Article(id=1217789889467306806, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1217789884081820362, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2406600, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1725292800000, receivedDateStr=2024-09-03, revisedDate=1744819200000, revisedDateStr=2025-04-17, acceptedDate=null, acceptedDateStr=null, onlineDate=1768273335091, onlineDateStr=2026-01-13, pubDate=1753632000000, pubDateStr=2025-07-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1768273335091, onlineIssueDateStr=2026-01-13, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1768273335091, creator=13701087609, updateTime=1768273335091, updator=13701087609, issue=Issue{id=1217789884081820362, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='21', pageStart='8761', pageEnd='9209', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1768273333807, creator=13701087609, updateTime=1768273602927, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1217791012932604619, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1217789884081820362, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1217791012932604620, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1217789884081820362, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=9102, endPage=9108, ext={EN=ArticleExt(id=1217789890738180978, articleId=1217789889467306806, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=CNN-LSTM Seepage Quantity Prediction Model of Earth-Rock Dam Based on Attention Mechanism, columnId=1156262735643005297, journalTitle=Science Technology and Engineering, columnName=Papers·Hydraulic Engineering, runingTitle=null, highlight=null, articleAbstract=

Seepage analysis is the key research content of dam safety and stability, and it is of great significance for dam disaster risk control by constructing a high-precision prediction model of seepage quantity for earth-rock dam. In order to further improve the seepage prediction capability of earth-rock dam, a prediction model combining long short-term memory neural(LSTM) networks, convolutional neural(CNN) networks, and attention mechanism (Attention) was proposed. Firstly, CNN was used to mine the deep features of the data, then the time series features of the seepage flow monitoring data was extracted through LSTM, and finally the attention mechanism to the pooling layer and the fully connected layer was added to determine the importance of different time features and assign weights. Through the application analysis of engineering examples, compared with CNN, LSTM and CNN-LSTM models, the CNN-LSTM-Attention model has better prediction effect, and its coefficient of determination R2 is as high as more than 0.98, and it can capture the spatial characteristics and temporal dependence of seepage data at the same time, which shows strong reliability and stability in the prediction of seepage flow of earth-rock dam.

, correspAuthors=Ming-dao YUAN, 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=Shi-wan LI, Ming-dao YUAN, Yun-qian XU, Shu ZHANG), CN=ArticleExt(id=1217789893284123618, articleId=1217789889467306806, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于注意力机制的CNN-LSTM土石坝渗流量预测模型, columnId=1156262735810777459, journalTitle=科学技术与工程, columnName=论文·水利工程, runingTitle=null, highlight=null, articleAbstract=

渗流分析是大坝安全与稳定的重点研究内容,通过构建高精度的土石坝渗流量预测模型对于大坝灾害风险管控具有重要意义。为了进一步提高土石坝渗流预测能力,提出了一种结合长短期记忆神经网络(long short-term memory,LSTM)、卷积神经网络(convolutional neural networks,CNN)和注意力机制(attention mechanism,Attention)的预测模型。该模型首先利用CNN挖掘数据的深层特征,然后通过LSTM提取渗流量监测数据的时间序列特征,最后将注意力机制添加到池化层和全连接层中,确定不同时间特征的重要性并分配权重。通过工程实例应用分析,与CNN、LSTM、CNN-LSTM模型相比,CNN-LSTM-Attention模型预测效果更好,其可决系数R2高达0.98以上,并且能够同时捕捉到渗流量数据的空间特征和时序依赖性,在土石坝渗流量预测中表现出了较强的可靠性与稳定性。

, correspAuthors=袁明道, authorNote=null, correspAuthorsNote=
* 袁明道(1972—),男,汉族,北京人,博士,教授级高级工程师。研究方向:水利工程管理。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=qbMyvxQbjCcw25Gxi3Sogw==, magXml=gGlMrlvF7EpmZC//eSWsQg==, pdfUrl=null, pdf=CZCgsKA913TaGJfFJOzkEw==, pdfFileSize=11578707, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=1Hnt7PKwaouUGQGyHQTSXA==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=plqDoN214u9aMYENw1+P2Q==, mapNumber=null, authorCompany=null, fund=null, authors=

李诗婉(1998—),女,汉族,重庆人,硕士。研究方向:大坝安全监测。E-mail:

, authorsList=李诗婉, 袁明道, 徐云乾, 张舒)}, authors=[Author(id=1217860109267555253, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=liswan0811@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1217860109452104645, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, authorId=1217860109267555253, language=EN, stringName=Shi-wan LI, firstName=Shi-wan, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Guangdong Research Institute of Water Resources and Hydropower, Guangzhou 510635, China
2 State and Local Joint Engineering Laboratory of Estuarine Hydraulic Technology, Guangzhou 510635, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1217860109603099605, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, authorId=1217860109267555253, language=CN, stringName=李诗婉, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 广东省水利水电科学研究院, 广州 510635
2 河口水利技术国家地方联合工程实验室, 广州 510635, bio={"content":"

李诗婉(1998—),女,汉族,重庆人,硕士。研究方向:大坝安全监测。E-mail:

"}, bioImg=null, bioContent=

李诗婉(1998—),女,汉族,重庆人,硕士。研究方向:大坝安全监测。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1217860108994925456, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, xref=1, ext=[AuthorCompanyExt(id=1217860109011702676, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860108994925456, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Guangdong Research Institute of Water Resources and Hydropower, Guangzhou 510635, China), AuthorCompanyExt(id=1217860109020091287, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860108994925456, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 广东省水利水电科学研究院, 广州 510635)]), AuthorCompany(id=1217860109112365986, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, xref=2, ext=[AuthorCompanyExt(id=1217860109124948901, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860109112365986, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 State and Local Joint Engineering Laboratory of Estuarine Hydraulic Technology, Guangzhou 510635, China), AuthorCompanyExt(id=1217860109133337510, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860109112365986, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 河口水利技术国家地方联合工程实验室, 广州 510635)])]), Author(id=1217860109791843301, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=gd13380025964@163.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1217860110085444603, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, authorId=1217860109791843301, language=EN, stringName=Ming-dao YUAN, firstName=Ming-dao, middleName=null, lastName=YUAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 Guangdong Research Institute of Water Resources and Hydropower, Guangzhou 510635, China
2 State and Local Joint Engineering Laboratory of Estuarine Hydraulic Technology, Guangzhou 510635, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1217860111440203784, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, authorId=1217860109791843301, language=CN, stringName=袁明道, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 广东省水利水电科学研究院, 广州 510635
2 河口水利技术国家地方联合工程实验室, 广州 510635, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1217860108994925456, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, xref=1, ext=[AuthorCompanyExt(id=1217860109011702676, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860108994925456, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Guangdong Research Institute of Water Resources and Hydropower, Guangzhou 510635, China), AuthorCompanyExt(id=1217860109020091287, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860108994925456, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 广东省水利水电科学研究院, 广州 510635)]), AuthorCompany(id=1217860109112365986, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, xref=2, ext=[AuthorCompanyExt(id=1217860109124948901, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860109112365986, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 State and Local Joint Engineering Laboratory of Estuarine Hydraulic Technology, Guangzhou 510635, China), AuthorCompanyExt(id=1217860109133337510, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860109112365986, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 河口水利技术国家地方联合工程实验室, 广州 510635)])]), Author(id=1217860111628947486, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, 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=1217860111796719666, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, authorId=1217860111628947486, language=EN, stringName=Yun-qian XU, firstName=Yun-qian, middleName=null, lastName=XU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Guangdong Research Institute of Water Resources and Hydropower, Guangzhou 510635, China
2 State and Local Joint Engineering Laboratory of Estuarine Hydraulic Technology, Guangzhou 510635, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1217860111905771584, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, authorId=1217860111628947486, language=CN, stringName=徐云乾, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 广东省水利水电科学研究院, 广州 510635
2 河口水利技术国家地方联合工程实验室, 广州 510635, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1217860108994925456, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, xref=1, ext=[AuthorCompanyExt(id=1217860109011702676, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860108994925456, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Guangdong Research Institute of Water Resources and Hydropower, Guangzhou 510635, China), AuthorCompanyExt(id=1217860109020091287, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860108994925456, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 广东省水利水电科学研究院, 广州 510635)]), AuthorCompany(id=1217860109112365986, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, xref=2, ext=[AuthorCompanyExt(id=1217860109124948901, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860109112365986, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 State and Local Joint Engineering Laboratory of Estuarine Hydraulic Technology, Guangzhou 510635, China), AuthorCompanyExt(id=1217860109133337510, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860109112365986, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 河口水利技术国家地方联合工程实验室, 广州 510635)])]), Author(id=1217860112216150098, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, orderNo=3, 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=1217860112392310884, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, authorId=1217860112216150098, language=EN, stringName=Shu ZHANG, firstName=Shu, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 Guangdong Research Institute of Water Resources and Hydropower, Guangzhou 510635, China
2 State and Local Joint Engineering Laboratory of Estuarine Hydraulic Technology, Guangzhou 510635, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1217860112597831790, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, authorId=1217860112216150098, language=CN, stringName=张舒, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 广东省水利水电科学研究院, 广州 510635
2 河口水利技术国家地方联合工程实验室, 广州 510635, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1217860108994925456, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, xref=1, ext=[AuthorCompanyExt(id=1217860109011702676, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860108994925456, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Guangdong Research Institute of Water Resources and Hydropower, Guangzhou 510635, China), AuthorCompanyExt(id=1217860109020091287, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860108994925456, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 广东省水利水电科学研究院, 广州 510635)]), AuthorCompany(id=1217860109112365986, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, xref=2, ext=[AuthorCompanyExt(id=1217860109124948901, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860109112365986, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 State and Local Joint Engineering Laboratory of Estuarine Hydraulic Technology, Guangzhou 510635, China), AuthorCompanyExt(id=1217860109133337510, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860109112365986, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 河口水利技术国家地方联合工程实验室, 广州 510635)])])], keywords=[Keyword(id=1217860112975319179, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, orderNo=1, keyword=earth-rock dam), Keyword(id=1217860113109536928, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, orderNo=2, keyword=seepage quantity), Keyword(id=1217860113252143276, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, orderNo=3, keyword=CNN-LSTM), Keyword(id=1217860113348612279, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, orderNo=4, keyword=attention mechanism), Keyword(id=1217860113449275593, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, orderNo=5, keyword=prediction), Keyword(id=1217860113545744597, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, orderNo=1, keyword=土石坝), Keyword(id=1217860113663185123, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, orderNo=2, keyword=渗流量), Keyword(id=1217860113826762992, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, orderNo=3, keyword=CNN-LSTM), Keyword(id=1217860113931620607, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, orderNo=4, keyword=注意力机制), Keyword(id=1217860114049061134, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, orderNo=5, keyword=预测)], refs=[Reference(id=1217860118998340243, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2022, volume=53, issue=3, pageStart=202, pageEnd=206, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=杨启贵, 周和清, 刘加龙, journalName=人民长江, refType=null, unstructuredReference=杨启贵, 周和清, 刘加龙. 东方山水库大坝管道型渗漏的应急抢险与除险加固[J]. 人民长江, 2022, 53(3): 202-206., articleTitle=东方山水库大坝管道型渗漏的应急抢险与除险加固, refAbstract=null), Reference(id=1217860119124169376, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2022, volume=53, issue=3, pageStart=202, pageEnd=206, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=Yang Qigui, Zhou Heqing, Liu Jialong, journalName=Yangtze River, refType=null, unstructuredReference=Yang Qigui, Zhou Heqing, Liu Jialong. Emergency rescue and reinforcement for pipeline leakage of Dongfangshan reservoir dam[J]. Yangtze River, 2022, 53(3): 202-206., articleTitle=Emergency rescue and reinforcement for pipeline leakage of Dongfangshan reservoir dam, refAbstract=null), Reference(id=1217860120046916264, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=41, issue=12, pageStart=90, pageEnd=92, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=胡孟凡, 欧斌, 张才溢, journalName=水电能源科学, refType=null, unstructuredReference=胡孟凡, 欧斌, 张才溢, 等. 基于PSO-BP模型的土石坝渗流预测研究[J]. 水电能源科学, 2023, 41(12): 90-92, 89., articleTitle=基于PSO-BP模型的土石坝渗流预测研究, refAbstract=null), Reference(id=1217860120223077042, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=41, issue=12, pageStart=90, pageEnd=92, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=Hu Mengfan, Ou Bin, Zhang Caiyi, journalName=Water Resources and Power, refType=null, unstructuredReference=Hu Mengfan, Ou Bin, Zhang Caiyi, et al. Analysis of causes of waterlogging based on real-time monitoring data[J]. Water Resources and Power, 2023, 41(12): 90-92, 89., articleTitle=Analysis of causes of waterlogging based on real-time monitoring data, refAbstract=null), Reference(id=1217860120416015041, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=242, issue=null, pageStart=122798, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=Liu B, Cen W, Zheng C, journalName=Expert Systems with Applications, refType=null, unstructuredReference=Liu B, Cen W, Zheng C, et al. A combined optimization prediction model for earth-rock dam seepage pressure using multi-machine learning fusion with decomposition data-driven[J]. Expert Systems with Applications, 2024, 242: 122798., articleTitle=A combined optimization prediction model for earth-rock dam seepage pressure using multi-machine learning fusion with decomposition data-driven, refAbstract=null), Reference(id=1217860120579592908, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=34, issue=5, pageStart=198, pageEnd=206, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=赵明哲, 樊牧, 沈立锋, journalName=水资源与水工程学报, refType=null, unstructuredReference=赵明哲, 樊牧, 沈立锋, 等. 基于GAMLSS-GLO及随机森林算法的土石坝渗流监测模型[J]. 水资源与水工程学报, 2023, 34(5): 198-206, 214., articleTitle=基于GAMLSS-GLO及随机森林算法的土石坝渗流监测模型, refAbstract=null), Reference(id=1217860120692839127, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=34, issue=5, pageStart=198, pageEnd=206, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Zhao Mingzhe, Fan Mu, Shen Lifeng, journalName=Journal of Water Resources & Water Engineering, refType=null, unstructuredReference=Zhao Mingzhe, Fan Mu, Shen Lifeng, et al. Seepage monitoring model of earth-rock dam based on GAMLSS-GLO and random forest algorithm[J]. Journal of Water Resources & Water Engineering, 2023, 34(5): 198-206, 214., articleTitle=Seepage monitoring model of earth-rock dam based on GAMLSS-GLO and random forest algorithm, refAbstract=null), Reference(id=1217860120843834081, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=11, pageStart=78, pageEnd=91, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=王晓玲, 朱开渲, 余红玲, journalName=水力发电学报, refType=null, unstructuredReference=王晓玲, 朱开渲, 余红玲, 等. 考虑时空相关性的大坝渗压组合深度学习预测模型[J]. 水力发电学报, 2023, 42(11): 78-91., articleTitle=考虑时空相关性的大坝渗压组合深度学习预测模型, refAbstract=null), Reference(id=1217860120990634733, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=11, pageStart=78, pageEnd=91, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=Wang Xiaoling, Zhu Kaixuan, Yu Hongling, journalName=Journal of Hydroelectric Engineering, refType=null, unstructuredReference=Wang Xiaoling, Zhu Kaixuan, Yu Hongling, et al. Combinatorial deep learning prediction model for dam seepage pressure considering spatiotemporal correlation[J]. Journal of Hydroelectric Engineering, 2023, 42(11): 78-91., articleTitle=Combinatorial deep learning prediction model for dam seepage pressure considering spatiotemporal correlation, refAbstract=null), Reference(id=1217860121175184118, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2022, volume=14, issue=16, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=Wang S, Yang B, Chen H, journalName=Water, refType=null, unstructuredReference=Wang S, Yang B, Chen H, et al. LSTM-based deformation prediction model of the embankment dam of the Danjiangkou hydropower station[J]. Water, 2022, 14(16): 2464., articleTitle=LSTM-based deformation prediction model of the embankment dam of the Danjiangkou hydropower station, refAbstract=null), Reference(id=1217860121296818946, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2020, volume=48, issue=5, pageStart=419, pageEnd=425, url=null, language=null, rfNumber=[7], rfOrder=10, authorNames=谷艳昌, 吴云星, 黄海兵, journalName=河海大学学报(自然科学版), refType=null, unstructuredReference=谷艳昌, 吴云星, 黄海兵, 等. 基于遗传算法优化支持向量机的大坝安全性态预测模型[J]. 河海大学学报(自然科学版), 2020, 48(5): 419-425., articleTitle=基于遗传算法优化支持向量机的大坝安全性态预测模型, refAbstract=null), Reference(id=1217860121410065166, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2020, volume=48, issue=5, pageStart=419, pageEnd=425, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=Gu Yanchang, Wu Yunxing, Huang Haibing, journalName=Journal of Hohai University(Natural Sciences), refType=null, unstructuredReference=Gu Yanchang, Wu Yunxing, Huang Haibing, et al. Prediction model of dam safety behavior based on genetic algorithm optimized support vector machine[J]. Journal of Hohai University(Natural Sciences), 2020, 48(5): 419-425., articleTitle=Prediction model of dam safety behavior based on genetic algorithm optimized support vector machine, refAbstract=null), Reference(id=1217860121577837341, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2019, volume=46, issue=3, pageStart=267, pageEnd=274, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=毛莺池, 王静, 陈小丽, journalName=计算机科学, refType=null, unstructuredReference=毛莺池, 王静, 陈小丽, 等. 基于特征组合与CNN的大坝缺陷识别与分类方法[J]. 计算机科学, 2019, 46(3): 267-274., articleTitle=基于特征组合与CNN的大坝缺陷识别与分类方法, refAbstract=null), Reference(id=1217860121712055083, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2019, volume=46, issue=3, pageStart=267, pageEnd=274, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=Mao Yingchi, Wang Jing, Chen Xiaoli, journalName=Computer Science, refType=null, unstructuredReference=Mao Yingchi, Wang Jing, Chen Xiaoli, et al. Dam defect recognition and classification based on feature combination and CNN[J]. Computer Science, 2019, 46(3): 267-274., articleTitle=Dam defect recognition and classification based on feature combination and CNN, refAbstract=null), Reference(id=1217860121888215862, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=232, issue=null, pageStart=120835, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=Pan J, Liu W, Liu C, journalName=Expert Systems with Applications, refType=null, unstructuredReference=Pan J, Liu W, Liu C, et al. Convolutional neural network-based spatiotemporal prediction for deformation behavior of arch dams[J]. Expert Systems with Applications, 2023, 232: 120835., articleTitle=Convolutional neural network-based spatiotemporal prediction for deformation behavior of arch dams, refAbstract=null), Reference(id=1217860122064376645, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=16, issue=3, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=Yang X, Xiang Y, Wang Y, journalName=Water, refType=null, unstructuredReference=Yang X, Xiang Y, Wang Y, et al. A dam safety state prediction and analysis method based on EMD-SSA-LSTM[J]. Water, 2024, 16(3): 395., articleTitle=A dam safety state prediction and analysis method based on EMD-SSA-LSTM, refAbstract=null), Reference(id=1217860122320229207, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2025, volume=24, issue=1, pageStart=466, pageEnd=480, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=Jiang Z, journalName=Structural Health Monitoring, refType=null, unstructuredReference=Jiang Z. Monitoring model group of seepage behavior of earth-rock dam based on the mutual information and support vector machine algorithms[J]. Structural Health Monitoring, 2025, 24(1): 466-480., articleTitle=Monitoring model group of seepage behavior of earth-rock dam based on the mutual information and support vector machine algorithms, refAbstract=null), Reference(id=1217860122588664680, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=19, issue=7, pageStart=99, pageEnd=105, url=null, language=null, rfNumber=[12], rfOrder=17, authorNames=王世杰, 肖萌, 华国威, journalName=中国安全生产科学技术, refType=null, unstructuredReference=王世杰, 肖萌, 华国威, 等. 基于CNN-GRU的混凝土大坝扬压力预测模型研究[J]. 中国安全生产科学技术, 2023, 19(7): 99-105., articleTitle=基于CNN-GRU的混凝土大坝扬压力预测模型研究, refAbstract=null), Reference(id=1217860122735465334, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=19, issue=7, pageStart=99, pageEnd=105, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=Wang Shijie, Xiao Meng, Hua Guowei, journalName=Journal of Safety Science and Technology, refType=null, unstructuredReference=Wang Shijie, Xiao Meng, Hua Guowei, et al. Research on uplift pressure prediction of concrete dam based on CNN-GRU[J]. Journal of Safety Science and Technology, 2023, 19(7): 99-105., articleTitle=Research on uplift pressure prediction of concrete dam based on CNN-GRU, refAbstract=null), Reference(id=1217860122840322944, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=26, pageStart=11401, pageEnd=11408, url=null, language=null, rfNumber=[13], rfOrder=19, authorNames=叶玉龙, 张研, 袁普龙, journalName=科学技术与工程, refType=null, unstructuredReference=叶玉龙, 张研, 袁普龙, 等. 基于优化变分模态分解的大坝变形组合预测模型: 以丰满水电站为例[J]. 科学技术与工程, 2024, 24(26): 11401-11408., articleTitle=基于优化变分模态分解的大坝变形组合预测模型: 以丰满水电站为例, refAbstract=null), Reference(id=1217860122949374858, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=26, pageStart=11401, pageEnd=11408, url=null, language=null, rfNumber=[13], rfOrder=20, authorNames=Ye Yulong, Zhang Yan, Yuan Pulong, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Ye Yulong, Zhang Yan, Yuan Pulong, et al. Combined prediction model of dam deformation based on optimal variational mode decomposition: taking Fengman hydropower station for example[J]. Science Technology and Engineering, 2024, 24(26): 11401-11408., articleTitle=Combined prediction model of dam deformation based on optimal variational mode decomposition: taking Fengman hydropower station for example, refAbstract=null), Reference(id=1217860123062621076, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2017, volume=28, issue=1, pageStart=162, pageEnd=169, url=null, language=null, rfNumber=[14], rfOrder=21, authorNames=Zhao B, Lu H, Chen S, journalName=Journal of Systems Engineering and Electronics, refType=null, unstructuredReference=Zhao B, Lu H, Chen S, et al. Convolutional neural networks for time series classification[J]. Journal of Systems Engineering and Electronics, 2017, 28(1): 162-169., articleTitle=Convolutional neural networks for time series classification, refAbstract=null), Reference(id=1217860123217810337, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=22, authorNames=文德春, journalName=基于多级小波分解交互网络的股票价格预测, refType=null, unstructuredReference=文德春. 基于多级小波分解交互网络的股票价格预测[D]. 济南: 山东大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1217860123339445167, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=23, authorNames=Wen Dechun, journalName=Stock price prediction based on multilevel wavelet decomposition interaction network, refType=null, unstructuredReference=Wen Dechun. Stock price prediction based on multilevel wavelet decomposition interaction network[D]. Jinan: Shandong University, 2023., articleTitle=null, refAbstract=null), Reference(id=1217860123440108466, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=250, pageEnd=257, url=null, language=null, rfNumber=[16], rfOrder=24, authorNames=翁鸣昊, 项兴华, 陈俊涛, journalName=中国农村水利水电, refType=null, unstructuredReference=翁鸣昊, 项兴华, 陈俊涛, 等. 基于LSTM与Transformer的大坝变形预测研究[J]. 中国农村水利水电, 2024(4): 250-257., articleTitle=基于LSTM与Transformer的大坝变形预测研究, refAbstract=null), Reference(id=1217860123565937594, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=4, pageStart=250, pageEnd=257, url=null, language=null, rfNumber=[16], rfOrder=25, authorNames=Weng Minghao, Xiang Xinghua, Chen Juntao, journalName=China Rural Water and Hydropower, refType=null, unstructuredReference=Weng Minghao, Xiang Xinghua, Chen Juntao, et al. Dam deformation prediction research based on LSTM and Transformer[J]. China Rural Water and Hydropower, 2024(4): 250-257., articleTitle=Dam deformation prediction research based on LSTM and Transformer, refAbstract=null), Reference(id=1217860123704349638, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=55, issue=3, pageStart=146, pageEnd=152, url=null, language=null, rfNumber=[17], rfOrder=26, authorNames=陈媛媛, 赵秉琨, 王慧, journalName=人民长江, refType=null, unstructuredReference=陈媛媛, 赵秉琨, 王慧, 等. 基于LSTM模型的时序InSAR地表形变预测[J]. 人民长江, 2024, 55(3): 146-152., articleTitle=基于LSTM模型的时序InSAR地表形变预测, refAbstract=null), Reference(id=1217860123830178767, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=55, issue=3, pageStart=146, pageEnd=152, url=null, language=null, rfNumber=[17], rfOrder=27, authorNames=Chen Yuanyuan, Zhao Bingkun, Wang Hui, journalName=Yangtze River, refType=null, unstructuredReference=Chen Yuanyuan, Zhao Bingkun, Wang Hui, et al. Time-series InSAR ground deformation prediction based on LSTM model[J]. Yangtze River, 2024, 55(3): 146-152., articleTitle=Time-series InSAR ground deformation prediction based on LSTM model, refAbstract=null), Reference(id=1217860123956007891, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=51, issue=S1, pageStart=1081, pageEnd=1087, url=null, language=null, rfNumber=[18], rfOrder=28, authorNames=翟运开, 乔正文, 乔岩, journalName=计算机科学, refType=null, unstructuredReference=翟运开, 乔正文, 乔岩. 基于LSTM和注意力机制的远程会诊需求预测[J]. 计算机科学, 2024, 51(S1): 1081-1087., articleTitle=基于LSTM和注意力机制的远程会诊需求预测, refAbstract=null), Reference(id=1217860125314962398, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=51, issue=S1, pageStart=1081, pageEnd=1087, url=null, language=null, rfNumber=[18], rfOrder=29, authorNames=Zhai Yunkai, Qiao Zhengwen, Qiao Yan, journalName=Computer Science, refType=null, unstructuredReference=Zhai Yunkai, Qiao Zhengwen, Qiao Yan. Forecasting teleconsultation demand based on LSTM and attention mechanism[J]. Computer Science, 2024, 51(S1): 1081-1087., articleTitle=Forecasting teleconsultation demand based on LSTM and attention mechanism, refAbstract=null), Reference(id=1217860125474345961, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2023, volume=284, issue=null, pageStart=128701, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=30, authorNames=Pan S, Yang B, Wang S, journalName=Energy, refType=null, unstructuredReference=Pan S, Yang B, Wang S, et al. Oil well production prediction based on CNN-LSTM model with self-attention mechanism[J]. Energy, 2023, 284: 128701., articleTitle=Oil well production prediction based on CNN-LSTM model with self-attention mechanism, refAbstract=null), Reference(id=1217860125587592173, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2019, volume=52, issue=11, pageStart=962, pageEnd=967, url=null, language=null, rfNumber=[20], rfOrder=31, authorNames=冯春燕, 庞琼, 谷艳昌, journalName=武汉大学学报(工学版), refType=null, unstructuredReference=冯春燕, 庞琼, 谷艳昌, 等. 基于卡尔曼滤波的土石坝渗流安全监控模型[J]. 武汉大学学报(工学版), 2019, 52(11): 962-967., articleTitle=基于卡尔曼滤波的土石坝渗流安全监控模型, refAbstract=null), Reference(id=1217860125759558649, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2019, volume=52, issue=11, pageStart=962, pageEnd=967, url=null, language=null, rfNumber=[20], rfOrder=32, authorNames=Feng Chunyan, Pang Qiong, Gu Yanchang, journalName=Engineering Journal of Wuhan University, refType=null, unstructuredReference=Feng Chunyan, Pang Qiong, Gu Yanchang, et al. Monitoring model of embankment dam seepage safety based on Kalman filtering[J]. Engineering Journal of Wuhan University, 2019, 52(11): 962-967., articleTitle=Monitoring model of embankment dam seepage safety based on Kalman filtering, refAbstract=null), Reference(id=1217860125893775360, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=46, issue=1, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[21], rfOrder=33, authorNames=傅露莹, 齐慧君, 李同春, journalName=三峡大学学报(自然科学版), refType=null, unstructuredReference=傅露莹, 齐慧君, 李同春, 等. 基于CEEMDAN和相关性分析的大坝位移预测[J]. 三峡大学学报(自然科学版), 2024, 46(1): 1-6., articleTitle=基于CEEMDAN和相关性分析的大坝位移预测, refAbstract=null), Reference(id=1217860126019604486, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, doi=null, pmid=null, pmcid=null, year=2024, volume=46, issue=1, pageStart=1, pageEnd=6, url=null, language=null, rfNumber=[21], rfOrder=34, authorNames=Fu Luying, Qi Huijun, Li Tongchun, journalName=Journal of China Three Gorges University(Natural Sciences), refType=null, unstructuredReference=Fu Luying, Qi Huijun, Li Tongchun, et al. Dam displacement prediction based on CEEMDAN and correlation analysis[J]. Journal of China Three Gorges University(Natural Sciences), 2024, 46(1): 1-6., articleTitle=Dam displacement prediction based on CEEMDAN and correlation analysis, refAbstract=null)], funds=[Fund(id=1217860118763459189, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, awardId=2024-07, language=CN, fundingSource=广东省水利科技创新项目(2024-07), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1217860108994925456, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, xref=1, ext=[AuthorCompanyExt(id=1217860109011702676, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860108994925456, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 Guangdong Research Institute of Water Resources and Hydropower, Guangzhou 510635, China), AuthorCompanyExt(id=1217860109020091287, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860108994925456, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 广东省水利水电科学研究院, 广州 510635)]), AuthorCompany(id=1217860109112365986, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, xref=2, ext=[AuthorCompanyExt(id=1217860109124948901, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860109112365986, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 State and Local Joint Engineering Laboratory of Estuarine Hydraulic Technology, Guangzhou 510635, China), AuthorCompanyExt(id=1217860109133337510, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, companyId=1217860109112365986, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 河口水利技术国家地方联合工程实验室, 广州 510635)])], figs=[ArticleFig(id=1217860114426548533, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Fig.1, caption=Structure of LSTM neural network, figureFileSmall=K0bafxPUjYTT8xg1qNpdmQ==, figureFileBig=NrxPYkoYacOsuGiWms3h8A==, tableContent=null), ArticleFig(id=1217860114548183365, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=图1, caption=LSTM神经网络结构

$\text { ⊗ }$代表点乘运算;\text { ⊕ }代表求和运算;xt-1xtxt+1分别为t-1、tt+1时刻的细胞输入;Ct-1CtCt+1分别为t-1、tt+1时刻的细胞状态;ht-1htht+1分别为t-1、tt+1时刻的隐藏层细胞输出;σ模块代表sigmoid激活函数(输出值域为[0,1]),见式(1);tanh模块代表tanh激活函数,见式(2)

, figureFileSmall=K0bafxPUjYTT8xg1qNpdmQ==, figureFileBig=NrxPYkoYacOsuGiWms3h8A==, tableContent=null), ArticleFig(id=1217860114661429590, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Fig.2, caption=CNN-LSTM-Attention model structure, figureFileSmall=+E5MSPiGkBe2fbKNqioOFw==, figureFileBig=rpRkuzdVReFQP0vXtcreUQ==, tableContent=null), ArticleFig(id=1217860115944886628, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=图2, caption=CNN-LSTM-Attention模型结构, figureFileSmall=+E5MSPiGkBe2fbKNqioOFw==, figureFileBig=rpRkuzdVReFQP0vXtcreUQ==, tableContent=null), ArticleFig(id=1217860116049744241, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Fig.3, caption=Monitoring layout of measuring weir points, figureFileSmall=A5N30sjMgaBY6WjxLWeXdA==, figureFileBig=pZrgQJ2KnsEHLXi0264T9A==, tableContent=null), ArticleFig(id=1217860116142018933, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=图3, caption=量水堰测点监测布置图, figureFileSmall=A5N30sjMgaBY6WjxLWeXdA==, figureFileBig=pZrgQJ2KnsEHLXi0264T9A==, tableContent=null), ArticleFig(id=1217860116242682239, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Fig.4, caption=Environmental quantity monitoring data, figureFileSmall=57nxWMVvbn0AT/9dCVwNhA==, figureFileBig=ufjpMGfh6re5Zl/KgBOpYw==, tableContent=null), ArticleFig(id=1217860116439814541, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=图4, caption=环境量监测数据, figureFileSmall=57nxWMVvbn0AT/9dCVwNhA==, figureFileBig=ufjpMGfh6re5Zl/KgBOpYw==, tableContent=null), ArticleFig(id=1217860116586615199, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Fig.5, caption=Measuring weir monitoring data, figureFileSmall=76ju5X16azsp7DfVa4vLyA==, figureFileBig=qIpZ6YHn4wc5P8BfcfeEZA==, tableContent=null), ArticleFig(id=1217860116720832941, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=图5, caption=量水堰监测数据, figureFileSmall=76ju5X16azsp7DfVa4vLyA==, figureFileBig=qIpZ6YHn4wc5P8BfcfeEZA==, tableContent=null), ArticleFig(id=1217860116888605121, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Fig.6, caption=Fitting and prediction effect of CNN model at WE1 measurement point, figureFileSmall=ElfH6vwaC8jF7OHUGgqP6Q==, figureFileBig=KS9tsH1iOBkzkpFuoxlWAg==, tableContent=null), ArticleFig(id=1217860117047988692, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=图6, caption=WE1测点CNN模型拟合预测效果, figureFileSmall=ElfH6vwaC8jF7OHUGgqP6Q==, figureFileBig=KS9tsH1iOBkzkpFuoxlWAg==, tableContent=null), ArticleFig(id=1217860117161234911, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Fig.7, caption=Fitting and prediction effect of LSTM model at WE1 measurement point, figureFileSmall=p9saSE7o6x19k2mAsOo97Q==, figureFileBig=EOwGf/sNx6wdVXxmH/JY2w==, tableContent=null), ArticleFig(id=1217860117312229869, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=图7, caption=WE1测点LSTM模型拟合预测效果, figureFileSmall=p9saSE7o6x19k2mAsOo97Q==, figureFileBig=EOwGf/sNx6wdVXxmH/JY2w==, tableContent=null), ArticleFig(id=1217860117412893175, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Fig.8, caption=Fitting and prediction effect of CNN-LSTM model at WE1 measurement point, figureFileSmall=iRX7a5vosRHZDysTBZrCXg==, figureFileBig=P+eOObJY7kwm7X/8gnCB4A==, tableContent=null), ArticleFig(id=1217860117580665358, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=图8, caption=WE1测点CNN-LSTM模型拟合预测效果, figureFileSmall=iRX7a5vosRHZDysTBZrCXg==, figureFileBig=P+eOObJY7kwm7X/8gnCB4A==, tableContent=null), ArticleFig(id=1217860117702300186, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Fig.9, caption=Fitting and prediction effect of CNN-LSTM-Attention model at WE1 measurement point, figureFileSmall=fdimQDBShwypq1HarcfpIg==, figureFileBig=7AwCpxwIJRbJi14WqQc/8g==, tableContent=null), ArticleFig(id=1217860117798769192, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=图9, caption=WE1测点CNN-LSTM-Attention模型拟合预测效果, figureFileSmall=fdimQDBShwypq1HarcfpIg==, figureFileBig=7AwCpxwIJRbJi14WqQc/8g==, tableContent=null), ArticleFig(id=1217860117912015414, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Fig.10, caption=Comparison of prediction ability of different models at WE5 measuring point, figureFileSmall=Kahoc49nHedZwsDbeawI4Q==, figureFileBig=WEGhf/N8Yx5k++uuwyt5xA==, tableContent=null), ArticleFig(id=1217860118125924928, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=图10, caption=WE5测点不同模型预测能力对比, figureFileSmall=Kahoc49nHedZwsDbeawI4Q==, figureFileBig=WEGhf/N8Yx5k++uuwyt5xA==, tableContent=null), ArticleFig(id=1217860118381777486, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=EN, label=Table 1, caption=

Fitting and prediction evaluation indexes of different models

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 R2 RMSE/(L·s-1) MSE/(L·s-1)
拟合 预测 拟合 预测 拟合 预测
CNN 0.978 0.942 0.334 0.450 0.111 0.203
LSTM 0.952 0.945 0.492 0.439 0.242 0.193
CNN-LSTM 0.982 0.962 0.304 0.367 0.092 0.135
CNN-LSTM-
Attention
0.993 0.984 0.185 0.235 0.034 0.055
), ArticleFig(id=1217860118524383833, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1217789889467306806, language=CN, label=表1, caption=

各个模型拟合预测评价指标

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 R2 RMSE/(L·s-1) MSE/(L·s-1)
拟合 预测 拟合 预测 拟合 预测
CNN 0.978 0.942 0.334 0.450 0.111 0.203
LSTM 0.952 0.945 0.492 0.439 0.242 0.193
CNN-LSTM 0.982 0.962 0.304 0.367 0.092 0.135
CNN-LSTM-
Attention
0.993 0.984 0.185 0.235 0.034 0.055
)], 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.1671-1815.2406600, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2406600, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2406600, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2406600, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于注意力机制的CNN-LSTM土石坝渗流量预测模型
收藏切换
PDF下载
李诗婉 1, 2 , 袁明道 1, 2, * , 徐云乾 1, 2 , 张舒 1, 2
科学技术与工程 | 论文·水利工程 2025,25(21): 9102-9108
收起
收藏切换
科学技术与工程 | 论文·水利工程 2025, 25(21): 9102-9108
基于注意力机制的CNN-LSTM土石坝渗流量预测模型
全屏
李诗婉1, 2 , 袁明道1, 2, * , 徐云乾1, 2, 张舒1, 2
作者信息
  • 1 广东省水利水电科学研究院, 广州 510635
  • 2 河口水利技术国家地方联合工程实验室, 广州 510635
  • 李诗婉(1998—),女,汉族,重庆人,硕士。研究方向:大坝安全监测。E-mail:

通讯作者:

* 袁明道(1972—),男,汉族,北京人,博士,教授级高级工程师。研究方向:水利工程管理。E-mail:
CNN-LSTM Seepage Quantity Prediction Model of Earth-Rock Dam Based on Attention Mechanism
Shi-wan LI1, 2 , Ming-dao YUAN1, 2, * , Yun-qian XU1, 2, Shu ZHANG1, 2
Affiliations
  • 1 Guangdong Research Institute of Water Resources and Hydropower, Guangzhou 510635, China
  • 2 State and Local Joint Engineering Laboratory of Estuarine Hydraulic Technology, Guangzhou 510635, China
出版时间: 2025-07-28 doi: 10.12404/j.issn.1671-1815.2406600
文章导航
收藏切换

渗流分析是大坝安全与稳定的重点研究内容,通过构建高精度的土石坝渗流量预测模型对于大坝灾害风险管控具有重要意义。为了进一步提高土石坝渗流预测能力,提出了一种结合长短期记忆神经网络(long short-term memory,LSTM)、卷积神经网络(convolutional neural networks,CNN)和注意力机制(attention mechanism,Attention)的预测模型。该模型首先利用CNN挖掘数据的深层特征,然后通过LSTM提取渗流量监测数据的时间序列特征,最后将注意力机制添加到池化层和全连接层中,确定不同时间特征的重要性并分配权重。通过工程实例应用分析,与CNN、LSTM、CNN-LSTM模型相比,CNN-LSTM-Attention模型预测效果更好,其可决系数R2高达0.98以上,并且能够同时捕捉到渗流量数据的空间特征和时序依赖性,在土石坝渗流量预测中表现出了较强的可靠性与稳定性。

土石坝  /  渗流量  /  CNN-LSTM  /  注意力机制  /  预测

Seepage analysis is the key research content of dam safety and stability, and it is of great significance for dam disaster risk control by constructing a high-precision prediction model of seepage quantity for earth-rock dam. In order to further improve the seepage prediction capability of earth-rock dam, a prediction model combining long short-term memory neural(LSTM) networks, convolutional neural(CNN) networks, and attention mechanism (Attention) was proposed. Firstly, CNN was used to mine the deep features of the data, then the time series features of the seepage flow monitoring data was extracted through LSTM, and finally the attention mechanism to the pooling layer and the fully connected layer was added to determine the importance of different time features and assign weights. Through the application analysis of engineering examples, compared with CNN, LSTM and CNN-LSTM models, the CNN-LSTM-Attention model has better prediction effect, and its coefficient of determination R2 is as high as more than 0.98, and it can capture the spatial characteristics and temporal dependence of seepage data at the same time, which shows strong reliability and stability in the prediction of seepage flow of earth-rock dam.

earth-rock dam  /  seepage quantity  /  CNN-LSTM  /  attention mechanism  /  prediction
李诗婉, 袁明道, 徐云乾, 张舒. 基于注意力机制的CNN-LSTM土石坝渗流量预测模型. 科学技术与工程, 2025 , 25 (21) : 9102 -9108 . DOI: 10.12404/j.issn.1671-1815.2406600
Shi-wan LI, Ming-dao YUAN, Yun-qian XU, Shu ZHANG. CNN-LSTM Seepage Quantity Prediction Model of Earth-Rock Dam Based on Attention Mechanism[J]. Science Technology and Engineering, 2025 , 25 (21) : 9102 -9108 . DOI: 10.12404/j.issn.1671-1815.2406600
大坝安全关乎人民生命财产安全、水资源利用效率、经济社会发展等多个方面,任何微小的潜在隐患都有可能导致大坝失事甚至是溃坝,造成巨大的损失。然而,渗漏是造成土石坝溃坝事故的重要原因之一[1]。由于土石坝材料各向异性,大坝渗流具有复杂性、随机性和时变性等特点,因此,构建渗流量预测模型是评估大坝运行安全性的重要手段[2]
传统的土石坝渗流预测主要是以物理驱动为基础,结合有限单元法构建数值模型,但该方法需要大量的工程数据,且耗费时间成本[3]。对于大坝渗流的不稳定性,影响其变化的主要因素包括水位、降雨量、温度和时间[4],构建基于数据驱动的统计模型在渗流预测中也具有较好的效果。
近年来,随着人工智能的不断发展,长短时神经网络(long short-term memory, LSTM)[5-6]、支持向量机(support vector machine, SVM)[7]、卷积神经网络(convolutional neural network, CNN)[8-9]等深度学习算法在预测模型中得到广泛应用。Yang等[10]引入Sparrow优化算法来优化LSTM神经网络计算过程,降低了模型参数选择的复杂性,验证了预测大坝安全状态方面的优越性。Jiang等[11]融合具有强大的相关性分析能力的互信息(mutual information, MI)算法和具有较强的非线性建模能力的SVM算法,提出了创新模型组的概念,有效提高了模型的精度,能够精确监测大坝的渗流状态。王世杰等[12]结合门控循环神经网络和卷积神经网络,构建的CNN-GRU模型实现了渗流形态预测。叶玉龙等[13]通过粒子群优化算法将监测量分解为周期项、趋势项和随机项分量,再采用趋势项和长短时记忆网络对各分量预测值重构加成得到最终预测值。在现有研究的基础上,为了进一步提高预测精度,同时考虑到在土石坝渗流量预测中,土石坝的渗流往往存在空间上和时间上的局部相关性,CNN能够提取这些特征并转换为适合LSTM处理的形式,LSTM能够帮助模型理解和预测这些复杂的时间序列模式。因此,现结合LSTM学习时间维度特征和CNN学习空间维度特征的优势,同时引入能够准确地捕捉到影响渗流关键因素的注意力机制,构建CNN-LSTM-Attention土石坝渗流量预测模型,并应用于实际工程,全面提升大坝安全监控水平。
卷积神经网络因其强大的特征提取特性而广泛应用于时间序列数据分析[14],其主要由输入层、卷积层、池化层、全连接层和输出层组成[15]。其中卷积层是卷积神经网络的核心,其通过滑动窗口分析数据,并使用卷积核从数据中提取特征。池化层功能是通过最大值池化或平均值池化对卷积层的输出进行下采样,减少特征维度,提高模型的鲁棒性。全连接层将每个神经元与使用前后的神经元连接起来,并计算特征的权重和偏差,以获得特征信息的输出。全连接层主要通过降低后续网络层的输入维数,减小模型的大小,加快计算效率,提高特征图的鲁棒性,防止过度拟合。
长短时记忆神经网络(LSTM)在重复的单一直链细胞结构的基础上,以特殊的遗忘门、输入门和输出门结构对信息数据的输入输出和单元状态加以控制,更好地处理了较长时间序列的问题[16],其整体结构如图 1所示。sigmoid激活函数如式(1)所示,tanh激活函数如式(2)所示。
$\sigma(x)=\frac{1}{1+\mathrm{e}^{-x}}$
$\tanh (x)=\frac{\mathrm{e}^{x}-\mathrm{e}^{-x}}{\mathrm{e}^{x}+\mathrm{e}^{-x}}$
LSTM模型的构建主要分为确定输入层输入变量、确定隐藏层神经元个数和确定输出层神经元数量3个关键步骤[17]
(1)确定输入层输入变量。输入层的神经元数量等于待处理数据中输入变量的数量,输入合适的自变量因子不仅能够有效地提高模型的训练精度,确保达到最优的预测性能,而且还能提高计算效率。
(2)确定隐藏层神经元个数。隐藏层神经元个数对LSTM模型的拟合精度有较大影响,过少的隐藏层神经元可能导致拟合效果不佳,而过多则可能降低计算效率并引发模型过拟合。
(3)确定输出层神经元数量。输出层的神经元的数量等于与每个输入关联的输出的数量,LSTM模型的输出应为预测日期的时变可靠指标。
注意力机制的关键是从大量信息中捕获出对当前任务有用的重要信息,突出影响的重要特征,减少无用特征的影响,使模型做出最优选择,提高模型的准确性,其本质是对关键信息分配更大的权重,选择丢弃不重要的信息,通过差异化权重分配提高信息处理效率[18]。注意力机制先计算查询向量和键向量的相似度,然后通过softmax函数得到每个向量的注意力权重,最后利用这些权重对值向量进行加权求和,计算公式为
$\text { Attention }(\boldsymbol{Q}, \boldsymbol{K}, \boldsymbol{V})=\operatorname{softmax}\left(\frac{\boldsymbol{Q} \boldsymbol{K}^{\mathrm{T}}}{\sqrt{d_{k}}}\right) \boldsymbol{V}$
式(3)中:Q为查询向量,Query;K为键向量,Key;V为值向量,Value;dkK向量的维度;softmax为归一化函数。
CNN-LSTM-Attention模型由输入层、CNN层、LSTM层、注意力层和输出层组成[19],其模型结构如图2所示。
(1)输入层:输入监测数据集,包括渗流量时间序列的测量值,以及其他可能影响渗流的因素,如水位、降雨量、时间等环境量。
(2)CNN层:从输入的时间序列数据中提取要素,捕获时间序列中的模式和特征,例如长期趋势或周期性变化。
(3)LSTM层:该层的任务是从卷积神经网络中提取特征,捕获时间序列数据中的动态变化和复杂的非线性关系。
(4)注意力层:帮助模型集中注意力在影响渗流量变化最显著的时间段或数据特征上,通过学习得到每个输入量的重要性权重,并在信息传递过程中更加关注重要的输入特征,从而提高预测的准确性。
(5)输出层:根据模型的设计输出渗流量的预测结果。
研究水库是一座以发电为主,兼有防洪、供水等综合利用效益的大型水库。工程拦河大坝采用心墙堆石坝,为了监测土石坝坝体渗流量,在堆石坝左右岸共布设6个量水堰,测点编号为WE1~WE6,6个量水堰运行正常,监测布置位置如图3所示。
影响大坝渗流的主要因素有上、下游水位、降雨、大坝的结构、排水系统、材料特性等[20]。考虑到数据来源的真实性、可靠性、准确性、完整性,主要选取水位、降雨、时效因子共8个影响因子,构建预测模型,其中水位、降雨监测数据如图4所示。
$\begin{aligned} Y_{W}(t)= & \left\{H_{1}(t)-H\left(t_{0}\right), H_{2}(t)-H\left(t_{0}\right),\right. \\ & H_{3}(t)-H\left(t_{0}\right), P_{1}(t), P_{2}(t), P_{3}(t), \\ & \left.\frac{t}{100}, \ln \frac{t}{100}\right\} \end{aligned}$
式(4)中:H(t0)为基准水位,通常选坝底高程;H1(t)、H2(t)、H3(t)分别为观测日当天、前5 d、前10 d的平均水位;P1(t)、P2(t)、P3(t)分别为观测日当天、前5 d、前10 d的平均降雨量;t为观测时刻距初始时刻的时间,d。
量水堰的自动化采集时间为每天06∶00,采用2019年1月1日—2023年12月31日的监测数据,共计1 826条数据,各测点监测数据历时过程线如图5所示。该土石坝渗流量具有周期变化规律,与库水位呈明显的正相关。总体上大坝左岸渗流量明显大于右岸,其中WE1、WE2、WE5、WE6监测数据与水位的相关性强,而WE3、WE4监测数据与水位的相关性相对较弱,这与其高程灌浆平洞监测区域基本位于两岸的相对抗水层有关。两岸渗流量主要集中在中高部位,在低水位时段经常出现量水堰无水流的情况,673~731 m高程区域渗流量最小。总体上监测数据变化符合土石坝渗流一般规律,数据可靠、无异常。
为了验证CNN-LSTM-Attention模型预测的准确性,按0.8∶0.2的比例选取2019年1月1日—2023年1月1日的数据作为训练集,2023年1月2日—12月31日的数据作为验证集。
CNN-LSTM-Attention模型参数设置中,CNN层设置了2个卷积层和1个最大池化层,卷积核数为64,卷积核大小为3,并采用Relu函数为卷积层的激活函数。LSTM层配置足够的隐藏单元和堆叠层数以增强模型的表达能力,同时为了防止出现过度拟合,引入dropout层,并设置为0.25。优化函数选择Adam算法,最大迭代次数取150,初始学习率设置为0.01。
为了综合评估模型的精度,引入平均绝对误差(mean absolute error,MAE)、均方根误差(root mean squared error,RMSE)、可决系数(R2)3个指标作为模型的评价标准[21],MAE、RMSE的值越接近0,R2的值越接近1,模型的预测精度越高,具体的计算公式如下。
$\text { MAE }=\frac{1}{m} \sum_{i=1}^{m}\left|y_{i}-\hat{y}_{i}\right|$
$\text { RMSE }=\sqrt{\frac{1}{m} \sum_{i=1}^{m}\left(y_{i}-\hat{y}_{i}\right)^{2}}$
$R^{2}=1-\frac{\sum_{i=1}^{m}\left(y_{i}-\hat{y}_{i}\right)^{2}}{\sum_{i=1}^{m}\left(y_{i}-\bar{y}\right)^{2}}$
式中:m为观测样本总数;yi为第i个实测值; y ^ i为第i个预测结果; y -为实测值的均值。
采用CNN-LSTM-Attention模型对渗流量测点WE1的监测数据进行预测,同时为了验证该模型在土石坝渗流量预测中的优越性,选择CNN、LSTM、CNN-LSTM模型作为分析的比较模型,结果如表1图 6~图9所示。可以看出,CNN-LSTM-Attention模型在拟合阶段R2为0.993,RMSE为0.185 L/s,MSE为0.034 L/s,在预测阶段R2为0.984,RMSE为0.235 L/s,MSE为0.055 L/s,3个评价指标均优于CNN、LSTM、CNN-LSTM模型,其在拟合和预测期间均表现出了较高的准确性。整体上来说,CNN-LSTM-Attention模型拟合预测效果>CNN-LSTM模型拟合预测效果>LSTM模型拟合预测效果≈CNN模型拟合预测效果。单一的LSTM模型无法准确及时地跟随渗流量真实值的变化进行调整,单一的CNN模型学习到的特征对应的区域有限。CNN-LSTM模型比单个CNN或LSTM模型更精确,其结合了处理时间序列数据分析的深度学习技术和分析空间特征的卷积神经网络,较好地预测了土石坝的渗流量。但是引入了注意力机制的CNN-LSTM模型的全局捕获能力得到了进一步提高,能够从大量显著影响预测结果的特征中获得对预测结果具有本质影响的关键特征,具有良好的鲁棒性和良好的泛化能力,预测曲线能够更好地响应真实曲线的变化趋势。
为了进一步验证模型的预测能力,选取WE5测点2019年1月1日—2023年12月31日的监测数据,分别按训练集∶预测集为0.9∶0.1、0.8∶0.2、0.7∶0.3、0.6∶0.4、0.5∶0.5的比例输入CNN、LSTM、CNN-LSTM、CNN-LSTM-Attention模型进行计算,计算结果如图10所示。
通过对比分析发现,随着训练集与预测集的比值减小,4种模型的预测能力均受到不同程度的影响。其中单一的CNN和LSTM模型受训练集、预测集长度的影响较大,当训练集占比从90%降至50%时,预测模型的R2降低15%左右。引入了注意力机制的CNN-LSTM模型预测精度虽然也受训练集占比影响,但其影响程度相对较小,当训练集占比从90%降至50%时,预测模型的R2依然保持在0.9以上,并且在相同训练集占比中具有最小的预测误差。因此,进一步说明CNN-LSTM-ATTENTION模型的预测能力强于CNN、LSTM、CNN-LSTM模型。
提出了一种基于注意力机制的CNN-LSTM土石坝渗流量预测模型,并将其应用于砾石土心墙堆石坝,结果如下。
(1)渗流量预测是一个典型的时间序列问题,CNN-LSTM-Attention模型通过输入水位、降雨、时效因子,结合处理时间序列数据分析的深度学习技术和分析空间特征的卷积神经网络,克服因素之间存在复杂的非线性关系,并引入注意力机制实现权重分配,在土石坝渗流量预测中表现出了较好的预测效果,其可决系数R2高达0.98以上。
(2)通过对比4种模型CNN、LSTM、CNN-LSTM、CNN-LSTM-Attention的预测效果和预测能力,CNN-LSTM-Attention模型预测精度最高,明显优于其他模型,并且预测值与实际值的变化相匹配,展示了其在捕获数据固有的时空复杂性方面的卓越能力,进一步证明了其在土石坝渗流量预测方面的准确性及鲁棒性。
  • 广东省水利科技创新项目(2024-07)
参考文献 引证文献
排序方式:
[1]
杨启贵, 周和清, 刘加龙. 东方山水库大坝管道型渗漏的应急抢险与除险加固[J]. 人民长江, 2022, 53(3): 202-206.
Yang Qigui, Zhou Heqing, Liu Jialong. Emergency rescue and reinforcement for pipeline leakage of Dongfangshan reservoir dam[J]. Yangtze River, 2022, 53(3): 202-206.
[2]
胡孟凡, 欧斌, 张才溢, 等. 基于PSO-BP模型的土石坝渗流预测研究[J]. 水电能源科学, 2023, 41(12): 90-92, 89.
Hu Mengfan, Ou Bin, Zhang Caiyi, et al. Analysis of causes of waterlogging based on real-time monitoring data[J]. Water Resources and Power, 2023, 41(12): 90-92, 89.
[3]
Liu B, Cen W, Zheng C, et al. A combined optimization prediction model for earth-rock dam seepage pressure using multi-machine learning fusion with decomposition data-driven[J]. Expert Systems with Applications, 2024, 242: 122798.
[4]
赵明哲, 樊牧, 沈立锋, 等. 基于GAMLSS-GLO及随机森林算法的土石坝渗流监测模型[J]. 水资源与水工程学报, 2023, 34(5): 198-206, 214.
Zhao Mingzhe, Fan Mu, Shen Lifeng, et al. Seepage monitoring model of earth-rock dam based on GAMLSS-GLO and random forest algorithm[J]. Journal of Water Resources & Water Engineering, 2023, 34(5): 198-206, 214.
[5]
王晓玲, 朱开渲, 余红玲, 等. 考虑时空相关性的大坝渗压组合深度学习预测模型[J]. 水力发电学报, 2023, 42(11): 78-91.
Wang Xiaoling, Zhu Kaixuan, Yu Hongling, et al. Combinatorial deep learning prediction model for dam seepage pressure considering spatiotemporal correlation[J]. Journal of Hydroelectric Engineering, 2023, 42(11): 78-91.
[6]
Wang S, Yang B, Chen H, et al. LSTM-based deformation prediction model of the embankment dam of the Danjiangkou hydropower station[J]. Water, 2022, 14(16): 2464.
[7]
谷艳昌, 吴云星, 黄海兵, 等. 基于遗传算法优化支持向量机的大坝安全性态预测模型[J]. 河海大学学报(自然科学版), 2020, 48(5): 419-425.
Gu Yanchang, Wu Yunxing, Huang Haibing, et al. Prediction model of dam safety behavior based on genetic algorithm optimized support vector machine[J]. Journal of Hohai University(Natural Sciences), 2020, 48(5): 419-425.
[8]
毛莺池, 王静, 陈小丽, 等. 基于特征组合与CNN的大坝缺陷识别与分类方法[J]. 计算机科学, 2019, 46(3): 267-274.
Mao Yingchi, Wang Jing, Chen Xiaoli, et al. Dam defect recognition and classification based on feature combination and CNN[J]. Computer Science, 2019, 46(3): 267-274.
[9]
Pan J, Liu W, Liu C, et al. Convolutional neural network-based spatiotemporal prediction for deformation behavior of arch dams[J]. Expert Systems with Applications, 2023, 232: 120835.
[10]
Yang X, Xiang Y, Wang Y, et al. A dam safety state prediction and analysis method based on EMD-SSA-LSTM[J]. Water, 2024, 16(3): 395.
[11]
Jiang Z. Monitoring model group of seepage behavior of earth-rock dam based on the mutual information and support vector machine algorithms[J]. Structural Health Monitoring, 2025, 24(1): 466-480.
[12]
王世杰, 肖萌, 华国威, 等. 基于CNN-GRU的混凝土大坝扬压力预测模型研究[J]. 中国安全生产科学技术, 2023, 19(7): 99-105.
Wang Shijie, Xiao Meng, Hua Guowei, et al. Research on uplift pressure prediction of concrete dam based on CNN-GRU[J]. Journal of Safety Science and Technology, 2023, 19(7): 99-105.
[13]
叶玉龙, 张研, 袁普龙, 等. 基于优化变分模态分解的大坝变形组合预测模型: 以丰满水电站为例[J]. 科学技术与工程, 2024, 24(26): 11401-11408.
Ye Yulong, Zhang Yan, Yuan Pulong, et al. Combined prediction model of dam deformation based on optimal variational mode decomposition: taking Fengman hydropower station for example[J]. Science Technology and Engineering, 2024, 24(26): 11401-11408.
[14]
Zhao B, Lu H, Chen S, et al. Convolutional neural networks for time series classification[J]. Journal of Systems Engineering and Electronics, 2017, 28(1): 162-169.
[15]
文德春. 基于多级小波分解交互网络的股票价格预测[D]. 济南: 山东大学, 2023.
Wen Dechun. Stock price prediction based on multilevel wavelet decomposition interaction network[D]. Jinan: Shandong University, 2023.
[16]
翁鸣昊, 项兴华, 陈俊涛, 等. 基于LSTM与Transformer的大坝变形预测研究[J]. 中国农村水利水电, 2024(4): 250-257.
Weng Minghao, Xiang Xinghua, Chen Juntao, et al. Dam deformation prediction research based on LSTM and Transformer[J]. China Rural Water and Hydropower, 2024(4): 250-257.
[17]
陈媛媛, 赵秉琨, 王慧, 等. 基于LSTM模型的时序InSAR地表形变预测[J]. 人民长江, 2024, 55(3): 146-152.
Chen Yuanyuan, Zhao Bingkun, Wang Hui, et al. Time-series InSAR ground deformation prediction based on LSTM model[J]. Yangtze River, 2024, 55(3): 146-152.
[18]
翟运开, 乔正文, 乔岩. 基于LSTM和注意力机制的远程会诊需求预测[J]. 计算机科学, 2024, 51(S1): 1081-1087.
Zhai Yunkai, Qiao Zhengwen, Qiao Yan. Forecasting teleconsultation demand based on LSTM and attention mechanism[J]. Computer Science, 2024, 51(S1): 1081-1087.
[19]
Pan S, Yang B, Wang S, et al. Oil well production prediction based on CNN-LSTM model with self-attention mechanism[J]. Energy, 2023, 284: 128701.
[20]
冯春燕, 庞琼, 谷艳昌, 等. 基于卡尔曼滤波的土石坝渗流安全监控模型[J]. 武汉大学学报(工学版), 2019, 52(11): 962-967.
Feng Chunyan, Pang Qiong, Gu Yanchang, et al. Monitoring model of embankment dam seepage safety based on Kalman filtering[J]. Engineering Journal of Wuhan University, 2019, 52(11): 962-967.
[21]
傅露莹, 齐慧君, 李同春, 等. 基于CEEMDAN和相关性分析的大坝位移预测[J]. 三峡大学学报(自然科学版), 2024, 46(1): 1-6.
Fu Luying, Qi Huijun, Li Tongchun, et al. Dam displacement prediction based on CEEMDAN and correlation analysis[J]. Journal of China Three Gorges University(Natural Sciences), 2024, 46(1): 1-6.
2025年第25卷第21期
PDF下载
181
75
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2406600
  • 接收时间:2024-09-03
  • 首发时间:2026-01-13
  • 出版时间:2025-07-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-09-03
  • 修回日期:2025-04-17
基金
广东省水利科技创新项目(2024-07)
作者信息
    1 广东省水利水电科学研究院, 广州 510635
    2 河口水利技术国家地方联合工程实验室, 广州 510635

通讯作者:

* 袁明道(1972—),男,汉族,北京人,博士,教授级高级工程师。研究方向:水利工程管理。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2406600
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