Article(id=1245407861216490258, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156262727438951343, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2402911, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1713628800000, receivedDateStr=2024-04-21, revisedDate=1722268800000, revisedDateStr=2024-07-30, acceptedDate=null, acceptedDateStr=null, onlineDate=1774857972672, onlineDateStr=2026-03-30, pubDate=1741363200000, pubDateStr=2025-03-08, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1774857972672, onlineIssueDateStr=2026-03-30, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1774857972672, creator=13701087609, updateTime=1774857972672, updator=13701087609, issue=Issue{id=1156262727438951343, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='7', pageStart='2193', pageEnd='3077', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1753604116544, creator=13701087609, updateTime=1753771263994, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1156963794699248405, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156262727438951343, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1156963794699248406, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156262727438951343, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=2683, endPage=2690, ext={EN=ArticleExt(id=1245407861812081449, articleId=1245407861216490258, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=2D Controlled Source Audio-frequency Magnetotelluric Inversion Based on Deep Learning, columnId=1156262729351549255, journalTitle=Science Technology and Engineering, columnName=Papers·Astronomy and Geosciences, runingTitle=null, highlight=null, articleAbstract=

Controlled-source audio-frequency magnetotellurics (CSAMT) uses artificial sources, providing strong anti-interference capabilities. It is widely used in oil exploration, mineral surveys and other areas. Traditional 2D inversion technology is mature, and deep learning has recently made some research advancements in geophysical exploration. There is still a research gap in applying deep learning to CSAMT inversion. Therefore, developing a 2D inversion algorithm for CSAMT based on deep learning is highly significant for advancing the use of deep learning in electromagnetic exploration. The characteristics of deep learning components such as convolutional layers, pooling layers, fully connected layers, and the UNet network were introduced. An explanation was provided on how to construct the training dataset, the UNet network used in this study, and how to set various training parameters. The network was saved after training. When the inversion was needed, the net was loaded and the algorithm could predict the result. Several theoretical models were designed for inversion, and the experiment results verified the reliability and effectiveness of the algorithm. The time of the deep learning inversion and the tranditional inversion was recorded. Building training set needed much time, but the time of deep learning inverison was much less than the tranditional inversion. The deep learning inversion is more efficient than the traditional inversion.

, correspAuthors=Han-dong TAN, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=Mao WANG, Han-dong TAN, Xing FU), CN=ArticleExt(id=1245407864127337405, articleId=1245407861216490258, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于深度学习的可控源音频大地电磁二维反演, columnId=1156262730077163858, journalTitle=科学技术与工程, columnName=论文·天文学、地球科学, runingTitle=null, highlight=null, articleAbstract=

可控源音频大地电磁(controlled-source audio-frequency magnetotellurics,CSAMT)采用人工场源,具有较强的抗干扰性,广泛应用于油气勘探、矿产普查等方面。传统的二维反演技术已发展成熟,深度学习目前在地球物理探测中有了一些研究进展,但深度学习在CSAMT反演中的研究还是空缺,因此开发基于深度学习的CSAMT二维反演算法对推进深度学习在电磁勘探中的发展非常有意义。对深度学习的卷积层、池化层、全连接层和UNet网络的特点进行了介绍;对如何构建训练集、本文所采用的UNet网络以及如何设置训练的各参数进行了阐述。将训练好的网络储存下来,在做反演计算时,将网络导入程序中,用网络对观测数据进行预测,得到反演结果。设计了多个理论模型进行反演试算,实验结果验证了算法的可靠性和有效性。对深度学习反演和数据空间OCAAM反演的计算时间进行了统计,在构建训练集和训练网络时需要较多时间,但采用训练好的网络反演的时间远低于传统反演的计算时间,具有反演速度快的特点。

, correspAuthors=谭捍东, authorNote=null, correspAuthorsNote=
* 谭捍东(1966—),男,汉族,湖北松滋人,博士,教授。研究方向:电磁法勘探。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=GhmWLURFhnIoC3w/2JJl8Q==, magXml=X5P7F3QOvRC9PLAtuXTkMQ==, pdfUrl=null, pdf=YS62NUTZ3I8GIvlZTcbQtA==, pdfFileSize=5363425, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=2YnDoRToXpuvFApiW++UOA==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=mtjSeDx/UTzGQqDpmhMCQw==, mapNumber=null, authorCompany=null, fund=null, authors=

汪茂(1982—),男,汉族,湖北天门人,博士,高级工程师。研究方向:地球物理中的机器学习和并行计算。E-mail:

, authorsList=汪茂, 谭捍东, 付兴)}, authors=[Author(id=1245407864660014068, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=wangmao@cugb.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1245407864781648895, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, authorId=1245407864660014068, language=EN, stringName=Mao WANG, firstName=Mao, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Information Engineering, China University of Geosciences, Beijing, 100083, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1245407864882311175, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, authorId=1245407864660014068, language=CN, stringName=汪茂, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 中国地质大学(北京)信息工程学院, 北京 100083, bio={"content":"

汪茂(1982—),男,汉族,湖北天门人,博士,高级工程师。研究方向:地球物理中的机器学习和并行计算。E-mail:

"}, bioImg=null, bioContent=

汪茂(1982—),男,汉族,湖北天门人,博士,高级工程师。研究方向:地球物理中的机器学习和并行计算。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1245407864416744406, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, xref=1, ext=[AuthorCompanyExt(id=1245407864425133015, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, companyId=1245407864416744406, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Information Engineering, China University of Geosciences, Beijing, 100083, China), AuthorCompanyExt(id=1245407864433521625, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, companyId=1245407864416744406, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国地质大学(北京)信息工程学院, 北京 100083)])]), Author(id=1245407864991363095, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=thd@cugb.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1245407865205272620, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, authorId=1245407864991363095, language=EN, stringName=Han-dong TAN, firstName=Han-dong, middleName=null, lastName=TAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, *, address=2 School of Geophysics and Information Technology, China University of Geosciences, Beijing, 100083, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1245407865394016318, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, authorId=1245407864991363095, language=CN, stringName=谭捍东, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, *, address=2 中国地质大学(北京)地球物理与信息技术学院, 北京 100083, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1245407864550962151, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, xref=2, ext=[AuthorCompanyExt(id=1245407864559350761, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, companyId=1245407864550962151, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Geophysics and Information Technology, China University of Geosciences, Beijing, 100083, China), AuthorCompanyExt(id=1245407864563545067, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, companyId=1245407864550962151, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国地质大学(北京)地球物理与信息技术学院, 北京 100083)])]), Author(id=1245407865490485319, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1245407865641480274, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, authorId=1245407865490485319, language=EN, stringName=Xing FU, firstName=Xing, middleName=null, lastName=FU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 School of Geophysics and Information Technology, China University of Geosciences, Beijing, 100083, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1245407865901527133, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, authorId=1245407865490485319, language=CN, stringName=付兴, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 中国地质大学(北京)地球物理与信息技术学院, 北京 100083, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1245407864550962151, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, xref=2, ext=[AuthorCompanyExt(id=1245407864559350761, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, companyId=1245407864550962151, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Geophysics and Information Technology, China University of Geosciences, Beijing, 100083, China), AuthorCompanyExt(id=1245407864563545067, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, companyId=1245407864550962151, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国地质大学(北京)地球物理与信息技术学院, 北京 100083)])])], keywords=[Keyword(id=1245407869215027313, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, orderNo=1, keyword=deep learning), Keyword(id=1245407869370216581, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, orderNo=2, keyword=UNet), Keyword(id=1245407869496045716, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, orderNo=3, keyword=controlled source audio-frequency magnetotelluric (CSAMT)), Keyword(id=1245407869672206508, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, orderNo=4, keyword=inversion), Keyword(id=1245407869781258428, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, orderNo=1, keyword=深度学习), Keyword(id=1245407869911281864, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, orderNo=2, keyword=UNet), Keyword(id=1245407870024528082, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, orderNo=3, keyword=CSAMT), Keyword(id=1245407870137774302, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, orderNo=4, keyword=反演)], refs=[Reference(id=1245407872507556353, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2004, volume=47, issue=4, pageStart=723, pageEnd=730, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=底青云, Martyn Unsworth, 王妙月, journalName=地球物理学报, refType=null, unstructuredReference=底青云, Martyn Unsworth, 王妙月. 复杂介质有限元法2.5维可控源音频大地电磁法数值模拟[J]. 地球物理学报, 2004, 47(4): 723-730., articleTitle=复杂介质有限元法2.5维可控源音频大地电磁法数值模拟, refAbstract=null), Reference(id=1245407872662745619, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2004, volume=47, issue=4, pageStart=723, pageEnd=730, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Di Qingyun, Martyn Unsworth, Wang Miaoyue, journalName=Chinese Journal of Geophysics, refType=null, unstructuredReference=Di Qingyun, Martyn Unsworth, Wang Miaoyue. Controlled source audio-frequency magnetotelluric 2.5 d numerical simulation of finite element method in complex medium[J]. Chinese Journal of Geophysics, 2004, 47(4): 723-730., articleTitle=Controlled source audio-frequency magnetotelluric 2.5 d numerical simulation of finite element method in complex medium, refAbstract=null), Reference(id=1245407872809546274, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2013, volume=35, issue=4, pageStart=435, pageEnd=441, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=孟庆奎, 林品荣, 徐宝利, journalName=物探化探计算技术, refType=null, unstructuredReference=孟庆奎, 林品荣, 徐宝利, . 张量CSAMT一维数值模拟分析[J]. 物探化探计算技术, 2013, 35(4): 435-441., articleTitle=张量CSAMT一维数值模拟分析, refAbstract=null), Reference(id=1245407872952152629, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2013, volume=35, issue=4, pageStart=435, pageEnd=441, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=Meng Qingkui, Lin Pinrong, Xu Baoli, journalName=Computing Techniques for Geophysical and Geochemical, refType=null, unstructuredReference=Meng Qingkui, Lin Pinrong, Xu Baoli, et al. An analysis of tensor CSAMT 1D numerical simulation[J]. Computing Techniques for Geophysical and Geochemical, 2013, 35(4): 435-441., articleTitle=An analysis of tensor CSAMT 1D numerical simulation, refAbstract=null), Reference(id=1245407873073787459, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2018, volume=37, issue=12, pageStart=894, pageEnd=901, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=Kim Y, Nakata N, journalName=The Leading Edge, refType=null, unstructuredReference=Kim Y, Nakata N. Geophysical inversion versus machine learning in inverse problems[J]. The Leading Edge, 2018, 37(12): 894-901., articleTitle=Geophysical inversion versus machine learning in inverse problems, refAbstract=null), Reference(id=1245407873233171030, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=张月, journalName=基于深度学习的磁法反演, refType=null, unstructuredReference=张月. 基于深度学习的磁法反演[D]. 荆州: 长江大学, 2020., articleTitle=null, refAbstract=null), Reference(id=1245407873363194468, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Zhang Yue, journalName=Magnetic inversion based on deep learning, refType=null, unstructuredReference=Zhang Yue. Magnetic inversion based on deep learning[D]. Jingzhou: Yangtze University, 2020., articleTitle=null, refAbstract=null), Reference(id=1245407873505800821, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=李丽丽, journalName=基于机器学习的大地电磁反演研究, refType=null, unstructuredReference=李丽丽. 基于机器学习的大地电磁反演研究[D]. 桂林: 桂林理工大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1245407873640018561, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=Li Lili, journalName=Guilin University of Technology, refType=null, unstructuredReference=Li Lili. Research on magnetotelluric inversion based on machine learning[D]. Guilin: Guilin University of Technology, 2023., articleTitle=null, refAbstract=null), Reference(id=1245407873749070476, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=43, issue=2, pageStart=344, pageEnd=354, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=王方, 能杰, 田慧潇, journalName=地质科技通报, refType=null, unstructuredReference=王方, 能杰, 田慧潇, . 基于深度学习的大地电磁二维反演方法[J]. 地质科技通报, 2024, 43(2): 344-354., articleTitle=基于深度学习的大地电磁二维反演方法, refAbstract=null), Reference(id=1245407873891676824, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=43, issue=2, pageStart=344, pageEnd=354, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=Wang Fang, Xiong Jie, Tian Huixiao, journalName=Bulletin of Geological Science and Technology, refType=null, unstructuredReference=Wang Fang, Xiong Jie, Tian Huixiao, et al. 2D magnetotelluric inversion method based on deep learning[J]. Bulletin of Geological Science and Technology, 2024, 43(2): 344-354., articleTitle=2D magnetotelluric inversion method based on deep learning, refAbstract=null), Reference(id=1245407874021700262, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=39, issue=4, pageStart=1457, pageEnd=1473, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=刘卫强, 林品荣, 陈儒军, journalName=地球物理学进展, refType=null, unstructuredReference=刘卫强, 林品荣, 陈儒军, . 基于深度学习与联合降噪的可控源音频大地电磁抗干扰数据处理方法研究[J]. 地球物理学进展, 2024, 39 (4): 1457-1473., articleTitle=基于深度学习与联合降噪的可控源音频大地电磁抗干扰数据处理方法研究, refAbstract=null), Reference(id=1245407874235609777, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=39, issue=4, pageStart=1457, pageEnd=1473, url=null, language=null, rfNumber=[7], rfOrder=12, authorNames=Liu Weiqiang, Lin Pinrong, Chen Rujun, journalName=Progress in Geophysics, refType=null, unstructuredReference=Liu Weiqiang, Lin Pinrong, Chen Rujun, et al. Anti-interference processing for CSAMT based on deep learning and joint denoising[J]. Progress in Geophysics, 2024, 39 (4): 1457-1473., articleTitle=Anti-interference processing for CSAMT based on deep learning and joint denoising, refAbstract=null), Reference(id=1245407874357244607, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2019, volume=84, issue=4, pageStart=583, pageEnd=599, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=Yang F S, Ma J W, journalName=Geophysics, refType=null, unstructuredReference=Yang F S, Ma J W. Deep-learning inversion: a next-generation seismic velocity model building method[J]. Geophysics, 2019, 84 (4): 583-599., articleTitle=Deep-learning inversion: a next-generation seismic velocity model building method, refAbstract=null), Reference(id=1245407874487268046, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2019, volume=218, issue=2, pageStart=817, pageEnd=832, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=Puzyrev V, journalName=Geophysical Journal International, refType=null, unstructuredReference=Puzyrev V. Deep learning electromagnetic inversion with convolutional neural networks[J]. Geophysical Journal International, 2019, 218 (2): 817-832., articleTitle=Deep learning electromagnetic inversion with convolutional neural networks, refAbstract=null), Reference(id=1245407874608902872, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=1992, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=纳比吉安, journalName=电磁法, refType=null, unstructuredReference=纳比吉安. 电磁法[M]. 北京: 地质出版社, 1992., articleTitle=null, refAbstract=null), Reference(id=1245407874785063657, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=1992, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=16, authorNames=Nabi Gian, journalName=Electromagnetic method, refType=null, unstructuredReference=Nabi Gian. Electromagnetic method[M]. Beijing: Geological Publishing House, 1992., articleTitle=null, refAbstract=null), Reference(id=1245407874990584575, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2009, volume=74, issue=2, pageStart=9, pageEnd=20, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=Key K, journalName=Geophysics, refType=null, unstructuredReference=Key K. 1D inversion of multicomponent, multifrequency marine CSEM data: methodogy and synthetic studies for resolving thin resistive layers[J]. Geophysics, 2009, 74(2): 9-20., articleTitle=1D inversion of multicomponent, multifrequency marine CSEM data: methodogy and synthetic studies for resolving thin resistive layers, refAbstract=null), Reference(id=1245407875108025098, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2014, volume=11, issue=3, pageStart=286, pageEnd=294, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=雷达, 张国鸿, 黄高元, journalName=工程地球物理学报, refType=null, unstructuredReference=雷达, 张国鸿, 黄高元, . 张量可控源音频大地电磁法的应用实例[J]. 工程地球物理学报, 2014, 11(3): 286-294., articleTitle=张量可控源音频大地电磁法的应用实例, refAbstract=null), Reference(id=1245407875212882708, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2014, volume=11, issue=3, pageStart=286, pageEnd=294, url=null, language=null, rfNumber=[12], rfOrder=19, authorNames=Lei Da, Zhang Guohong, Huang Gaoyuan, journalName=Chinese Journal of Engineering Geophysics, refType=null, unstructuredReference=Lei Da, Zhang Guohong, Huang Gaoyuan, et al. Application examples of tensor CSAMT[J]. Chinese Journal of Engineering Geophysics, 2014, 11(3): 286-294., articleTitle=Application examples of tensor CSAMT, refAbstract=null), Reference(id=1245407875363877670, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=20, authorNames=桂兵, journalName=可控源音频大地电磁法张量数据二维反演研究, refType=null, unstructuredReference=桂兵. 可控源音频大地电磁法张量数据二维反演研究[D]. 北京: 中国地质大学(北京), 2014., articleTitle=null, refAbstract=null), Reference(id=1245407875472929589, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=21, authorNames=Gui Bing, journalName=Research of 2D CSAMT tensor data inversion, refType=null, unstructuredReference=Gui Bing. Research of 2D CSAMT tensor data inversion[D]. Beijing: China University of Geosciences (Beijing), 2014., articleTitle=null, refAbstract=null), Reference(id=1245407875590370111, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2013, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=22, authorNames=李晋津, journalName=可控源音频大地电磁法二维OCCAM反演研究, refType=null, unstructuredReference=李晋津. 可控源音频大地电磁法二维OCCAM反演研究[D]. 北京: 中国地质大学(北京), 2013., articleTitle=null, refAbstract=null), Reference(id=1245407875724587854, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2013, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=23, authorNames=Li Jinjin, journalName=A research on OCCAM inversion of 2D CSAMT, refType=null, unstructuredReference=Li Jinjin. A research on OCCAM inversion of 2D CSAMT[D]. Beijing: China University of Geosciences (Beijing), 2013., articleTitle=null, refAbstract=null), Reference(id=1245407875879777112, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2009, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=24, authorNames=何梅兴, journalName=可控源音频大地电磁二维OCCAM反演研究, refType=null, unstructuredReference=何梅兴. 可控源音频大地电磁二维OCCAM反演研究[D]. 武汉: 中国地质大学, 2009., articleTitle=null, refAbstract=null), Reference(id=1245407875993023333, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2009, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=25, authorNames=He Meixing, journalName=A study on CSAMT 2D OCCAM inversion, refType=null, unstructuredReference=He Meixing. A study on CSAMT 2D OCCAM inversion[D]. Wuhan: China University of Geosciences, 2009., articleTitle=null, refAbstract=null), Reference(id=1245407876089492331, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=26, authorNames=王堃鹏, journalName=张量CSAMT三维主轴各向异性正反演研究, refType=null, unstructuredReference=王堃鹏. 张量CSAMT三维主轴各向异性正反演研究[D]. 北京: 中国地质大学(北京), 2017., articleTitle=null, refAbstract=null), Reference(id=1245407876206932855, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2017, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=27, authorNames=Wang Kunpeng, journalName=Research on forward modeling and inversion of tensor CSAMT in 3D axial anisotropic media, refType=null, unstructuredReference=Wang Kunpeng. Research on forward modeling and inversion of tensor CSAMT in 3D axial anisotropic media[D]. Beijing: China University of Geosciences (Beijing), 2017., articleTitle=null, refAbstract=null), Reference(id=1245407876311790460, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2012, volume=55, issue=11, pageStart=3829, pageEnd=3839, url=null, language=null, rfNumber=[17], rfOrder=28, authorNames=林昌洪, 谭捍东, 舒晴, journalName=地球物理学报, refType=null, unstructuredReference=林昌洪, 谭捍东, 舒晴, . 可控源音频大地电磁三维共轭梯度反演研究[J]. 地球物理学报, 2012, 55 (11): 3829-3839., articleTitle=可控源音频大地电磁三维共轭梯度反演研究, refAbstract=null), Reference(id=1245407876446008199, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2012, volume=55, issue=11, pageStart=3829, pageEnd=3839, url=null, language=null, rfNumber=[17], rfOrder=29, authorNames=Lin Changhong, Tan Handong, Shu Qing, journalName=Chinese Journal of Geophysics, refType=null, unstructuredReference=Lin Changhong, Tan Handong, Shu Qing, et al. 3D conjugate gradient inversion of CSAMT data[J]. Chinese Journal of Geophysics, 2012, 55(11): 3829-3839., articleTitle=3D conjugate gradient inversion of CSAMT data, refAbstract=null), Reference(id=1245407876542477202, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2020, volume=35, issue=2, pageStart=642, pageEnd=655, url=null, language=null, rfNumber=[18], rfOrder=30, authorNames=王昊, 严加永, 付光明, journalName=地球物理学进展, refType=null, unstructuredReference=王昊, 严加永, 付光明, . 深度学习在地球物理中的应用现状与前景[J]. 地球物理学进展, 2020, 35(2): 642-655., articleTitle=深度学习在地球物理中的应用现状与前景, refAbstract=null), Reference(id=1245407876680889242, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2020, volume=35, issue=2, pageStart=642, pageEnd=655, url=null, language=null, rfNumber=[18], rfOrder=31, authorNames=Wang Hao, Yan Jiayong, Fu Guangming, journalName=Progress in Geophysics, refType=null, unstructuredReference=Wang Hao, Yan Jiayong, Fu Guangming, et al. Current status and application prospect of deep learning in grophysics[J]. Progress in Geophysics, 2020, 35(2): 642-655., articleTitle=Current status and application prospect of deep learning in grophysics, refAbstract=null), Reference(id=1245407876836078499, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=17, pageStart=7059, pageEnd=7065, url=null, language=null, rfNumber=[19], rfOrder=32, authorNames=林开颜, 牛程远, 张浩平, journalName=科学技术与工程, refType=null, unstructuredReference=林开颜, 牛程远, 张浩平, . 基于深度学习的景观植物颜色特征提取方法[J]. 科学技术与工程, 2024, 24(17): 7059-7065., articleTitle=基于深度学习的景观植物颜色特征提取方法, refAbstract=null), Reference(id=1245407876978684852, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=17, pageStart=7059, pageEnd=7065, url=null, language=null, rfNumber=[19], rfOrder=33, authorNames=Lin Kaiyan, Niu Chengyuan, Zhang Haoping, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Lin Kaiyan, Niu Chengyuan, Zhang Haoping, et al. A method for extracting color characteristics of landscape plants based on deep learning[J]. Science Technology and Engineering, 2024, 24(17): 7059-7065., articleTitle=A method for extracting color characteristics of landscape plants based on deep learning, refAbstract=null), Reference(id=1245407877091931070, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2020, volume=48, issue=6, pageStart=1230, pageEnd=1239, url=null, language=null, rfNumber=[20], rfOrder=34, authorNames=罗会兰, 陈鸿坤, journalName=电子学报, refType=null, unstructuredReference=罗会兰, 陈鸿坤. 基于深度学习的目标检测研究综述[J]. 电子学报, 2020, 48(6): 1230-1239., articleTitle=基于深度学习的目标检测研究综述, refAbstract=null), Reference(id=1245407877200982982, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2020, volume=48, issue=6, pageStart=1230, pageEnd=1239, url=null, language=null, rfNumber=[20], rfOrder=35, authorNames=Luo Huilan, Chen Hongkun, journalName=Acta Electronica Sinica, refType=null, unstructuredReference=Luo Huilan, Chen Hongkun. Survey of object detection based on deep learning[J]. Acta Electronica Sinica, 2020, 48(6): 1230-1239., articleTitle=Survey of object detection based on deep learning, refAbstract=null), Reference(id=1245407877347783631, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=60, issue=5, pageStart=30, pageEnd=46, url=null, language=null, rfNumber=[21], rfOrder=36, authorNames=苏晨阳, 武文红, 牛恒茂, journalName=计算机工程与应用, refType=null, unstructuredReference=苏晨阳, 武文红, 牛恒茂, . 深度学习的工人多种不安全行为识别方法综述[J]. 计算机工程与应用, 2024, 60(5): 30-46., articleTitle=深度学习的工人多种不安全行为识别方法综述, refAbstract=null), Reference(id=1245407877477807063, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=60, issue=5, pageStart=30, pageEnd=46, url=null, language=null, rfNumber=[21], rfOrder=37, authorNames=Shu Chenyang, Wu Wenhong, Niu Hengmao, journalName=Computer Engineering and Applications, refType=null, unstructuredReference=Shu Chenyang, Wu Wenhong, Niu Hengmao, et al. Review of deep learning approaches for recognizing multiple unsafe behaviors in workers[J]. Computer Engineering and Applications, 2024, 60(5): 30-46., articleTitle=Review of deep learning approaches for recognizing multiple unsafe behaviors in workers, refAbstract=null), Reference(id=1245407877603636192, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=22, pageStart=9203, pageEnd=9211, url=null, language=null, rfNumber=[22], rfOrder=38, authorNames=洪腾蛟, 丁凤娟, 王鹏, journalName=科学技术与工程, refType=null, unstructuredReference=洪腾蛟, 丁凤娟, 王鹏, . 深度学习在轴承故障诊断领域的应用研究[ J]. 科学技术与工程, 2021, 21(22): 9203-9211., articleTitle=深度学习在轴承故障诊断领域的应用研究[ J], refAbstract=null), Reference(id=1245407877708493800, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=22, pageStart=9203, pageEnd=9211, url=null, language=null, rfNumber=[22], rfOrder=39, authorNames=Hong Tengjiao, Ding Fengjuan, Wang Peng, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Hong Tengjiao, Ding Fengjuan, Wang Peng, et al. Application of deep learning in bearing fault diagnosis[J]. Science Technology and Engineering, 2021, 21(22): 9203-9211., articleTitle=Application of deep learning in bearing fault diagnosis, refAbstract=null), Reference(id=1245407877830128627, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2023, volume=28, issue=1, pageStart=3, pageEnd=36, url=null, language=null, rfNumber=[23], rfOrder=40, authorNames=唐霖峰, 张浩, 徐涵, journalName=中国图象图形学报, refType=null, unstructuredReference=唐霖峰, 张浩, 徐涵, . 基于深度学习的图像融合方法综述[J]. 中国图象图形学报, 2023, 28(1): 3-36., articleTitle=基于深度学习的图像融合方法综述, refAbstract=null), Reference(id=1245407878018872320, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2023, volume=28, issue=1, pageStart=3, pageEnd=36, url=null, language=null, rfNumber=[23], rfOrder=41, authorNames=Tang Linfeng, Zhang Hao, Xu Han, journalName=Journal of Image and Graphics, refType=null, unstructuredReference=Tang Linfeng, Zhang Hao, Xu Han, et al. Deep learning-based image fusion: a survey[J]. Journal of Image and Graphics, 2023, 28(1): 3-36., articleTitle=Deep learning-based image fusion: a survey, refAbstract=null), Reference(id=1245407878132117513, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=16, pageStart=6726, pageEnd=6733, url=null, language=null, rfNumber=[24], rfOrder=42, authorNames=金星, 周凯翔, 于海洲, journalName=科学技术与工程, refType=null, unstructuredReference=金星, 周凯翔, 于海洲, . 基于卷积神经网络与 Transformer 的电能质量扰动分类方法[J]. 科学技术与工程, 2024, 24(16): 6726-6733., articleTitle=基于卷积神经网络与 Transformer 的电能质量扰动分类方法, refAbstract=null), Reference(id=1245407878257946642, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=16, pageStart=6726, pageEnd=6733, url=null, language=null, rfNumber=[24], rfOrder=43, authorNames=Jin Xing, Zhou Kaixiang, Yu Haizhou, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Jin Xing, Zhou Kaixiang, Yu Haizhou, et al. Power quality disturbances classification method based on convolutional neural network and Transformer[J]. Science Technology and Engineering, 2024, 24(16): 6726-6733., articleTitle=Power quality disturbances classification method based on convolutional neural network and Transformer, refAbstract=null), Reference(id=1245407878387970080, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[25], rfOrder=44, authorNames=蒙祖强, 欧元汉, journalName=深度学习理论与应用, refType=null, unstructuredReference=蒙祖强, 欧元汉. 深度学习理论与应用[M]. 北京: 清华大学出版社, 2023., articleTitle=null, refAbstract=null), Reference(id=1245407878488633387, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[25], rfOrder=45, authorNames=Meng Zuqiang, Ou Yuanhan, journalName=Deep learning theory and application, refType=null, unstructuredReference=Meng Zuqiang, Ou Yuanhan. Deep learning theory and application[M]. Beijing: Tsinghua University Press, 2023., articleTitle=null, refAbstract=null), Reference(id=1245407878597685301, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[26], rfOrder=46, authorNames=邱锡鹏, journalName=神经网络与深度学习, refType=null, unstructuredReference=邱锡鹏. 神经网络与深度学习[M]. 北京: 机械工业出版社, 2020., articleTitle=null, refAbstract=null), Reference(id=1245407878715125822, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[26], rfOrder=47, authorNames=Qiu Xipeng, journalName=Neural networks and deep learning, refType=null, unstructuredReference=Qiu Xipeng. Neural networks and deep learning[M]. Beijing: China Machine Press, 2020., articleTitle=null, refAbstract=null), Reference(id=1245407878845149253, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[27], rfOrder=48, authorNames=安百州, journalName=银川盆地综合地球物理反演及地热成因机制研究, refType=null, unstructuredReference=安百州. 银川盆地综合地球物理反演及地热成因机制研究[D]. 长春: 吉林大学, 2020., articleTitle=null, refAbstract=null), Reference(id=1245407878979366992, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2020, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[27], rfOrder=49, authorNames=An Baizhou, journalName=Study on geophysical inversion and geothermal genetic mechanism in Yinchuan Basin, refType=null, unstructuredReference=An Baizhou. Study on geophysical inversion and geothermal genetic mechanism in Yinchuan Basin[D]. Changchun: Jilin University, 2020., articleTitle=null, refAbstract=null), Reference(id=1245407879101001821, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=39, issue=1, pageStart=162, pageEnd=173, url=null, language=null, rfNumber=[28], rfOrder=50, authorNames=马新月, 张冰, 徐嘉亮, journalName=地球物理学进展, refType=null, unstructuredReference=马新月, 张冰, 徐嘉亮, . 基于深度学习的速度场建模方法[J]. 地球物理学进展, 2024, 39(1): 162-173., articleTitle=基于深度学习的速度场建模方法, refAbstract=null), Reference(id=1245407879218442343, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, doi=null, pmid=null, pmcid=null, year=2024, volume=39, issue=1, pageStart=162, pageEnd=173, url=null, language=null, rfNumber=[28], rfOrder=51, authorNames=Ma Xinyue, Zhang Bing, Xu Jialiang, journalName=Progress in Geophysics, refType=null, unstructuredReference=Ma Xinyue, Zhang Bing, Xu Jialiang, et al. Velocity field modeling method based on deep learning[J]. Progress in Geophysics, 2024, 39(1): 162-173., articleTitle=Velocity field modeling method based on deep learning, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1245407864416744406, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, xref=1, ext=[AuthorCompanyExt(id=1245407864425133015, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, companyId=1245407864416744406, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Information Engineering, China University of Geosciences, Beijing, 100083, China), AuthorCompanyExt(id=1245407864433521625, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, companyId=1245407864416744406, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国地质大学(北京)信息工程学院, 北京 100083)]), AuthorCompany(id=1245407864550962151, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, xref=2, ext=[AuthorCompanyExt(id=1245407864559350761, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, companyId=1245407864550962151, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Geophysics and Information Technology, China University of Geosciences, Beijing, 100083, China), AuthorCompanyExt(id=1245407864563545067, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, companyId=1245407864550962151, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 中国地质大学(北京)地球物理与信息技术学院, 北京 100083)])], figs=[ArticleFig(id=1245407870284574963, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, label=Fig.1, caption=2D model, figureFileSmall=GTHmdD6kp8TDErUfKlmkng==, figureFileBig=2YnDoRToXpuvFApiW++UOA==, tableContent=null), ArticleFig(id=1245407870460735747, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, label=图1, caption=二维正演模型, figureFileSmall=GTHmdD6kp8TDErUfKlmkng==, figureFileBig=2YnDoRToXpuvFApiW++UOA==, tableContent=null), ArticleFig(id=1245407870762725663, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, label=Fig.2, caption=Training model diagram, figureFileSmall=6WG7UoGiX7YdnfFv4oWHRg==, figureFileBig=AImeK7ynR9Qj9TyVt8qeUg==, tableContent=null), ArticleFig(id=1245407870880166186, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, label=图2, caption=训练模型图, figureFileSmall=6WG7UoGiX7YdnfFv4oWHRg==, figureFileBig=AImeK7ynR9Qj9TyVt8qeUg==, tableContent=null), ArticleFig(id=1245407871001801014, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, label=Fig.3, caption=UNet Neural network the algorithm adopted, figureFileSmall=BBC1kZ/xb4yfd0iFmdJRgA==, figureFileBig=4Ut8roUSdJPS+d9e12ngFQ==, tableContent=null), ArticleFig(id=1245407871089881407, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, label=图3, caption=算法所采用的UNet神经网络, figureFileSmall=BBC1kZ/xb4yfd0iFmdJRgA==, figureFileBig=4Ut8roUSdJPS+d9e12ngFQ==, tableContent=null), ArticleFig(id=1245407871198933326, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, label=Fig.4, caption=Single low resistance model inversion result, figureFileSmall=NsU09T6i/xTu+Rut7BhPyQ==, figureFileBig=B/wS32LFbXIDAvVmPeT5wg==, tableContent=null), ArticleFig(id=1245407871333151068, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, label=图4, caption=单个低阻模型反演结果图, figureFileSmall=NsU09T6i/xTu+Rut7BhPyQ==, figureFileBig=B/wS32LFbXIDAvVmPeT5wg==, tableContent=null), ArticleFig(id=1245407871479951727, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, label=Fig.5, caption=Two low resistance model inversion result, figureFileSmall=4TNVysGMAwbZtUdodWMSyA==, figureFileBig=noIf6OgSeEwgCVamIXyPiA==, tableContent=null), ArticleFig(id=1245407871639335299, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, label=图5, caption=两个低阻模型反演结果图, figureFileSmall=4TNVysGMAwbZtUdodWMSyA==, figureFileBig=noIf6OgSeEwgCVamIXyPiA==, tableContent=null), ArticleFig(id=1245407871723221393, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, label=Fig.6, caption=Layered low resistance model inversion result, figureFileSmall=qy94NLcSKCaqQb4zDs3Arg==, figureFileBig=MPE29AQQk4nTNFN2tnCEtg==, tableContent=null), ArticleFig(id=1245407871840661920, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, label=图6, caption=层状低阻模型反演结果图, figureFileSmall=qy94NLcSKCaqQb4zDs3Arg==, figureFileBig=MPE29AQQk4nTNFN2tnCEtg==, tableContent=null), ArticleFig(id=1245407871966491062, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, label=Table 1, caption=

Neural network setting parameter table

, figureFileSmall=null, figureFileBig=null, tableContent=
名称 名称 数值
数据集 训练集 2 251
测试集 400
训练参数 学习率 0.001
激活函数 RELU
优化器 adam
DropPeriod 20
迭代次数 500
训练环境 CPU
), ArticleFig(id=1245407872109097418, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, label=表1, caption=

神经网络设置参数表

, figureFileSmall=null, figureFileBig=null, tableContent=
名称 名称 数值
数据集 训练集 2 251
测试集 400
训练参数 学习率 0.001
激活函数 RELU
优化器 adam
DropPeriod 20
迭代次数 500
训练环境 CPU
), ArticleFig(id=1245407872218149338, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=EN, label=Table 2, caption=

The running time of deep learning several parts and data space occam inversion

, figureFileSmall=null, figureFileBig=null, tableContent=
反演 名称 耗时/min
基于深度学
习的反演
构建数据集 2 440
训练网络 268
反演预测 0.25
传统反演 数据空间occam反演 890
), ArticleFig(id=1245407872331395563, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1245407861216490258, language=CN, label=表2, caption=

深度学习各部分用时和数据空间occam反演用时统计表

, figureFileSmall=null, figureFileBig=null, tableContent=
反演 名称 耗时/min
基于深度学
习的反演
构建数据集 2 440
训练网络 268
反演预测 0.25
传统反演 数据空间occam反演 890
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2402911, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2402911, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2402911, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2402911, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于深度学习的可控源音频大地电磁二维反演
收藏切换
PDF下载
汪茂 1 , 谭捍东 2, * , 付兴 2
科学技术与工程 | 论文·天文学、地球科学 2025,25(7): 2683-2690
收起
收藏切换
科学技术与工程 | 论文·天文学、地球科学 2025, 25(7): 2683-2690
基于深度学习的可控源音频大地电磁二维反演
全屏
汪茂1 , 谭捍东2, * , 付兴2
作者信息
  • 1 中国地质大学(北京)信息工程学院, 北京 100083
  • 2 中国地质大学(北京)地球物理与信息技术学院, 北京 100083
  • 汪茂(1982—),男,汉族,湖北天门人,博士,高级工程师。研究方向:地球物理中的机器学习和并行计算。E-mail:

通讯作者:

* 谭捍东(1966—),男,汉族,湖北松滋人,博士,教授。研究方向:电磁法勘探。E-mail:
2D Controlled Source Audio-frequency Magnetotelluric Inversion Based on Deep Learning
Mao WANG1 , Han-dong TAN2, * , Xing FU2
Affiliations
  • 1 School of Information Engineering, China University of Geosciences, Beijing, 100083, China
  • 2 School of Geophysics and Information Technology, China University of Geosciences, Beijing, 100083, China
出版时间: 2025-03-08 doi: 10.12404/j.issn.1671-1815.2402911
文章导航
收藏切换

可控源音频大地电磁(controlled-source audio-frequency magnetotellurics,CSAMT)采用人工场源,具有较强的抗干扰性,广泛应用于油气勘探、矿产普查等方面。传统的二维反演技术已发展成熟,深度学习目前在地球物理探测中有了一些研究进展,但深度学习在CSAMT反演中的研究还是空缺,因此开发基于深度学习的CSAMT二维反演算法对推进深度学习在电磁勘探中的发展非常有意义。对深度学习的卷积层、池化层、全连接层和UNet网络的特点进行了介绍;对如何构建训练集、本文所采用的UNet网络以及如何设置训练的各参数进行了阐述。将训练好的网络储存下来,在做反演计算时,将网络导入程序中,用网络对观测数据进行预测,得到反演结果。设计了多个理论模型进行反演试算,实验结果验证了算法的可靠性和有效性。对深度学习反演和数据空间OCAAM反演的计算时间进行了统计,在构建训练集和训练网络时需要较多时间,但采用训练好的网络反演的时间远低于传统反演的计算时间,具有反演速度快的特点。

深度学习  /  UNet  /  CSAMT  /  反演

Controlled-source audio-frequency magnetotellurics (CSAMT) uses artificial sources, providing strong anti-interference capabilities. It is widely used in oil exploration, mineral surveys and other areas. Traditional 2D inversion technology is mature, and deep learning has recently made some research advancements in geophysical exploration. There is still a research gap in applying deep learning to CSAMT inversion. Therefore, developing a 2D inversion algorithm for CSAMT based on deep learning is highly significant for advancing the use of deep learning in electromagnetic exploration. The characteristics of deep learning components such as convolutional layers, pooling layers, fully connected layers, and the UNet network were introduced. An explanation was provided on how to construct the training dataset, the UNet network used in this study, and how to set various training parameters. The network was saved after training. When the inversion was needed, the net was loaded and the algorithm could predict the result. Several theoretical models were designed for inversion, and the experiment results verified the reliability and effectiveness of the algorithm. The time of the deep learning inversion and the tranditional inversion was recorded. Building training set needed much time, but the time of deep learning inverison was much less than the tranditional inversion. The deep learning inversion is more efficient than the traditional inversion.

deep learning  /  UNet  /  controlled source audio-frequency magnetotelluric (CSAMT)  /  inversion
汪茂, 谭捍东, 付兴. 基于深度学习的可控源音频大地电磁二维反演. 科学技术与工程, 2025 , 25 (7) : 2683 -2690 . DOI: 10.12404/j.issn.1671-1815.2402911
Mao WANG, Han-dong TAN, Xing FU. 2D Controlled Source Audio-frequency Magnetotelluric Inversion Based on Deep Learning[J]. Science Technology and Engineering, 2025 , 25 (7) : 2683 -2690 . DOI: 10.12404/j.issn.1671-1815.2402911
可控源音频大地电磁法(controlled source audio-frequency magnetotelluric, CSAMT)最大特点就是采用了人工场源,大大增加了信号强度,从而弥补了天然场源电磁信号微弱[1-2]、不易观测等缺点,已经被广泛应用于油气勘探、矿产普查、水文环境等方面,并且发挥了巨大的作用。此外,它还具有工作效率高、高阻屏蔽作用小、水平和垂直分辨率高等优点。
近年来,计算机技术的迅速发展,可控源音频大地电磁测深传统的二维正、反演算法的研究已经趋于成熟;随着人工智能的快速发展,深度学习,神经网络技术开始应用在各领域中。Kim等[3]提出的深度学习(deep learning)的优化算法和反演中的算法类似,许多基于深度学习的地球物理反演方法也随之发展起来;张月[4]设计卷积神经网络(convolutional neural networks,CNN)神经网络实现了深度学习的磁法反演。李丽丽[5]基于粒子群优化算法实现了MT(magnetotelluric)二维反演。 王方等[6]实现了基于深度学习的大地电磁二维反演算法。刘卫强等[7]实现了基于深度学习与联合降噪的可控源音频大地电磁抗干扰数据处理方法研究。Yang等[8]提出了一种CNN方法,可以直接从叠前地震数据中反演速度模型,与全波形反演相比,它减少了许多的计算时间。Puzyrev[9]将深度学习应用于海洋频率域可控电磁数据的一维反演。深度学习在可控源音频大地电磁反演过程中的研究主要是在降噪方面,目前将神经网络技术应用在可控源音频大地电磁二维反演中的研究还是空缺。此外,传统的CSAMT二维反演也比较耗时,根据深度学习算法的特点,实现的算法可以提高反演的计算效率。在干扰较强的地区采用基于深度学习的CSAMT方法进行矿产、水的探测具有较强的抗干扰性和更好的探测效果。采用已训练好的网络进行反演的计算速度也要快于传统的反演方法。
在本项研究中,实现了基于深度学习的CSAMT二维反演算法,本文中将具体阐述所取得的进展。
在进行二维正演模拟时建立如图1所示模型,场源放置在坐标原点,构造走向方向y轴,电性参数沿构造走向方向不变,仅在x-z平面中变化。
取电磁场随时间变化的因子为eiωt,忽略位移电流影响的时候[10],二次场满足的麦克斯韦方程组为
$\nabla \times \boldsymbol{E}^{s}=-\mathrm{i} \omega \mu_{0} \boldsymbol{H}^{s}$
$\nabla \times \boldsymbol{H}^{\mathrm{s}}=\sigma \boldsymbol{E}^{\mathrm{s}}+\Delta \sigma \boldsymbol{E}$
式中:E为电场;H为磁场;ω为角频率;μ0为真空中磁导率;σ为二维电导率;Δσ为异常电导率,Δσ=σ-σ0,σ0为背景电导率;EsE的二次场;HsH的二次场。沿y方向利用傅里叶变换公式,即
$\stackrel{\wedge }{F}$(x,ky,z)=${\int }_{-\infty }^{\infty } $F(x,y,z)e-ikyydy
从频率空间域过渡到频率波数域,得
$\frac{\partial \hat{H}_{x}^{\mathrm{s}}}{\partial z}-\frac{\partial \hat{H}_{z}^{\mathrm{s}}}{\partial x}=\sigma \hat{E}_{y}^{\mathrm{s}}+\sigma \hat{E}_{y}^{\mathrm{s}}$
$\begin{aligned}\nabla \times \boldsymbol{E}^{\mathrm{s}}= & \left(\frac{\partial E_{z}^{\mathrm{s}}}{\partial y}-\frac{\partial E_{z}^{\mathrm{s}}}{\partial z}\right) \boldsymbol{i}+\left(\frac{\partial E_{x}^{\mathrm{s}}}{\partial z}-\frac{\partial E_{z}^{\mathrm{s}}}{\partial x}\right) \boldsymbol{j}+ \\& \left(\frac{\partial E_{y}^{\mathrm{s}}}{\partial x}-\frac{\partial E_{x}^{\mathrm{s}}}{\partial y}\right) \boldsymbol{k}\end{aligned}$
$\begin{aligned}\nabla \times \boldsymbol{H}^{\mathrm{s}}= & \left(\frac{\partial H_{z}^{\mathrm{s}}}{\partial y}-\frac{\partial H_{y}^{\mathrm{s}}}{\partial z}\right) \boldsymbol{i}+\left(\frac{\partial H_{x}^{\mathrm{s}}}{\partial z}-\frac{\partial H_{z}^{\mathrm{s}}}{\partial x}\right) \boldsymbol{j}+ \\& \left(\frac{\partial H_{y}^{\mathrm{s}}}{\partial x}-\frac{\partial H_{x}^{\mathrm{s}}}{\partial y}\right) \boldsymbol{k}\end{aligned}$
解方程组,得
$\stackrel{\wedge }{{{E}^{s}}_{x}}$=-$\frac{i{k}_{y}}{{k}_{e}^{2}}\frac{\stackrel{\wedge }{\partial {{E}^{s}}_{y}}}{\partial x}$-$\frac{\stackrel{\wedge }{z}}{{k}_{e}^{2}}\frac{\stackrel{\wedge }{\partial {{H}^{s}}_{y}}}{\partial z}$-$\frac{\stackrel{\wedge }{z}}{{k}_{e}^{2}}$Δ$\stackrel{\wedge }{\sigma {{E}^{p}}_{x}}$
$\stackrel{\wedge }{{{H}^{s}}_{z}}$=-$\frac{\stackrel{\wedge }{y}}{{k}_{e}^{2}}\frac{\stackrel{\wedge }{\partial {{E}^{s}}_{y}}}{\partial x}$-$\frac{i{k}_{y}}{{k}_{e}^{2}}\frac{\stackrel{\wedge }{\partial {{H}^{s}}_{y}}}{\partial z}$-$\frac{i{k}_{y}}{{k}_{e}^{2}}$Δ$\stackrel{\wedge }{\sigma {{E}^{p}}_{x}}$
$\stackrel{\wedge }{{{E}^{s}}_{z}}$=$\frac{\stackrel{\wedge }{z}}{{k}_{e}^{2}}\frac{\stackrel{\wedge }{\partial {{H}^{s}}_{y}}}{\partial x}$-$\frac{i{k}_{y}}{{k}_{e}^{2}}\frac{\stackrel{\wedge }{\partial {{E}^{s}}_{y}}}{\partial z}$-$\frac{\stackrel{\wedge }{z}}{{k}_{e}^{2}}$Δ$\stackrel{\wedge }{\sigma {{E}^{p}}_{z}}$
$\stackrel{\wedge }{{{H}^{s}}_{x}}$=-$\frac{i{k}_{y}}{{k}_{e}^{2}}\frac{\stackrel{\wedge }{\partial {{H}^{s}}_{y}}}{\partial x}$+$\frac{\stackrel{\wedge }{y}}{{k}_{e}^{2}}\frac{\stackrel{\wedge }{\partial {{E}^{s}}_{y}}}{\partial z}$+$\frac{i{k}_{y}}{{k}_{e}^{2}}$Δ$\stackrel{\wedge }{\sigma {{E}^{p}}_{z}}$
式(7)~式(10)为波数域中${\stackrel{\wedge }{E}}_{x}$${\stackrel{\wedge }{E}}_{z}$${\stackrel{\wedge }{H}}_{x}$${\stackrel{\wedge }{H}}_{z}$的二次场的计算公式;i为虚数单位; $\stackrel{\wedge }{z}$=iμ0ω为阻抗率;$\stackrel{\wedge }{y}$=σ为导纳率;${k}_{e}^{2}$=${k}_{y}^{2}$=k2,${k}_{}^{2}$=-iω${\mu }_{0}^{}$σ。将式(8)、式(10)代入式(4)得
$\begin{array}{l} \frac{\partial}{\partial x}\left(\frac{\hat{y}}{k_{\mathrm{e}}^{2}} \frac{\partial \hat{E}_{y}^{\mathrm{s}}}{\partial x}\right)+\frac{\partial}{\partial z}\left(\frac{\hat{y}}{k_{\mathrm{e}}^{2}} \frac{\partial \hat{E}_{y}^{\mathrm{s}}}{\partial z}\right)-\hat{y} \hat{E}_{y}^{\mathrm{s}}+ \\ \mathrm{i} k_{y}\left[\frac{\partial}{\partial x}\left(\frac{1}{k_{\mathrm{e}}^{2}}\right) \frac{\partial \hat{H}_{y}^{\mathrm{s}}}{\partial z}-\frac{\partial}{\partial z}\left(\frac{1}{k_{\mathrm{e}}^{2}}\right) \frac{\partial \hat{H}_{y}^{\mathrm{s}}}{\partial x}\right]= \\ \Delta \sigma \hat{E}_{x}^{\mathrm{p}}-\mathrm{i} k_{y}\left[\frac{\partial}{\partial x}\left(\frac{\Delta \sigma}{k_{\mathrm{e}}^{2}} \hat{E}_{x}^{\mathrm{p}}\right)+\frac{\partial}{\partial z}\left(\frac{\Delta \sigma}{k_{\mathrm{e}}^{2}} \hat{E}_{z}^{\mathrm{p}}\right)\right] \end{array}$
$\begin{array}{l}\frac{\partial}{\partial x}\left(\frac{\hat{z}}{k_{\mathrm{e}}^{2}} \frac{\partial \hat{H}_{y}^{\mathrm{s}}}{\partial x}\right)+\frac{\partial}{\partial z}\left(\frac{\hat{z}}{k_{\mathrm{e}}^{2}} \frac{\partial \hat{H}_{y}^{\mathrm{s}}}{\partial z}\right)-\hat{z} \hat{H}_{y}^{\mathrm{s}}+ \\\quad \mathrm{i} k_{y}\left[-\frac{\partial}{\partial x}\left(\frac{1}{k_{\mathrm{e}}^{2}}\right) \frac{\partial \hat{E}_{y}^{\mathrm{s}}}{\partial z}+-\frac{\partial}{\partial z}\left(\frac{1}{k_{\mathrm{e}}^{2}}\right) \frac{\partial \hat{E}_{y}^{\mathrm{s}}}{\partial x}\right]= \\\quad \frac{\partial}{\partial x}\left(\frac{\hat{z} \Delta \sigma}{k_{\mathrm{e}}^{2}} \hat{E}_{z}^{\mathrm{p}}\right)-\frac{\partial}{\partial z}\left(\frac{\hat{z} \Delta \sigma}{k_{\mathrm{e}}^{2}} \hat{E}_{x}^{\mathrm{p}}\right)\end{array}$
式(11)和式(12)为正演计算所要求的电磁场偏微分方程组[11]。通过解方程组可求得电磁场分量各方向的二次场。通过有限差分形成的矩阵方程为
$\left[\begin{array}{ll}{k}_{1}& {k}_{2}\\ {k}_{3}& {k}_{4}\end{array}\right]\left[\begin{array}{l}{{E}^{s}}_{y}\\ {{H}^{s}}_{y}\end{array}\right]$=$\left[\begin{array}{l}{s}_{1}\\ {s}_{2}\end{array}\right]$
式(13)中:k为系数矩阵[12-13];EsyHsy是二次场值,各项分别为
$\left\{\begin{array}{l}{k}_{1}=\frac{\stackrel{\wedge }{y}}{{k}_{e}^{2}}\left({\xi }_{i}{\xi }_{j}\frac{b}{8a}\frac{6+2{\eta }_{i}{\eta }_{j}}{3}+{\eta }_{i}{\eta }_{j}\frac{a}{8b}\frac{6+{\xi }_{i}{\xi }_{j}}{3}\right)+\\   \stackrel{\wedge }{y}\frac{ab}{64}\frac{6+2{\xi }_{i}{\xi }_{j}}{3}\frac{6+2{\eta }_{i}{\eta }_{j}}{3}\\ {k}_{2}=\frac{1}{4}\frac{iky}{{k}_{e}^{2}}({\xi }_{i}{\eta }_{j}-{\xi }_{j}{\eta }_{i})\\ {k}_{3}=-\frac{1}{4}\frac{iky}{{k}_{e}^{2}}({\xi }_{i}{\eta }_{j}-{\xi }_{j}{\eta }_{i})\\ {k}_{4}=\frac{\stackrel{\wedge }{z}}{{k}_{e}^{2}}\left({\xi }_{i}{\xi }_{j}\frac{b}{8a}\frac{6+2{\eta }_{i}{\eta }_{j}}{3}+{\eta }_{i}{\eta }_{j}\frac{a}{8b}\frac{6+{\xi }_{i}{\xi }_{j}}{3}\right)+\\   \stackrel{\wedge }{z}\frac{ab}{64}\frac{6+2{\xi }_{i}{\xi }_{j}}{3}\frac{6+2{\eta }_{i}{\eta }_{j}}{3}\end{array}\right.$
$\begin{array}{l}\left\{\begin{array}{l}s_{1}=-\frac{a b}{4} \Delta \sigma \hat{E}_{y}^{\mathrm{p}}-\frac{\mathrm{i} k y \Delta \sigma}{k_{\mathrm{e}}^{2}}\left(\frac{b \xi_{i}}{2} \hat{E}_{x}^{\mathrm{p}}+\frac{a \eta_{i}}{2} \hat{E}_{z}^{\mathrm{p}}\right) \\s_{2}=-\frac{\hat{z} \Delta \sigma}{k_{\mathrm{e}}^{2}}\left(\frac{a \eta_{i}}{2} \hat{E}_{x}^{\mathrm{p}}+\frac{b \xi_{i}}{2} \hat{E}_{z}^{\mathrm{p}}\right)\end{array}\right.\end{array}$
式中:ξiξjηiηj为单元中的变量。给出一次场的值,即可求解方程(13)。
目前多种传统的CSAMT的反演方法,均取得了较好的反演效果。数据空间OCCAM反演方法通过数学公式将模型迭代序列的计算从模型空间(M×M)转化到数据空间形式(N×N),M为网格剖分数量,N为地面测点数量和频率数量的乘积。达到减少计算量 [14-15],提高计算效率的目的。通过一系列变换,将模型空间数据转换为数据空间数据。在数据空间反演中仅需解N×N阶的方程,在实际应用中观测数据的个数要远小于模型数据的个数,所以数据空间反演中的计算量要远小于模型空间中的计算量。
有限内存拟牛顿法避免了直接计算和存储雅克比矩阵,通过一次“拟正演”来解方程,直接得到目标函数的梯度,该方法只需要储存前几次计算得到了目标函数梯度值就可以近似的逼近本次迭代的海森矩阵(或逆矩阵),提高了反演的计算效率[16]
非线性共轭梯度法避免了直接计算雅可比矩阵,而是通过解“拟正演”问题来计算一个中间值,从而提高了反演的计算效率[17]
深度学习以其强大的特征提取和表达能力主导了计算机视觉领域的发展[18-19],并在诸如图像分类,目标检测[20-22]和语义分割等视觉任务上展现了显著的性能优势[23]
卷积神经网络是一种具有局部连接、权重共享等特性的深层前馈神经网络,受生物学上感受野机制的启发,其采用局部感受野、权值共享和时空亚采样的思想,利用卷积和池化的连接方式提取输入信息的局部特征从而进行进一步的识别和分类[24]
卷积的相关研究在模式识别领域尤其是手写数字识别中取得了良好的识别效果,近年来在算力强大的硬件支持下,可计算更大的样本数据集,深层网络训练技术不断进步,使得卷积神经网络快速发展。卷积神经网络在图像领域已取得巨大成就,主要使用在计算视觉中的各种任务上。在卷积神网络中涉及到卷积层、池化层和全连接层。
(1)卷积层:卷积是卷积神经网络的核心部分,它是模拟人脑在对图像进行识别时[25-26],先通过对图像的各个局部特征进行感知的过程,这个过程叫做卷积。经过卷积运算可以提取到相应特征,使特性增强。
(2)池化层:卷积层虽然可以显著减少网络中连接的数量,但特征映射组中的神经元个数并没有显著减少。池化操作可以压缩神经网络和参数的量从而减少计算量并保留有用信息,从而减小网络过拟合的风险,池化层是利用局部相关性的原理进行下采样,包括最大池化和平均池化等。最常用的最大池化是指整个图片分割成若干个等大小的块,每块内取最大数字并舍弃其他节点,最后保持原本的图片分割成若干个等大小的块,每块内取平均值,从而保持原本的图片结构。
(3)全连接层:全连接层是指当前层每个神经元都与前一层所有神经元相连接,当前层神经元的输入是一层神经元的输出,每个连接伴随一个权值。借助这一思想,卷积神经网络的全连接层可以整合卷积层或者池化层中具有类别区分性的局部信息,且最后一层全连接层起到了分类器的作用,能够对前面层学习到的特征进一步地识别,关注最强的特征组合,划分信号并归属于某个特定的类。
(4) UNet网络:UNet网络是一种在医学图像分割领域常用的深度学习网络结构[27]。它的特点是采用对称的U形结构,包括一个Encoder部分和一个Decoder部分。
Encoder部分:负责从输入图像中提取特征。这个部分通常包含多个卷积层,每个卷积层后面可能跟随一个池化层(如最大池化)以降低特征图的维度。这些卷积层的核大小通常为3×3,步长为1,且不使用填充(padding),以保持特征图的尺寸不变。
Decoder部分:负责将上采样后的特征图与Encoder部分提取的特征进行融合,逐步还原图像的原始尺寸。上采样操作通常使用转置卷积来实现,同时可能会结合1×1卷积层来输出最终的分类结果。
UNet网络的一个关键特点是它能够对像素点进行分类,即确定图像中的每个像素点是前景还是背景。这与之前的图像分类网络如LeNet或ResNet不同,后者通常处理整幅图像的分类。UNet网络的设计思想是Encoder-Decoder结构,这种结构简单而有效。由于在医学领域样本收集可能较为困难,UNet网络的设计者还采用了图像增强的方法,以在数据集有限的情况下获得更好的精度。
UNet网络的具体实现会有所不同,在某些情况下下采样操作后可能会使用不同的上采样方法,或者在上采样过程中可能会结合其他技巧来保持特征的细节。此外,UNet网络的设计层数也可能根据具体的数据集和任务需求而有所不同。
要训练有效的卷积神经网络,需要大量的训练集,神经网络不断的迭代和学习需要大量样本。在本次构建训练集中,将地下半空间划分为63×37个网格,由于电磁的特点,两边的网格要大一些,中间的网格为20 m(高)×40 m(宽)。在训练集构建过程中,地质体模型设置了不同的形状,不同大小[28],实验选择了12种单一模型对UNet网络进行了训练,其中前8种的地质体电阻率P=10 Ω·m,地质体的尺寸分别为3×3(深度×宽度)、3×5、4×2、4×3、7×5、8×3、8×5和8×8 。对前6种模型进行了展示,如图2所示设置了28个测点,背景电阻率为100 Ω·m。对每种模型移动他的横向和纵向位置,分别生成正演结果。另外还构建了4个地质体,电阻率不是10 Ω·m的模型,尺寸分别为3×3(深度×宽度)、5×2、5×5和2×2,电阻率值分别为20 、20、30、30 Ω·m。
将上述模型每次沿横向或纵向移动一个单元格,会开形成一个新的模型,由此算法构建了2 651个单个低阻体模型,调用CSAMT正演程序,将模型数据代入程序中进行计算得到正演结果,正演结果包括视电阻率和相位。在调用正演时为了加快计算速度,采用了平台自带的并行计算功能,所使用计算机CPU为酷睿I7,每次可以计算7个正演模型。正演采用了24个频率,测点为28个。正演结果(视电阻率和相位)存储在train数组中,将理论模型的电阻率存储在Y中,然后取Y中的28×24个(横向28个,从27~54个网格;纵向24个,从5~28层)电阻率作为标签。
在数据集中取400组数据,作为网络的验证集来验证所训练的网络是否能达到预测电阻率值的效果。
本算法设计了UNet卷积神经网络来对数据集进行训练,对24×28的电阻率数据首先进行卷积计算,得到20×24的64层的数据,进行池化操作后再进行卷积计算得到6×8的128层的数据,池化操作后再进行卷积计算得到规模为1的数据,有256层,进行上采样和反卷积后得到10×12的128层数据,再进行上采样和反卷积后得到24×28的电阻率和相位的数据。如图3所示。
在设置好网络后,通过trainoptions可以设置训练过程中的各项参数,如学习率,最大训练回合数,测试数据集,验证频率,采用CPU或GPU训练。调用trainNetwork训练网络,需要传入训练集,设置好网络和训练参数。对神经网络进行训练,完成训练后将神经网络保存下来。具体神经网络的数值设置见表1
在对数据进行预测时,可以调用load函数将训练好的神经网络载入程序中,调用predict函数预测数据,然后将理论模型和预测到的电阻率值画在一个图中进行对比。
基于深度学习的CSAMT反演包含构建数据集,训练网络,反演预测3个部分,各部分耗时情况见表2。三部分耗时分别为2 440、268、0.25 min。构建训练集的过程比较耗时,训练网络的过程也用了268 min,但是网络训练出来以后,对数据进行预测仅需0.25 min,前两部分对于算法来说只需要执行一次即可,后期预测数据用时较少。而用传统的反演算法,如数据空间occam的反演算法采用5进程进行反演需要约890 min,只要进行反演都需要这么多时间。从这个实验可以看出,如得到训练好的卷积神经网络后,反演更加高效。
假设一个低阻异常棱柱体,其顶面埋深240 m,大小为240 m×160 m。背景电阻率为100 Ω·m,异常体电阻率为10 Ω·m。模型网格大小为63×37,数据采集点为地表28~55点处。24个频率为1 000、900、800、…、100、90、…、10、8、4、2、1 Hz。XZ方向剖分单元数:横向(x)63,纵向(z)37,模型如图4(a)所示。将正演结果作为观测数据,采用已训练好的深度学习网络预测的目标体如图4(b)所示,所预测目标体与理论模型的目标体位置比较一致,且电阻率值也恢复到了10 Ω·m。
假设两个低阻异常棱柱体,其顶面埋深从左至右分别为200、240 m,从左至右大小分别为240 m×120 m、200 m×120 m。背景电阻率为100 Ω·m,异常体电阻率为10 Ω·m。模型网格大小为63×37,数据采集点为地表28~55点处,24个频率同上。XZ方向剖分单元数:横向(x)63,纵向(z)37。模型如图5(a)所示。将正演结果作为观测数据,采用已训练好的深度学习网络预测的目标体如图5(b)所示,所预测目标体与理论模型的目标体位置比较一致,且电阻率值也恢复到了10 Ω·m。
假设一个层状低阻异常体,其顶面埋深240 m,厚度为120 m。背景电阻率为100 Ω·m,异常体电阻率为10 Ω·m。模型网格和数据采集点同上,24个频率同上。XZ方向剖分网格单元数:横向(x)63,纵向(z)37。模型如图6(a)所示。采用已训练好的深度学习网络预测的目标体如图6(b)所示,所预测目标体与理论模型的层状目标体比较一致,且电阻率值也恢复到了10 Ω·m。
本文中采用UNet神经网络,设计2 000余个模型,并使用正演结果对神经网络进行训练,采用训练好的网络可对观测数据进行预测,从而得到反演结果。采用单个模型,多个模型及层状模型的合成数据进行反演试算,可较好的恢复目标的形态和电阻率值。对深度学习反演和数据空间OCAAM反演的计算时间进行了统计,在构建训练集和训练网络时需要较多时间,但采用训练好的网络反演的时间远低于传统反演的计算时间。
综上所述,得出如下结论。
(1)实现了基于深度学习的可控音频大地电磁的二维反演算法。
(2)设计了多个理论模型进行反演试算,验证了深度学习在CSAMT反演中的可行性和有效性。
(3)将深度学习的CSAMT二维反演用时与传统的数据空间OCCAM反演用时进行统计和比较。深度学习算法的运行时间远低于传统反演的计算时间,具有反演速度快的特点。
本论文所做的工作为今后深度学习在电磁反演中的进一步研究奠定了基础。
参考文献 引证文献
排序方式:
[1]
底青云, Martyn Unsworth, 王妙月. 复杂介质有限元法2.5维可控源音频大地电磁法数值模拟[J]. 地球物理学报, 2004, 47(4): 723-730.
Di Qingyun, Martyn Unsworth, Wang Miaoyue. Controlled source audio-frequency magnetotelluric 2.5 d numerical simulation of finite element method in complex medium[J]. Chinese Journal of Geophysics, 2004, 47(4): 723-730.
[2]
孟庆奎, 林品荣, 徐宝利, . 张量CSAMT一维数值模拟分析[J]. 物探化探计算技术, 2013, 35(4): 435-441.
Meng Qingkui, Lin Pinrong, Xu Baoli, et al. An analysis of tensor CSAMT 1D numerical simulation[J]. Computing Techniques for Geophysical and Geochemical, 2013, 35(4): 435-441.
[3]
Kim Y, Nakata N. Geophysical inversion versus machine learning in inverse problems[J]. The Leading Edge, 2018, 37(12): 894-901.
[4]
张月. 基于深度学习的磁法反演[D]. 荆州: 长江大学, 2020.
Zhang Yue. Magnetic inversion based on deep learning[D]. Jingzhou: Yangtze University, 2020.
[5]
李丽丽. 基于机器学习的大地电磁反演研究[D]. 桂林: 桂林理工大学, 2023.
Li Lili. Research on magnetotelluric inversion based on machine learning[D]. Guilin: Guilin University of Technology, 2023.
[6]
王方, 能杰, 田慧潇, . 基于深度学习的大地电磁二维反演方法[J]. 地质科技通报, 2024, 43(2): 344-354.
Wang Fang, Xiong Jie, Tian Huixiao, et al. 2D magnetotelluric inversion method based on deep learning[J]. Bulletin of Geological Science and Technology, 2024, 43(2): 344-354.
[7]
刘卫强, 林品荣, 陈儒军, . 基于深度学习与联合降噪的可控源音频大地电磁抗干扰数据处理方法研究[J]. 地球物理学进展, 2024, 39 (4): 1457-1473.
Liu Weiqiang, Lin Pinrong, Chen Rujun, et al. Anti-interference processing for CSAMT based on deep learning and joint denoising[J]. Progress in Geophysics, 2024, 39 (4): 1457-1473.
[8]
Yang F S, Ma J W. Deep-learning inversion: a next-generation seismic velocity model building method[J]. Geophysics, 2019, 84 (4): 583-599.
[9]
Puzyrev V. Deep learning electromagnetic inversion with convolutional neural networks[J]. Geophysical Journal International, 2019, 218 (2): 817-832.
[10]
纳比吉安. 电磁法[M]. 北京: 地质出版社, 1992.
Nabi Gian. Electromagnetic method[M]. Beijing: Geological Publishing House, 1992.
[11]
Key K. 1D inversion of multicomponent, multifrequency marine CSEM data: methodogy and synthetic studies for resolving thin resistive layers[J]. Geophysics, 2009, 74(2): 9-20.
[12]
雷达, 张国鸿, 黄高元, . 张量可控源音频大地电磁法的应用实例[J]. 工程地球物理学报, 2014, 11(3): 286-294.
Lei Da, Zhang Guohong, Huang Gaoyuan, et al. Application examples of tensor CSAMT[J]. Chinese Journal of Engineering Geophysics, 2014, 11(3): 286-294.
[13]
桂兵. 可控源音频大地电磁法张量数据二维反演研究[D]. 北京: 中国地质大学(北京), 2014.
Gui Bing. Research of 2D CSAMT tensor data inversion[D]. Beijing: China University of Geosciences (Beijing), 2014.
[14]
李晋津. 可控源音频大地电磁法二维OCCAM反演研究[D]. 北京: 中国地质大学(北京), 2013.
Li Jinjin. A research on OCCAM inversion of 2D CSAMT[D]. Beijing: China University of Geosciences (Beijing), 2013.
[15]
何梅兴. 可控源音频大地电磁二维OCCAM反演研究[D]. 武汉: 中国地质大学, 2009.
He Meixing. A study on CSAMT 2D OCCAM inversion[D]. Wuhan: China University of Geosciences, 2009.
[16]
王堃鹏. 张量CSAMT三维主轴各向异性正反演研究[D]. 北京: 中国地质大学(北京), 2017.
Wang Kunpeng. Research on forward modeling and inversion of tensor CSAMT in 3D axial anisotropic media[D]. Beijing: China University of Geosciences (Beijing), 2017.
[17]
林昌洪, 谭捍东, 舒晴, . 可控源音频大地电磁三维共轭梯度反演研究[J]. 地球物理学报, 2012, 55 (11): 3829-3839.
Lin Changhong, Tan Handong, Shu Qing, et al. 3D conjugate gradient inversion of CSAMT data[J]. Chinese Journal of Geophysics, 2012, 55(11): 3829-3839.
[18]
王昊, 严加永, 付光明, . 深度学习在地球物理中的应用现状与前景[J]. 地球物理学进展, 2020, 35(2): 642-655.
Wang Hao, Yan Jiayong, Fu Guangming, et al. Current status and application prospect of deep learning in grophysics[J]. Progress in Geophysics, 2020, 35(2): 642-655.
[19]
林开颜, 牛程远, 张浩平, . 基于深度学习的景观植物颜色特征提取方法[J]. 科学技术与工程, 2024, 24(17): 7059-7065.
Lin Kaiyan, Niu Chengyuan, Zhang Haoping, et al. A method for extracting color characteristics of landscape plants based on deep learning[J]. Science Technology and Engineering, 2024, 24(17): 7059-7065.
[20]
罗会兰, 陈鸿坤. 基于深度学习的目标检测研究综述[J]. 电子学报, 2020, 48(6): 1230-1239.
Luo Huilan, Chen Hongkun. Survey of object detection based on deep learning[J]. Acta Electronica Sinica, 2020, 48(6): 1230-1239.
[21]
苏晨阳, 武文红, 牛恒茂, . 深度学习的工人多种不安全行为识别方法综述[J]. 计算机工程与应用, 2024, 60(5): 30-46.
Shu Chenyang, Wu Wenhong, Niu Hengmao, et al. Review of deep learning approaches for recognizing multiple unsafe behaviors in workers[J]. Computer Engineering and Applications, 2024, 60(5): 30-46.
[22]
洪腾蛟, 丁凤娟, 王鹏, . 深度学习在轴承故障诊断领域的应用研究[ J]. 科学技术与工程, 2021, 21(22): 9203-9211.
Hong Tengjiao, Ding Fengjuan, Wang Peng, et al. Application of deep learning in bearing fault diagnosis[J]. Science Technology and Engineering, 2021, 21(22): 9203-9211.
[23]
唐霖峰, 张浩, 徐涵, . 基于深度学习的图像融合方法综述[J]. 中国图象图形学报, 2023, 28(1): 3-36.
Tang Linfeng, Zhang Hao, Xu Han, et al. Deep learning-based image fusion: a survey[J]. Journal of Image and Graphics, 2023, 28(1): 3-36.
[24]
金星, 周凯翔, 于海洲, . 基于卷积神经网络与 Transformer 的电能质量扰动分类方法[J]. 科学技术与工程, 2024, 24(16): 6726-6733.
Jin Xing, Zhou Kaixiang, Yu Haizhou, et al. Power quality disturbances classification method based on convolutional neural network and Transformer[J]. Science Technology and Engineering, 2024, 24(16): 6726-6733.
[25]
蒙祖强, 欧元汉. 深度学习理论与应用[M]. 北京: 清华大学出版社, 2023.
Meng Zuqiang, Ou Yuanhan. Deep learning theory and application[M]. Beijing: Tsinghua University Press, 2023.
[26]
邱锡鹏. 神经网络与深度学习[M]. 北京: 机械工业出版社, 2020.
Qiu Xipeng. Neural networks and deep learning[M]. Beijing: China Machine Press, 2020.
[27]
安百州. 银川盆地综合地球物理反演及地热成因机制研究[D]. 长春: 吉林大学, 2020.
An Baizhou. Study on geophysical inversion and geothermal genetic mechanism in Yinchuan Basin[D]. Changchun: Jilin University, 2020.
[28]
马新月, 张冰, 徐嘉亮, . 基于深度学习的速度场建模方法[J]. 地球物理学进展, 2024, 39(1): 162-173.
Ma Xinyue, Zhang Bing, Xu Jialiang, et al. Velocity field modeling method based on deep learning[J]. Progress in Geophysics, 2024, 39(1): 162-173.
2025年第25卷第7期
PDF下载
139
60
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2402911
  • 接收时间:2024-04-21
  • 首发时间:2026-03-30
  • 出版时间:2025-03-08
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-04-21
  • 修回日期:2024-07-30
基金
作者信息
    1 中国地质大学(北京)信息工程学院, 北京 100083
    2 中国地质大学(北京)地球物理与信息技术学院, 北京 100083

通讯作者:

* 谭捍东(1966—),男,汉族,湖北松滋人,博士,教授。研究方向:电磁法勘探。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2402911
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏