, the introduction of angle-gather difference can force the deep learning network to focus on the lateral differences, thus improving the lateral accuracy of the prediction profile. The proposed deep learning network includes input and output blocks. First, angle gathers and the angle-gather difference are fed into two separate input blocks with ResNet architecture and Unet architecture, respectively. Then, three elastic parameters, including P- and S-wave velocities and density, are simultaneously predicted based on the idea of multi-task learning by using three separate output blocks with the same convolutional network layers. Experimental and field data tests demonstrate the effectiveness of the proposed method in improving the prediction accuracy of seismic elastic parameters.</div></div>","PeriodicalId":19938,"journal":{"name":"Petroleum Science","volume":"21 6","pages":"Pages 4001-4009"},"PeriodicalIF":6.0000,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Petroleum Science","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1995822624001687","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENERGY & FUELS","Score":null,"Total":0}
引用次数: 0

Abstract

Pre-stack seismic inversion is an effective way to investigate the characteristics of hydrocarbon-bearing reservoirs. Multi-parameter application is the key to identifying reservoir lithology and fluid in pre-stack inversion. However, multi-parameter inversion may bring coupling effects on the parameters and destabilize the inversion. In addition, the lateral recognition accuracy of geological structures receives great attention. To address these challenges, a multi-task learning network considering the angle-gather difference is proposed in this work. The deep learning network is usually assumed as a black box and it is unclear what it can learn. However, the introduction of angle-gather difference can force the deep learning network to focus on the lateral differences, thus improving the lateral accuracy of the prediction profile. The proposed deep learning network includes input and output blocks. First, angle gathers and the angle-gather difference are fed into two separate input blocks with ResNet architecture and Unet architecture, respectively. Then, three elastic parameters, including P- and S-wave velocities and density, are simultaneously predicted based on the idea of multi-task learning by using three separate output blocks with the same convolutional network layers. Experimental and field data tests demonstrate the effectiveness of the proposed method in improving the prediction accuracy of seismic elastic parameters.
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
利用角-集差横向约束的地震弹性参数反演多任务学习
叠前地震反演是研究含油气储层特征的有效方法。在叠前反演中,多参数应用是确定储层岩性和流体的关键。然而,多参数反演可能会带来参数耦合效应,破坏反演的稳定性。此外,地质结构的横向识别精度也备受关注。为应对这些挑战,本研究提出了一种考虑角-集差的多任务学习网络。深度学习网络通常被认为是一个黑盒子,不清楚它能学到什么。然而,引入角度聚集差可以迫使深度学习网络关注横向差异,从而提高预测剖面的横向精度。拟议的深度学习网络包括输入和输出块。首先,角度采集和角度采集差值分别被输入到两个独立的输入块中,这两个输入块分别采用 ResNet 架构和 Unet 架构。然后,基于多任务学习的思想,使用具有相同卷积网络层的三个独立输出块,同时预测三个弹性参数,包括 P 波和 S 波速度以及密度。实验和现场数据测试证明了所提方法在提高地震弹性参数预测精度方面的有效性。
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
Petroleum Science
Petroleum Science 地学-地球化学与地球物理
CiteScore
7.70
自引率
16.10%
发文量
311
审稿时长
63 days
期刊介绍: Petroleum Science is the only English journal in China on petroleum science and technology that is intended for professionals engaged in petroleum science research and technical applications all over the world, as well as the managerial personnel of oil companies. It covers petroleum geology, petroleum geophysics, petroleum engineering, petrochemistry & chemical engineering, petroleum mechanics, and economic management. It aims to introduce the latest results in oil industry research in China, promote cooperation in petroleum science research between China and the rest of the world, and build a bridge for scientific communication between China and the world.
期刊最新文献
Characterization of petrophysical and seismic properties for CO2 storage with sensitivity analysis Exceptional functionalized dual-acidic ionic liquid: High-efficiency catalytic reaction medium for oxidation desulfurization Machine learning-based grayscale analyses for lithofacies identification of the Shahejie formation, Bohai Bay Basin, China New alkylbenzene parameters to identify organic matter sources for source rocks of light oils and condensates from the Tarim Basin and Beibuwan Basin Integrating well logs, 3D seismic, and earthquake data for comprehensive prediction of 3D in-situ stress orientations: A case study from the Weiyuan area in the Sichuan Basin, China
×
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