Article(id=1211297842019299501, tenantId=1146029695717560320, journalId=1149651085930835976, issueId=1211297835618799960, articleNumber=null, orderNo=null, doi=10.12284/hyxb2023027, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1657036800000, receivedDateStr=2022-07-06, revisedDate=1664121600000, revisedDateStr=2022-09-26, acceptedDate=null, acceptedDateStr=null, onlineDate=1766725510363, onlineDateStr=2025-12-26, pubDate=1680192000000, pubDateStr=2023-03-31, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1766725510363, onlineIssueDateStr=2025-12-26, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1766725510363, creator=13701087609, updateTime=1766725510363, updator=13701087609, issue=Issue{id=1211297835618799960, tenantId=1146029695717560320, journalId=1149651085930835976, year='2023', volume='45', issue='4', pageStart='1', pageEnd='178', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1766725508837, creator=13701087609, updateTime=1766924525177, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1212132570683281639, tenantId=1146029695717560320, journalId=1149651085930835976, issueId=1211297835618799960, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1212132570683281640, tenantId=1146029695717560320, journalId=1149651085930835976, issueId=1211297835618799960, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=165, endPage=178, ext={EN=ArticleExt(id=1211297842359038140, articleId=1211297842019299501, tenantId=1146029695717560320, journalId=1149651085930835976, language=EN, title=Water temperature prediction in the Sansha Bay based on the integration of differential regression model and transportable long short-term memory network, columnId=1194652705852465724, journalTitle=Haiyang Xuebao, columnName=Article, runingTitle=null, highlight=null, articleAbstract=

Water temperature prediction is a key technology to ensure the production of coastal fisheries and environmental safety. The existing numerical models have high development costs with insufficient business applications. This study develops a prediction method of water temperature through integrating differential regression (DR) and transferable long short-term memory (TLSTM). Taking the water temperature of Xiamen Bay (source domain, with a large number of data) and Sansha Bay (target domain, with less data) as the research object, the DR model is established based on the data of monitoring water temperature and forecast temperature in the Sansha Bay, and the TLSTM model is established based on the long-term monitoring data of water temperature in the Xiamen Bay. The pure differential regression model, mixed differential regression model and TLSTM model are integrated into the DR-TLSTM model of Sansha Bay by using variable weight algorithm, and the performance of the model is evaluated, the results are compared with the LSTM model based on only a small amount of monitoring data in the Sansha Bay. The results show that: (1) the prediction accuracy of TLSTM model is better than that of LSTM model based on a small amount of data in the target domain; (2) the DR-TLSTM model has high prediction accuracy, and the root mean square error of prediction in the next 1−7 days is 0.13−0.77℃, and the root mean square error of prediction in the next 1−3 days is less than 0.4℃; (3) the DR-TLSTM model can effectively predict the sudden rise or fall trend of water temperature, and the root mean square error of predicting the sudden change point of water temperature is 0.29−1.09℃. Based on the DR-TLSTM model, the operational information service of water temperature early warning and forecast in the Sansha Bay is realized.

, correspAuthors=Nengwang Chen, authorNote=null, correspAuthorsNote=null, copyrightStatement=Copyright © 2023 Pratacultural Science. All rights reserved., 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=Xiaoqian Lai, Yiqi Yu, Zhongyao Liang, Huorong Chen, Nengwang Chen), CN=ArticleExt(id=1211297845748035910, articleId=1211297842019299501, tenantId=1146029695717560320, journalId=1149651085930835976, language=CN, title=基于差分回归模型和可迁移长短期记忆网络集成的三沙湾水温预测, columnId=1149698756456657529, journalTitle=海洋学报, columnName=论文, runingTitle=null, highlight=null, articleAbstract=

水温预测是保障近海渔业生产和环境安全的关键技术。现有的数值模型开发成本大,业务化应用不足。本文提出了一种集成差分回归(Differential Regression, DR)和可迁移长短期记忆网络(Transferable Long Short-Term Memory, TLSTM)的水温预测方法。以厦门湾(源域,数据多)和三沙湾(目标域,数据少)水温为研究对象,根据三沙湾在线监测水温和预报气温数据建立了DR模型,根据厦门湾长时间监测水温数据建立了TLSTM模型,采用变权算法将纯差分回归模型、混差分回归模型和TLSTM模型集成为三沙湾DR-TLSTM模型,对模型性能进行了评估,并与仅根据三沙湾少量监测数据建立的LSTM模型效果进行了对比。结果表明:(1) TLSTM模型的预测精度优于基于目标域少量数据建立的LSTM模型;(2) DR-TLSTM集成模型具有较高的预测精度,未来1~7 d预测的均方根误差为0.13~0.77℃,未来1~3 d预测的均方根误差小于0.4℃;(3) DR-TLSTM集成模型可有效预测水温骤升或骤降趋势,对水温突变点的预测均方根误差为0.29~1.09℃。基于本文建立的DR-TLSTM集成模型,实现了三沙湾渔业水域的水温预警预报业务化信息服务。

, correspAuthors=陈能汪, authorNote=null, correspAuthorsNote=
*陈能汪(1976-),男,教授,主要从事海陆界面生态环境研究。E-mail:
, copyrightStatement=版权所有©《海洋学报》编辑部 2023, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=MsCaE+1+Wk0CX31Y8kjkyw==, magXml=UuEbKYQKSiUWuIVOIBRzuQ==, pdfUrl=null, pdf=4xhYcfrS/8+y0td49hNKHA==, pdfFileSize=2983563, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=B3WwhNd9jvtnL8sn0AkxFg==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=8zk5iKwqoYpP/al7lS9vjQ==, mapNumber=null, authorCompany=null, fund=null, authors=

赖晓倩(1998-),女,福建省龙岩市人,助理工程师,主要从事大数据建模研究。E-mail:

, authorsList=赖晓倩, 余镒琦, 梁中耀, 陈火荣, 陈能汪)}, authors=[Author(id=1215313998476661569, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=xiaoqianl1_2020@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1215313998564741959, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, authorId=1215313998476661569, language=EN, stringName=Xiaoqian Lai, firstName=Xiaoqian, middleName=null, lastName=Lai, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=null, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1215313998652822350, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, authorId=1215313998476661569, language=CN, stringName=赖晓倩, firstName=晓倩, middleName=null, lastName=赖, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 厦门大学 近海海洋环境科学国家重点实验室,福建 厦门 361102, bio={"content":"

赖晓倩(1998-),女,福建省龙岩市人,助理工程师,主要从事大数据建模研究。E-mail:

"}, bioImg=null, bioContent=

赖晓倩(1998-),女,福建省龙岩市人,助理工程师,主要从事大数据建模研究。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1215313997960762137, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=1, ext=[AuthorCompanyExt(id=1215313997969150746, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313997960762137, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 厦门大学 近海海洋环境科学国家重点实验室,福建 厦门 361102)])]), Author(id=1215313998732514132, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, orderNo=1, 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=1215313998883509086, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, authorId=1215313998732514132, language=EN, stringName=Yiqi Yu, firstName=Yiqi, middleName=null, lastName=Yu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=null, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1215313998967395176, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, authorId=1215313998732514132, language=CN, stringName=余镒琦, firstName=镒琦, middleName=null, lastName=余, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 厦门大学 福建省海陆界面生态环境重点实验室,福建 厦门 361102, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1215313998111757096, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=2, ext=[AuthorCompanyExt(id=1215313998120145704, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313998111757096, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 厦门大学 福建省海陆界面生态环境重点实验室,福建 厦门 361102)])]), Author(id=1215313999042892651, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, 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=1215313999122584435, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, authorId=1215313999042892651, language=EN, stringName=Zhongyao Liang, firstName=Zhongyao, middleName=null, lastName=Liang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=null, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1215313999202276215, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, authorId=1215313999042892651, language=CN, stringName=梁中耀, firstName=中耀, middleName=null, lastName=梁, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2 厦门大学 福建省海陆界面生态环境重点实验室,福建 厦门 361102, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1215313998111757096, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=2, ext=[AuthorCompanyExt(id=1215313998120145704, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313998111757096, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 厦门大学 福建省海陆界面生态环境重点实验室,福建 厦门 361102)])]), Author(id=1215313999290356604, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, 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=1215313999386825604, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, authorId=1215313999290356604, language=EN, stringName=Huorong Chen, firstName=Huorong, middleName=null, lastName=Chen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=null, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1215313999487488906, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, authorId=1215313999290356604, language=CN, stringName=陈火荣, firstName=火荣, middleName=null, lastName=陈, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 福建省渔业资源监测中心,福建 福州 350003, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1215313998308889392, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=3, ext=[AuthorCompanyExt(id=1215313998317278001, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313998308889392, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 福建省渔业资源监测中心,福建 福州 350003)])]), Author(id=1215313999571374990, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=nwchen@xmu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1215313999680426900, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, authorId=1215313999571374990, language=EN, stringName=Nengwang Chen, firstName=Nengwang, middleName=null, lastName=Chen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=null, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1215313999764312982, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, authorId=1215313999571374990, language=CN, stringName=陈能汪, firstName=能汪, middleName=null, lastName=陈, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 厦门大学 近海海洋环境科学国家重点实验室,福建 厦门 361102
2 厦门大学 福建省海陆界面生态环境重点实验室,福建 厦门 361102, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1215313997960762137, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=1, ext=[AuthorCompanyExt(id=1215313997969150746, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313997960762137, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 厦门大学 近海海洋环境科学国家重点实验室,福建 厦门 361102)]), AuthorCompany(id=1215313998111757096, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=2, ext=[AuthorCompanyExt(id=1215313998120145704, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313998111757096, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 厦门大学 福建省海陆界面生态环境重点实验室,福建 厦门 361102)])])], keywords=[Keyword(id=1215313999932085154, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, orderNo=1, keyword=water temperature prediction), Keyword(id=1215314000041137061, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, orderNo=2, keyword=regression model), Keyword(id=1215314000112440232, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, orderNo=3, keyword=LSTM model), Keyword(id=1215314000187937708, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, orderNo=4, keyword=transfer learning), Keyword(id=1215314000263435184, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, orderNo=5, keyword=variable weight integration), Keyword(id=1215314000351515571, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, orderNo=1, keyword=水温预测), Keyword(id=1215314000464761781, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, orderNo=2, keyword=回归模型), Keyword(id=1215314000536064952, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, orderNo=3, keyword=LSTM模型), Keyword(id=1215314001819521981, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, orderNo=4, keyword=迁移学习), Keyword(id=1215314002025042883, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, orderNo=5, keyword=变权集成)], refs=[Reference(id=1215314005103661112, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=null, journalName=null, refType=null, unstructuredReference=田梓杨, 黄洪辉, 齐占会, 等. 大鹏澳海湾养殖生态环境的化学计量分析[J]. 中国水产科学, 2012, 19(5): 881−888., articleTitle=null, refAbstract=null), Reference(id=1215314006378729530, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=1, authorNames=null, journalName=null, refType=null, unstructuredReference=Tian Ziyang, Huang Honghui, Qi Zhanhui, et al. Using chemometric methods to study the mariculture eco-environment in Dapeng Cove, China[J]. Journal of Fishery Sciences of China, 2012, 19(5): 881−888., articleTitle=null, refAbstract=null), Reference(id=1215314006454227007, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=2, authorNames=null, journalName=null, refType=null, unstructuredReference=程建新, 肖佳媚, 陈明茹, 等. 兴化湾海湾生态系统退化评价[J]. 厦门大学学报(自然科学版), 2012, 51(5): 944−950., articleTitle=null, refAbstract=null), Reference(id=1215314006546501698, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=Cheng Jianxin, Xiao Jiamei, Chen Mingru, et al. Ecosystem degradation assessment of Xinghua Bay[J]. Journal of Xiamen University (Natural Science), 2012, 51(5): 944−950., articleTitle=null, refAbstract=null), Reference(id=1215314006630387783, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=Rabalais N N. Human impacts on fisheries across the land-sea interface[J]. Proceedings of the National Academy of Sciences of the United States of America, 2015, 112(26): 7892−7893., articleTitle=null, refAbstract=null), Reference(id=1215314006743633993, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=肖应凯, 刘卫国, 马云麒, 等. 鹿角珊瑚δ18O, Sr/Ca和Mg/Ca比值与海水表面温度的相关性研究——不同海水温度下的珊瑚养殖实验[J]. 中国科学: 地球科学, 2014, 57(5): 1048−1060., articleTitle=null, refAbstract=null), Reference(id=1215314006819131471, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=Xiao Yingkai, Liu Weiguo, Ma Yunqi, et al. Correlation between δ18O, Sr/Ca and Mg/Ca of coral Acropora and seawater temperature from coral culture experiments[J]. Science China Earth Sciences, 2014, 57(5): 1048−1060., articleTitle=null, refAbstract=null), Reference(id=1215314006936571986, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=胡石建, 李诗翰. 海洋热浪研究进展与展望[J]. 地球科学进展, 2022, 37(1): 51−64., articleTitle=null, refAbstract=null), Reference(id=1215314007016263768, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=Hu Shijian, Li Shihan. Progress and prospect of marine heatwave study[J]. Advances in Earth Science, 2022, 37(1): 51−64., articleTitle=null, refAbstract=null), Reference(id=1215314007108538460, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=Biswas B K, Svirezhev Y M, Bala B K. A model to predict climate-change impact on fish catch in the world oceans[J]. IEEE Transactions on Systems, Man, and Cybernetics-Part A: Systems and Humans, 2005, 35(6): 773−783., articleTitle=null, refAbstract=null), Reference(id=1215314007188230240, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=Gianelli I, Ortega L, Marín Y, et al. Evidence of ocean warming in Uruguay's fisheries landings: the mean temperature of the catch approach[J]. Marine Ecology Progress Series, 2019, 625: 115−125., articleTitle=null, refAbstract=null), Reference(id=1215314007284699235, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Wenyu, Zhou Chenghu, Shao Quanqin, et al. Remote sensing of sea surface temperature and chlorophyll-a: implications for squid fisheries in the north-west Pacific Ocean[J]. International Journal of Remote Sensing, 2010, 31(17/18): 4515−4530., articleTitle=null, refAbstract=null), Reference(id=1215314007410528358, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=Wheeland L J, Morgan M J. Age-specific shifts in Greenland halibut (Reinhardtius hippoglossoides) distribution in response to changing ocean climate[J]. ICES Journal of Marine Science, 2020, 77(1): 230−240., articleTitle=null, refAbstract=null), Reference(id=1215314007502803053, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=Zisserson B, Cook A. Impact of bottom water temperature change on the southernmost snow crab fishery in the Atlantic Ocean[J]. Fisheries Research, 2017, 195: 12−18., articleTitle=null, refAbstract=null), Reference(id=1215314007607660657, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=卢峰本, 黄滢, 周启强, 等. 海水养殖的气象风险分析及预报[J]. 气象, 2006, 32(11): 113−117., articleTitle=null, refAbstract=null), Reference(id=1215314007695741045, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=Lu Fengben, Huang Ying, Zhou Qiqiang, et al. Meteorological risk analysis and forecast to mariculture production[J]. Meteorological Monthly, 2006, 32(11): 113−117., articleTitle=null, refAbstract=null), Reference(id=1215314007779627125, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=Xie Jiang, Zhang Jiyuan, Yu Jie, et al. An adaptive scale sea surface temperature predicting method based on deep learning with attention mechanism[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 17(5): 740−744., articleTitle=null, refAbstract=null), Reference(id=1215314007888679034, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735−1780., articleTitle=null, refAbstract=null), Reference(id=1215314007989342332, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=刘建伟, 宋志妍. 循环神经网络研究综述[J]. 控制与决策, 2022, 37(11): 2753−2768., articleTitle=null, refAbstract=null), Reference(id=1215314008110977151, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Jianwei, Song Zhiyan. Overview of recurrent neural networks[J]. Control and Decision, 2022, 37(11): 2753−2768., articleTitle=null, refAbstract=null), Reference(id=1215314008199057539, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Qin, Wang Hui, Dong Junyu, et al. Prediction of sea surface temperature using long short-term memory[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(10): 1745−1749., articleTitle=null, refAbstract=null), Reference(id=1215314008299720835, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=Hou Siyun, Li Wen’gen, Liu Tianying, et al. D2CL: a dense dilated convolutional LSTM model for sea surface temperature prediction[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 12514−12523., articleTitle=null, refAbstract=null), Reference(id=1215314008492658824, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=Yang Yuting, Dong Junyu, Sun Xin, et al. A CFCC-LSTM model for sea surface temperature prediction[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15(2): 207−211., articleTitle=null, refAbstract=null), Reference(id=1215314008597516427, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=张昆. 基于深度学习的深海遥感技术研究[D]. 厦门: 厦门大学, 2019., articleTitle=null, refAbstract=null), Reference(id=1215314008970809488, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Kun. Research on deep-ocean remote sensing by deep learning[D]. Xiamen: Xiamen University, 2019., articleTitle=null, refAbstract=null), Reference(id=1215314009109221527, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=Kim M, Yang H, Kim J. Sea surface temperature and high water temperature occurrence prediction using a long short-term memory model[J]. Remote Sensing, 2020, 12(21): 3654., articleTitle=null, refAbstract=null), Reference(id=1215314009193107610, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=李嘉康, 李其杰, 赵颖, 等. 基于CEEMD-BP神经网络的海温异常预测研究[J]. 数学的实践与认识, 2017, 47(24): 163−171., articleTitle=null, refAbstract=null), Reference(id=1215314009256022175, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Jiakang, Li Qijie, Zhao Ying, et al. Sea surface temperature anomaly forecast research based on CEEMD-BP neural network[J]. Mathematics in Practice and Theory, 2017, 47(24): 163−171., articleTitle=null, refAbstract=null), Reference(id=1215314009339908259, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=董璐, 周天军. 20世纪太平洋海温变化中人为因子与自然因子贡献的模拟研究[J]. 海洋学报, 2014, 36(3): 48−60., articleTitle=null, refAbstract=null), Reference(id=1215314009436377255, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=Dong Lu, Zhou Tianjun. Contributions of natural and anthropogenic forcings to the twentieth-century Pacific sea surface temperature variability simulated by a climate system model[J]. Haiyang Xuebao, 2014, 36(3): 48−60., articleTitle=null, refAbstract=null), Reference(id=1215314009507680424, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=22, rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=田远洋, 徐显涛, 彭安帮, 等. 训练数据量对LSTM网络学习性能影响分析[J]. 水文, 2022, 42(1): 29−34, 22., articleTitle=null, refAbstract=null), Reference(id=1215314009574789291, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=22, rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=Tian Yuanyang, Xu Xiantao, Peng Anbang, et al. Effects of training data on the study performance of LSTM network[J]. Journal of China Hydrology, 2022, 42(1): 29−34, 22., articleTitle=null, refAbstract=null), Reference(id=1215314010807914670, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhuang Fuzhen, Qi Zhiyuan, Duan Keyu, et al. A comprehensive survey on transfer learning[J]. Proceedings of the IEEE, 2021, 109(1): 43−76., articleTitle=null, refAbstract=null), Reference(id=1215314010900189364, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=黎英, 宋佩华. 迁移学习在医学图像分类中的研究进展[J]. 中国图象图形学报, 2022, 27(3): 672−686., articleTitle=null, refAbstract=null), Reference(id=1215314010971492531, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=34, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Ying, Song Peihua. Review of transfer learning in medical image classification[J]. Journal of Image and Graphics, 2022, 27(3): 672−686., articleTitle=null, refAbstract=null), Reference(id=1215314011076350137, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=25, rfOrder=35, authorNames=null, journalName=null, refType=null, unstructuredReference=张晓, 李业刚, 王栋, 等. 基于迁移学习的社交评论命名实体识别[J]. 计算机应用与软件, 2022, 39(1): 143−150., articleTitle=null, refAbstract=null), Reference(id=1215314011160236219, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=25, rfOrder=36, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Xiao, Li Yegang, Wang Dong, et al. Social comment named entity recognition based on transfer learning[J]. Computer Applications and Software, 2022, 39(1): 143−150., articleTitle=null, refAbstract=null), Reference(id=1215314011294453956, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=37, authorNames=null, journalName=null, refType=null, unstructuredReference=王义刚, 宋志尧, 姜翠玲, 等. 福建省海湾数模与环境研究: 三沙湾[M]. 北京: 海洋出版社, 2009., articleTitle=null, refAbstract=null), Reference(id=1215314011399311559, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=38, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Yigang, Song Zhiyao, Jiang Cuiling, et al. Numerical Modeling and Environmental Study of Gulf in Fujian Province: Sansha Bay[M]. Beijing: China Ocean Press, 2009., articleTitle=null, refAbstract=null), Reference(id=1215314011483197643, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=39, authorNames=null, journalName=null, refType=null, unstructuredReference=厦门市地方志编纂委员会. 厦门市志[M]. 北京: 方志出版社, 2004., articleTitle=null, refAbstract=null), Reference(id=1215314011588055245, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=40, authorNames=null, journalName=null, refType=null, unstructuredReference=Xiamen Local Chronicle Compilation Committee. County Annals of Xiamen[M]. Beijing: Local Records Publishing House, 2004., articleTitle=null, refAbstract=null), Reference(id=1215314011676135632, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=41, authorNames=null, journalName=null, refType=null, unstructuredReference=欧阳俊峰. 福建沿海滩涂微型真核生物群落结构及对生物修复的响应[D]. 厦门: 集美大学, 2016., articleTitle=null, refAbstract=null), Reference(id=1215314011768410325, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=42, authorNames=null, journalName=null, refType=null, unstructuredReference=Ouyang Junfeng. Microeukaryotic community structure and response to bioremediation in intertidal mudflats of Fujian coast[D]. Xiamen: Jimei University, 2016., articleTitle=null, refAbstract=null), Reference(id=1215314011848102102, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=43, authorNames=null, journalName=null, refType=null, unstructuredReference=王翠, 郭晓峰, 方婧, 等. 闽浙沿岸流扩展范围的季节特征及其对典型海湾的影响[J]. 应用海洋学学报, 2018, 37(1): 1−8., articleTitle=null, refAbstract=null), Reference(id=1215314011936182487, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=44, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Cui, Guo Xiaofeng, Fang Jing, et al. Characteristics of seasonal spatial expansion of Fujian and Zhejiang Coastal Current and their bay effects[J]. Journal of Applied Oceanography, 2018, 37(1): 1−8., articleTitle=null, refAbstract=null), Reference(id=1215314012007485657, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=30, rfOrder=45, authorNames=null, journalName=null, refType=null, unstructuredReference=Yu Yong, Si Xiaosheng, Hu Changhua, et al. A review of recurrent neural networks: LSTM cells and network architectures[J]. Neural Computation, 2019, 31(7): 1235−1270., articleTitle=null, refAbstract=null), Reference(id=1215314012112343261, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=31, rfOrder=46, authorNames=null, journalName=null, refType=null, unstructuredReference=Pan S J, Yang Qiang. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345−1359., articleTitle=null, refAbstract=null), Reference(id=1215314012221395168, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=32, rfOrder=47, authorNames=null, journalName=null, refType=null, unstructuredReference=福建省科学技术厅. 大黄鱼养殖[M]. 北京: 海洋出版社, 2004., articleTitle=null, refAbstract=null), Reference(id=1215314012326252772, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=32, rfOrder=48, authorNames=null, journalName=null, refType=null, unstructuredReference=Province Department of Science and Technology of Fujian. Breeding and farming of Pseudosciaena Crocea[M]. Beijing: China Ocean Press, 2004., articleTitle=null, refAbstract=null), Reference(id=1215314012414333161, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=33, rfOrder=49, authorNames=null, journalName=null, refType=null, unstructuredReference=Yosinski J, Clune J, Bengio Y, et al. How transferable are features in deep neural networks?[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal: MIT Press, 2014., articleTitle=null, refAbstract=null), Reference(id=1215314012540162284, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=34, rfOrder=50, authorNames=null, journalName=null, refType=null, unstructuredReference=刘明, 王红蕾, 索良泽. 基于变权组合模型的中长期负荷概率密度预测[J]. 电力系统及其自动化学报, 2019, 31(7): 88−94., articleTitle=null, refAbstract=null), Reference(id=1215314012636631279, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=34, rfOrder=51, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Ming, Wang Honglei, Suo Liangze. Medium- and long-term load probability density forecasting based on variable weight combination model[J]. Proceedings of the CSU-EPSA, 2019, 31(7): 88−94., articleTitle=null, refAbstract=null), Reference(id=1215314012741488884, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=35, rfOrder=52, authorNames=null, journalName=null, refType=null, unstructuredReference=康俊锋, 谭建林, 方雷, 等. XGBoost-LSTM变权组合模型支持下短期PM2.5浓度预测—以上海为例[J]. 中国环境科学, 2021, 41(9): 4016−4025., articleTitle=null, refAbstract=null), Reference(id=1215314012842152186, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=35, rfOrder=53, authorNames=null, journalName=null, refType=null, unstructuredReference=Kang Junfeng, Tan Jianlin, Fang Lei, et al. Short-term PM2.5 concentration prediction based on XGBoost and LSTM variable weight combination model: a case study of Shanghai[J]. China Environmental Science, 2021, 41(9): 4016−4025., articleTitle=null, refAbstract=null), Reference(id=1215314012938621178, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=36, rfOrder=54, authorNames=null, journalName=null, refType=null, unstructuredReference=Benthuysen J A, Oliver E C J, Feng Ming, et al. Extreme marine warming across tropical Australia during austral summer 2015−2016[J]. Journal of Geophysical Research: Oceans, 2018, 123(2): 1301−1326., articleTitle=null, refAbstract=null), Reference(id=1215314013060255997, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=37, rfOrder=55, authorNames=null, journalName=null, refType=null, unstructuredReference=齐庆华, 蔡榕硕. 中国近海海表温度变化的极端特性及其气候特征研究[J]. 海洋学报, 2019, 41(7): 36−51., articleTitle=null, refAbstract=null), Reference(id=1215314013144142079, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=37, rfOrder=56, authorNames=null, journalName=null, refType=null, unstructuredReference=Qi Qinghua, Cai Rongshuo. Analysis on climate characteristics of sea surface temperature extremes in coastal China seas[J]. Haiyang Xuebao, 2019, 41(7): 36−51., articleTitle=null, refAbstract=null)], funds=[Fund(id=1215314004944277554, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, awardId=null, language=CN, fundingSource=福建省海洋经济发展补助资金项目(ZHHY-2019-1);国家重点研发计划(2016YFC0502901), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1215313997960762137, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=1, ext=[AuthorCompanyExt(id=1215313997969150746, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313997960762137, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 厦门大学 近海海洋环境科学国家重点实验室,福建 厦门 361102)]), AuthorCompany(id=1215313998032065313, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=1, ext=[AuthorCompanyExt(id=1215313998036259618, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313998032065313, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1State Key Laboratory of Marine Environmental Science, Xiamen University, Xiamen 361102, China)]), AuthorCompany(id=1215313998111757096, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=2, ext=[AuthorCompanyExt(id=1215313998120145704, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313998111757096, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 厦门大学 福建省海陆界面生态环境重点实验室,福建 厦门 361102)]), AuthorCompany(id=1215313998195643178, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=2, ext=[AuthorCompanyExt(id=1215313998204031788, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313998195643178, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Fujian Provincial Key Laboratory for Coastal Ecology and Environmental Studies, Xiamen University, Xiamen 361102, China)]), AuthorCompany(id=1215313998308889392, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=3, ext=[AuthorCompanyExt(id=1215313998317278001, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313998308889392, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 福建省渔业资源监测中心,福建 福州 350003)]), AuthorCompany(id=1215313998392775484, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, xref=3, ext=[AuthorCompanyExt(id=1215313998401164094, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, companyId=1215313998392775484, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Fishery Resources Monitoring Center of Fujian Province, Fuzhou 350003, China)])], figs=[ArticleFig(id=1215314002272506826, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Fig. 1, caption=Modeling process of DR-TLSTM model of Sansha Bay based on differential regression (DR) and transferable long short-term memory (TLSTM), figureFileSmall=XQxKrutXwvo/wTb4HCDpug==, figureFileBig=d0JfJieoWLsAiPXQdGPCag==, tableContent=null), ArticleFig(id=1215314002368975823, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=图1, caption=基于差分回归(DR)与可迁移长短期记忆网络(TLSTM)集成的三沙湾DR-TLSTM模型建模流程, figureFileSmall=XQxKrutXwvo/wTb4HCDpug==, figureFileBig=d0JfJieoWLsAiPXQdGPCag==, tableContent=null), ArticleFig(id=1215314002494804944, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Fig. 2, caption=The influence of transfer learning on the prediction of water temperature for the next 1−7 days in the Sansha Bay

a, c, e. Means and standard deviations of root mean square error (RMSE), mean absolute error (MAE), goodness of fit (R2) of 100 replicates at the maximum temperature; b, d, f. means and standard deviations of root mean square error (RMSE), mean absolute error (MAE), goodness of fit (R2) of 100 replicates at the minimum temperature

, figureFileSmall=lAMIzrObRZaKOG0J0k+ThQ==, figureFileBig=TJxCZBuCH74TnFMMdZKXTA==, tableContent=null), ArticleFig(id=1215314002578691028, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=图2, caption=迁移学习对模型预测三沙湾未来1~7 d水温的影响

a, c, e. 最高温100次重复测试的均方根误差(RMSE)、平均绝对误差(MAE)、拟合优度(R2)的平均值和标准差;b, d, f. 最低温100次重复测试的均方根误差(RMSE)、平均绝对误差(MAE)、拟合优度(R2)的平均值和标准差

, figureFileSmall=lAMIzrObRZaKOG0J0k+ThQ==, figureFileBig=TJxCZBuCH74TnFMMdZKXTA==, tableContent=null), ArticleFig(id=1215314002662577114, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Fig. 3, caption=Influence of variable-weight combination on the predicted water temperature for the next 1− 7 days in the Sansha Bay

a, c, e. Means and standard deviations of root mean square error (RMSE), mean absolute error (MAE), goodness of fit (R2) of 100 replicates at the maximum temperature; b, d, f. means and standard deviations of root mean square error (RMSE), mean absolute error (MAE), goodness of fit (R2) of 100 replicates at the minimum temperature

, figureFileSmall=a87PPaAqdbWlSUHnPECqnw==, figureFileBig=31mEzw9K05Lr2/TdtpuykA==, tableContent=null), ArticleFig(id=1215314002738074588, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=图3, caption=变权组合对模型预测三沙湾未来1~7 d水温的影响

a, c, e. 最高温100次重复测试的均方根误差(RMSE)、平均绝对误差(MAE)、拟合优度(R2)的平均值和标准差;b, d, f. 最低温100次重复测试的均方根误差(RMSE)、平均绝对误差(MAE)、拟合优度(R2)的平均值和标准差

, figureFileSmall=a87PPaAqdbWlSUHnPECqnw==, figureFileBig=31mEzw9K05Lr2/TdtpuykA==, tableContent=null), ArticleFig(id=1215314002809377761, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Fig. 4, caption=Influence of variable-weight combination on the prediction of water temperature sudden change point for the next 1−7 days in the Sansha Bay

a, c, e. Means and standard deviations of root mean square error (RMSE), mean absolute error (MAE), goodness of fit (R2) of 100 replicates at the maximum temperature; b, d, f. means and standard deviations of root mean square error (RMSE), mean absolute error (MAE), goodness of fit (R2) of 100 replicates at the minimum temperature

, figureFileSmall=P4gOUGEpb8ayr6D09aHsog==, figureFileBig=icJPmcGOXNxzg2FmuKuZ/Q==, tableContent=null), ArticleFig(id=1215314002884875236, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=图4, caption=变权组合对模型预测三沙湾未来1~7 d水温突变点的影响

a, c, e. 最高温100次重复测试的均方根误差(RMSE)、平均绝对误差(MAE)、拟合优度(R2)的平均值和标准差;b, d, f. 最低温100次重复测试的均方根误差(RMSE)、平均绝对误差(MAE)、拟合优度(R2)的平均值和标准差

, figureFileSmall=P4gOUGEpb8ayr6D09aHsog==, figureFileBig=icJPmcGOXNxzg2FmuKuZ/Q==, tableContent=null), ArticleFig(id=1215314002972955624, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Fig. 5, caption=Comparison of observed and predicted water temperature for the next 1−7 days in the Sansha Bay (2021−2022)

a, b. Comparison results on day 1; c, d. comparison results on day 2; e, f. comparison results on day 3; g, h. comparison results on day 4; i, j. comparison results on day 5; k, l. comparison results on day 6; m, n. comparison results on day 7

, figureFileSmall=PbfhVjpLoiglK79gY7wOYw==, figureFileBig=LpElzhtBsg3WseeoFqhfTg==, tableContent=null), ArticleFig(id=1215314003052647402, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=图5, caption=三沙湾未来1~7 d水温实测值和模型预测值对比(2021−2022年)

a, b. 第1天对比结果;c, d. 第2天对比结果;e, f. 第3天对比结果;g, h. 第4天对比结果;i, j. 第5天对比结果;k, l. 第6天对比结果;m, n. 第7天对比结果

, figureFileSmall=PbfhVjpLoiglK79gY7wOYw==, figureFileBig=LpElzhtBsg3WseeoFqhfTg==, tableContent=null), ArticleFig(id=1215314003157505006, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Fig. 6, caption=Operational information system interface based on water temperature early warning and forecast in the Sansha Bay by DR-TLSTM model based on differential regression (DR) and transferable long short-term memory (TLSTM), figureFileSmall=OMvISACLTBwbHa5Z9v19vQ==, figureFileBig=lL1as8PcgEDyDy9BKhaEFw==, tableContent=null), ArticleFig(id=1215314003262362611, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=图6, caption=基于差分回归(DR)与可迁移长短期记忆网络(TLSTM)集成的DR-TLSTM模型应用于三沙湾水温预警预报业务化的信息系统界面, figureFileSmall=OMvISACLTBwbHa5Z9v19vQ==, figureFileBig=lL1as8PcgEDyDy9BKhaEFw==, tableContent=null), ArticleFig(id=1215314003350442994, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Table 1, caption=

Structures and mathematical expressions of long short-term memory (LSTM) model

, figureFileSmall=null, figureFileBig=null, tableContent=
结构数学表达式
注:式中,$ {i}_{t} $为输入门;$ {f}_{t} $为遗忘门;$ {o}_{t} $为输出门;$ {\widehat{C}}_{t} $t时刻临时记忆单元的输出;$ {C}_{t} $t时刻记忆单元的输出;$ {h}_{t} $t时刻隐藏单元的输出;$ {x}_{t} $t时刻的输入数据;$ \sigma $为Sigmoid激活函数;$ \mathrm{t}\mathrm{a}\mathrm{n}\mathrm{h} $为双曲正切激活函数;$ {W}_{i} $$ {W}_{f} $$ {W}_{c} $$ {W}_{o} $为权重矩阵;$ {b}_{i} $$ {b}_{f} $$ {b}_{c} $$ {b}_{o} $为偏置向量;$ \odot $为逐元素点积运算。
输入门$ {i}_{t}=\sigma \left({W}_{i}\cdot \left[{h}_{t-1},{x}_{t}\right]+{b}_{i}\right) $
遗忘门$ {f}_{t}=\sigma \left({W}_{f}\cdot \left[{h}_{t-1},{x}_{t}\right]+{b}_{f}\right) $
输出门$ {o}_{t}=\sigma \left({W}_{o}\cdot \left[{h}_{t-1},{x}_{t}\right]+{b}_{o}\right) $
临时记忆单元${\widehat{C} }_{t}=\mathrm{tanh}\left({W}_{c}\cdot \left[{h}_{t-1},{x}_{t}\right]+{b}_{c}\right)$
记忆单元$ {C}_{t}={f}_{t}\odot {C}_{t-1}+{i}_{t}\odot {\widehat{C}}_{t} $
隐藏单元$ {h}_{t}={o}_{t}\odot \mathrm{t}\mathrm{a}\mathrm{n}\mathrm{h}\left({C}_{t}\right) $
), ArticleFig(id=1215314003530798074, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=表1, caption=

长短期记忆网络(LSTM)模型结构和数学表达式

, figureFileSmall=null, figureFileBig=null, tableContent=
结构数学表达式
注:式中,$ {i}_{t} $为输入门;$ {f}_{t} $为遗忘门;$ {o}_{t} $为输出门;$ {\widehat{C}}_{t} $t时刻临时记忆单元的输出;$ {C}_{t} $t时刻记忆单元的输出;$ {h}_{t} $t时刻隐藏单元的输出;$ {x}_{t} $t时刻的输入数据;$ \sigma $为Sigmoid激活函数;$ \mathrm{t}\mathrm{a}\mathrm{n}\mathrm{h} $为双曲正切激活函数;$ {W}_{i} $$ {W}_{f} $$ {W}_{c} $$ {W}_{o} $为权重矩阵;$ {b}_{i} $$ {b}_{f} $$ {b}_{c} $$ {b}_{o} $为偏置向量;$ \odot $为逐元素点积运算。
输入门$ {i}_{t}=\sigma \left({W}_{i}\cdot \left[{h}_{t-1},{x}_{t}\right]+{b}_{i}\right) $
遗忘门$ {f}_{t}=\sigma \left({W}_{f}\cdot \left[{h}_{t-1},{x}_{t}\right]+{b}_{f}\right) $
输出门$ {o}_{t}=\sigma \left({W}_{o}\cdot \left[{h}_{t-1},{x}_{t}\right]+{b}_{o}\right) $
临时记忆单元${\widehat{C} }_{t}=\mathrm{tanh}\left({W}_{c}\cdot \left[{h}_{t-1},{x}_{t}\right]+{b}_{c}\right)$
记忆单元$ {C}_{t}={f}_{t}\odot {C}_{t-1}+{i}_{t}\odot {\widehat{C}}_{t} $
隐藏单元$ {h}_{t}={o}_{t}\odot \mathrm{t}\mathrm{a}\mathrm{n}\mathrm{h}\left({C}_{t}\right) $
), ArticleFig(id=1215314003627267069, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Table 2, caption=

Main parameters and set values of long short-term memory (LSTM) model

, figureFileSmall=null, figureFileBig=null, tableContent=
主要参数设定值
注:MSE为均方误差,Adam为基于适应性低阶矩估计的一阶梯度优化算法结果。
LSTM层1神经元个数48
正则化层1正则化系数0.2
LSTM层2神经元个数32
正则化层2正则化系数0.2
损失函数MSE
优化器Adam
批处理大小32
迭代轮次150
), ArticleFig(id=1215314003748900865, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=表2, caption=

长短期记忆网络(LSTM)模型主要参数和设定值

, figureFileSmall=null, figureFileBig=null, tableContent=
主要参数设定值
注:MSE为均方误差,Adam为基于适应性低阶矩估计的一阶梯度优化算法结果。
LSTM层1神经元个数48
正则化层1正则化系数0.2
LSTM层2神经元个数32
正则化层2正则化系数0.2
损失函数MSE
优化器Adam
批处理大小32
迭代轮次150
), ArticleFig(id=1215314003853758470, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Table 3, caption=

Classification of water temperature warning levels in the Sansha Bay

, figureFileSmall=null, figureFileBig=null, tableContent=
温度T<10℃10℃≤T<12℃12℃≤T<14℃14℃≤T≤28℃28℃<T≤30℃30℃<T≤32℃T>32℃
等级红色低温橙色低温黄色低温正常黄色高温橙色高温红色高温
), ArticleFig(id=1215314003950227467, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=表3, caption=

三沙湾水温预警等级的划分

, figureFileSmall=null, figureFileBig=null, tableContent=
温度T<10℃10℃≤T<12℃12℃≤T<14℃14℃≤T≤28℃28℃<T≤30℃30℃<T≤32℃T>32℃
等级红色低温橙色低温黄色低温正常黄色高温橙色高温红色高温
), ArticleFig(id=1215314004034113549, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Table 4, caption=

Statistics of the means of 100 replicates in transfer learning test

, figureFileSmall=null, figureFileBig=null, tableContent=
预测天数评价指标目标域LSTM模型TLSTM模型(未微调)TLSTM模型
最高温最低温最高温最低温最高温最低温
第1天RMSE/℃5.425.690.420.310.380.28
MAE/℃5.305.540.360.280.350.25
R20.8230.8580.9900.9940.9910.995
第2天RMSE/℃5.575.890.520.390.470.34
MAE/℃5.455.720.430.330.430.30
R20.8370.8330.9680.9860.9680.987
第3天RMSE/℃5.746.020.620.470.570.43
MAE/℃5.615.870.520.400.510.37
R20.7950.8120.9310.9710.9310.971
第4天RMSE/℃5.966.170.750.570.680.52
MAE/℃5.846.020.610.460.610.44
R20.7400.8050.8840.9500.8860.951
第5天RMSE/℃6.226.370.840.640.760.59
MAE/℃6.096.220.690.520.680.50
R20.6710.7950.8340.9320.8390.933
第6天RMSE/℃6.426.530.910.710.830.65
MAE/℃6.306.380.750.580.750.56
R20.6650.7600.7830.9130.7940.914
第7天RMSE/℃6.616.700.970.790.880.73
MAE/℃6.496.540.800.650.810.64
R20.6530.7490.7370.8820.7590.884
), ArticleFig(id=1215314004109611027, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=表4, caption=

迁移学习测试时100次重复实验的平均值统计

, figureFileSmall=null, figureFileBig=null, tableContent=
预测天数评价指标目标域LSTM模型TLSTM模型(未微调)TLSTM模型
最高温最低温最高温最低温最高温最低温
第1天RMSE/℃5.425.690.420.310.380.28
MAE/℃5.305.540.360.280.350.25
R20.8230.8580.9900.9940.9910.995
第2天RMSE/℃5.575.890.520.390.470.34
MAE/℃5.455.720.430.330.430.30
R20.8370.8330.9680.9860.9680.987
第3天RMSE/℃5.746.020.620.470.570.43
MAE/℃5.615.870.520.400.510.37
R20.7950.8120.9310.9710.9310.971
第4天RMSE/℃5.966.170.750.570.680.52
MAE/℃5.846.020.610.460.610.44
R20.7400.8050.8840.9500.8860.951
第5天RMSE/℃6.226.370.840.640.760.59
MAE/℃6.096.220.690.520.680.50
R20.6710.7950.8340.9320.8390.933
第6天RMSE/℃6.426.530.910.710.830.65
MAE/℃6.306.380.750.580.750.56
R20.6650.7600.7830.9130.7940.914
第7天RMSE/℃6.616.700.970.790.880.73
MAE/℃6.496.540.800.650.810.64
R20.6530.7490.7370.8820.7590.884
), ArticleFig(id=1215314004214468630, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Table 5, caption=

Statistics of the means of 100 replicates in variable-weight combination test

, figureFileSmall=null, figureFileBig=null, tableContent=
预测天数评价指标TLSTM模型纯差分回归模型混差分回归模型DR-TLSTM集成模型
最高温最低温最高温最低温最高温最低温最高温最低温
第1天RMSE/℃0.380.280.160.140.150.150.130.13
MAE/℃0.350.250.120.110.110.110.110.11
R20.9910.9950.9940.9960.9950.9950.9950.996
第2天RMSE/℃0.470.340.300.250.310.250.260.22
MAE/℃0.430.300.230.300.220.170.220.18
R20.9680.9870.9780.9890.9780.9890.9780.989
第3天RMSE/℃0.570.430.450.370.480.370.390.33
MAE/℃0.510.370.340.290.330.260.330.27
R20.9310.9710.9460.9750.9470.9750.9470.975
第4天RMSE/℃0.680.520.600.480.630.480.510.44
MAE/℃0.610.440.460.380.430.350.430.36
R20.8860.9510.9020.9560.9060.9560.9050.956
第5天RMSE/℃0.760.590.710.570.740.570.600.51
MAE/℃0.680.500.550.470.530.440.520.43
R20.8390.9330.8630.9390.8700.9410.8670.940
第6天RMSE/℃0.830.650.810.670.830.670.680.59
MAE/℃0.750.560.660.580.630.540.620.52
R20.7940.9140.8220.9170.8340.9220.8310.921
第7天RMSE/℃0.880.730.930.790.950.790.770.69
MAE/℃0.810.640.770.680.740.650.710.63
R20.7590.8840.7750.8840.7930.8890.7910.890
), ArticleFig(id=1215314004319326234, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=表5, caption=

变权组合测试时100次重复实验的平均值统计

, figureFileSmall=null, figureFileBig=null, tableContent=
预测天数评价指标TLSTM模型纯差分回归模型混差分回归模型DR-TLSTM集成模型
最高温最低温最高温最低温最高温最低温最高温最低温
第1天RMSE/℃0.380.280.160.140.150.150.130.13
MAE/℃0.350.250.120.110.110.110.110.11
R20.9910.9950.9940.9960.9950.9950.9950.996
第2天RMSE/℃0.470.340.300.250.310.250.260.22
MAE/℃0.430.300.230.300.220.170.220.18
R20.9680.9870.9780.9890.9780.9890.9780.989
第3天RMSE/℃0.570.430.450.370.480.370.390.33
MAE/℃0.510.370.340.290.330.260.330.27
R20.9310.9710.9460.9750.9470.9750.9470.975
第4天RMSE/℃0.680.520.600.480.630.480.510.44
MAE/℃0.610.440.460.380.430.350.430.36
R20.8860.9510.9020.9560.9060.9560.9050.956
第5天RMSE/℃0.760.590.710.570.740.570.600.51
MAE/℃0.680.500.550.470.530.440.520.43
R20.8390.9330.8630.9390.8700.9410.8670.940
第6天RMSE/℃0.830.650.810.670.830.670.680.59
MAE/℃0.750.560.660.580.630.540.620.52
R20.7940.9140.8220.9170.8340.9220.8310.921
第7天RMSE/℃0.880.730.930.790.950.790.770.69
MAE/℃0.810.640.770.680.740.650.710.63
R20.7590.8840.7750.8840.7930.8890.7910.890
), ArticleFig(id=1215314004428378141, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Table 6, caption=

Statistics of the means of 100 replicates in sudden change point prediction test

, figureFileSmall=null, figureFileBig=null, tableContent=
预测天数评价指标TLSTM模型纯差分回归模型混差分回归模型DR-TLSTM集成模型
最高温最低温最高温最低温最高温最低温最高温最低温
第1天RMSE/℃0.440.290.420.320.400.340.350.29
MAE/℃0.400.390.360.300.330.290.330.29
R20.9920.9920.9950.9930.9960.9930.9960.994
第2天RMSE/℃0.710.610.750.630.780.630.650.57
MAE/℃0.660.600.580.600.560.520.570.56
R20.9400.9350.9540.9510.9540.9510.9540.951
第3天RMSE/℃1.070.771.020.671.080.670.890.61
MAE/℃0.950.720.800.750.760.690.760.71
R20.7980.7050.8060.7470.8080.7470.8070.747
第4天RMSE/℃1.310.951.010.811.060.810.870.74
MAE/℃1.130.950.850.920.800.850.800.87
R20.7290.6930.7820.7800.7850.7810.7840.781
第5天RMSE/℃1.320.951.180.991.230.991.010.90
MAE/℃1.040.940.890.950.850.880.840.88
R20.6660.6720.7880.7290.7940.7310.7920.730
第6天RMSE/℃1.331.001.191.071.231.071.010.95
MAE/℃1.031.051.001.070.941.000.940.98
R20.6930.6910.7120.6940.7220.6980.7200.698
第7天RMSE/℃1.231.031.311.111.341.121.090.98
MAE/℃1.071.061.021.020.970.970.950.95
R20.6300.4500.6360.5260.6510.5290.6490.530
), ArticleFig(id=1215314004545818658, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=表6, caption=

突变点测试时100次重复实验的平均值统计

, figureFileSmall=null, figureFileBig=null, tableContent=
预测天数评价指标TLSTM模型纯差分回归模型混差分回归模型DR-TLSTM集成模型
最高温最低温最高温最低温最高温最低温最高温最低温
第1天RMSE/℃0.440.290.420.320.400.340.350.29
MAE/℃0.400.390.360.300.330.290.330.29
R20.9920.9920.9950.9930.9960.9930.9960.994
第2天RMSE/℃0.710.610.750.630.780.630.650.57
MAE/℃0.660.600.580.600.560.520.570.56
R20.9400.9350.9540.9510.9540.9510.9540.951
第3天RMSE/℃1.070.771.020.671.080.670.890.61
MAE/℃0.950.720.800.750.760.690.760.71
R20.7980.7050.8060.7470.8080.7470.8070.747
第4天RMSE/℃1.310.951.010.811.060.810.870.74
MAE/℃1.130.950.850.920.800.850.800.87
R20.7290.6930.7820.7800.7850.7810.7840.781
第5天RMSE/℃1.320.951.180.991.230.991.010.90
MAE/℃1.040.940.890.950.850.880.840.88
R20.6660.6720.7880.7290.7940.7310.7920.730
第6天RMSE/℃1.331.001.191.071.231.071.010.95
MAE/℃1.031.051.001.070.941.000.940.98
R20.6930.6910.7120.6940.7220.6980.7200.698
第7天RMSE/℃1.231.031.311.111.341.121.090.98
MAE/℃1.071.061.021.020.970.970.950.95
R20.6300.4500.6360.5260.6510.5290.6490.530
), ArticleFig(id=1215314004663259175, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=EN, label=Table 7, caption=

Statistical results of DR-TLSTM model based on differential regression (DR) and transferable long short-term memory (TLSTM) for water temperature early warning accuracy in the Sansha Bay

, figureFileSmall=null, figureFileBig=null, tableContent=
设备编号第1天第2天第3天第4天第5天第6天第7天
1#渔排基96.27%94.03%91.04%88.06%84.33%82.84%82.09%
2#渔排基95.52%94.03%91.04%88.06%85.82%83.58%80.60%
3#渔排基97.76%95.52%91.79%91.79%88.06%87.31%79.10%
), ArticleFig(id=1215314004776505387, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1211297842019299501, language=CN, label=表7, caption=

基于差分回归(DR)与可迁移长短期记忆网络(TLSTM)集成的DR-TLSTM模型对三沙湾水温预警准确率的统计结果

, figureFileSmall=null, figureFileBig=null, tableContent=
设备编号第1天第2天第3天第4天第5天第6天第7天
1#渔排基96.27%94.03%91.04%88.06%84.33%82.84%82.09%
2#渔排基95.52%94.03%91.04%88.06%85.82%83.58%80.60%
3#渔排基97.76%95.52%91.79%91.79%88.06%87.31%79.10%
)], attaches=null, journal=Journal(id=1146441459026210850, delFlag=0, nameCn=海洋学报, nameEn=Haiyang Xuebao, nameHistory1=null, nameHistory2=null, issn=0253-4193, eissn=null, cn=11-2055/P, coden=null, periodic=0, language=CN, oaType=否, 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=uYi7hkkrve+l8pIcwqcaQQ==, journalPrice=null, startedYear=null, abbrevIsoEn=null, journalRemark=null, publicationField=null, createdTime=1751262543687, updatedTime=1761729782936, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=H, firstLetterEn=H, subjectCode=Natural Sciences, subjectName=Natural Sciences, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=uYi7hkkrve+l8pIcwqcaQQ==, picEn=C0WLQb7uW3ok8EkkVOAGuw==, jcr=null, cjcr=null, exts=[JournalExt(id=1190344242636624294, language=CN, name=海洋学报, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.hyxbocean.cn/, createdTime=1761729782971, updatedTime=1761729782971, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.hyxb.org.cn/aos/ch/author/login.aspx, submissionEditorUrl=http://www.hyxb.org.cn/aos/ch/login.aspx, submissionReviewUrl=http://www.hyxb.org.cn/aos/ch/auditor/login.aspx, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1190344242712121767, language=EN, name=Haiyang Xuebao, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.aosocean.com/, createdTime=1761729782989, updatedTime=1761729782989, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.hyxb.org.cn/aos/ch/author/login.aspx, submissionEditorUrl=http://www.hyxb.org.cn/aos/ch/login.aspx, submissionReviewUrl=http://www.hyxb.org.cn/aos/ch/auditor/login.aspx, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1149651085930835976, websiteList=[Website(id=1188165202219512001, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1149651085930835976, 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/hyxb/CN, language=CN, createTime=1761210259251, createBy=18614031015, updateTime=1761210330879, updateBy=18614031015, name=海洋学报-中文, tplId=1146099689490845704, title=海洋学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1188166688563413602, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202219512001, code=articleTextType, value=kx, createTime=1761210613623, updateTime=1761210613623, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166688538247775, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202219512001, code=banner, value=null, createTime=1761210613617, updateTime=1761210613617, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166688529859166, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202219512001, code=logo, value=https://castjournals.cast.org.cn/joweb/hyxb/CN/file/pic?fileId=BDEio/cxHnid8OD4QxrAYQ==, createTime=1761210613615, updateTime=1761210613615, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166688555024993, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202219512001, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/hyxb/CN/file/pic, createTime=1761210613621, updateTime=1761210613621, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166688546636384, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202219512001, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1761210613619, updateTime=1761210613619, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166688575996515, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202219512001, code=themeColor, value=null, createTime=1761210613626, updateTime=1761210613626, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166688596968036, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202219512001, code=themeStyle, value=null, createTime=1761210613631, updateTime=1761210613631, creator=18614031015, updator=18614031015)]), Website(id=1188165202282426564, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1149651085930835976, 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/hyxb/EN, language=EN, createTime=1761210259266, createBy=18614031015, updateTime=1761210377920, updateBy=18614031015, name=海洋学报-英文, tplId=1146101810881728533, title=Haiyang Xuebao, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1188166798101856873, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202282426564, code=articleTextType, value=kx, createTime=1761210639739, updateTime=1761210639739, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166798076691046, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202282426564, code=banner, value=null, createTime=1761210639733, updateTime=1761210639733, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166798068302437, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202282426564, code=logo, value=https://castjournals.cast.org.cn/joweb/hyxb/EN/file/pic?fileId=BDEio/cxHnid8OD4QxrAYQ==, createTime=1761210639731, updateTime=1761210639731, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166798093468264, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202282426564, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/hyxb/EN/file/pic, createTime=1761210639737, updateTime=1761210639737, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166798085079655, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202282426564, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1761210639735, updateTime=1761210639735, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166798106051178, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202282426564, code=themeColor, value=null, createTime=1761210639740, updateTime=1761210639740, creator=18614031015, updator=18614031015), WebsiteProps(id=1188166798110245483, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1188165202282426564, code=themeStyle, value=null, createTime=1761210639741, updateTime=1761210639741, creator=18614031015, updator=18614031015)])], journalTitle=海洋学报, weixinUrl=null, journalUrl=http://www.hyxbocean.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Haiyang Xuebao, journalPhotoCn=uYi7hkkrve+l8pIcwqcaQQ==, journalPhotoEn=C0WLQb7uW3ok8EkkVOAGuw==, journalFirstLetter=H, 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=, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/hyxb/CN/10.12284/hyxb2023027, detailUrlEn=https://castjournals.cast.org.cn/joweb/hyxb/EN/10.12284/hyxb2023027, pdfUrlCn=https://castjournals.cast.org.cn/joweb/hyxb/CN/PDF/10.12284/hyxb2023027, pdfUrlEn=https://castjournals.cast.org.cn/joweb/hyxb/EN/PDF/10.12284/hyxb2023027, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于差分回归模型和可迁移长短期记忆网络集成的三沙湾水温预测
收藏切换
PDF下载
赖晓倩 1 , 余镒琦 2 , 梁中耀 2 , 陈火荣 3 , 陈能汪 1, 2, *
海洋学报 | 论文 2023,45(4): 165-178
收起
收藏切换
海洋学报 | 论文 2023, 45(4): 165-178
基于差分回归模型和可迁移长短期记忆网络集成的三沙湾水温预测
全屏
赖晓倩1 , 余镒琦2, 梁中耀2, 陈火荣3, 陈能汪1, 2, *
作者信息
  • 1 厦门大学 近海海洋环境科学国家重点实验室,福建 厦门 361102
  • 2 厦门大学 福建省海陆界面生态环境重点实验室,福建 厦门 361102
  • 3 福建省渔业资源监测中心,福建 福州 350003
  • 赖晓倩(1998-),女,福建省龙岩市人,助理工程师,主要从事大数据建模研究。E-mail:

通讯作者:

*陈能汪(1976-),男,教授,主要从事海陆界面生态环境研究。E-mail:
Water temperature prediction in the Sansha Bay based on the integration of differential regression model and transportable long short-term memory network
Xiaoqian Lai1 , Yiqi Yu2, Zhongyao Liang2, Huorong Chen3, Nengwang Chen1, 2, *
Affiliations
  • 1State Key Laboratory of Marine Environmental Science, Xiamen University, Xiamen 361102, China
  • 2Fujian Provincial Key Laboratory for Coastal Ecology and Environmental Studies, Xiamen University, Xiamen 361102, China
  • 3Fishery Resources Monitoring Center of Fujian Province, Fuzhou 350003, China
出版时间: 2023-03-31 doi: 10.12284/hyxb2023027
文章导航
收藏切换

水温预测是保障近海渔业生产和环境安全的关键技术。现有的数值模型开发成本大,业务化应用不足。本文提出了一种集成差分回归(Differential Regression, DR)和可迁移长短期记忆网络(Transferable Long Short-Term Memory, TLSTM)的水温预测方法。以厦门湾(源域,数据多)和三沙湾(目标域,数据少)水温为研究对象,根据三沙湾在线监测水温和预报气温数据建立了DR模型,根据厦门湾长时间监测水温数据建立了TLSTM模型,采用变权算法将纯差分回归模型、混差分回归模型和TLSTM模型集成为三沙湾DR-TLSTM模型,对模型性能进行了评估,并与仅根据三沙湾少量监测数据建立的LSTM模型效果进行了对比。结果表明:(1) TLSTM模型的预测精度优于基于目标域少量数据建立的LSTM模型;(2) DR-TLSTM集成模型具有较高的预测精度,未来1~7 d预测的均方根误差为0.13~0.77℃,未来1~3 d预测的均方根误差小于0.4℃;(3) DR-TLSTM集成模型可有效预测水温骤升或骤降趋势,对水温突变点的预测均方根误差为0.29~1.09℃。基于本文建立的DR-TLSTM集成模型,实现了三沙湾渔业水域的水温预警预报业务化信息服务。

水温预测  /  回归模型  /  LSTM模型  /  迁移学习  /  变权集成

Water temperature prediction is a key technology to ensure the production of coastal fisheries and environmental safety. The existing numerical models have high development costs with insufficient business applications. This study develops a prediction method of water temperature through integrating differential regression (DR) and transferable long short-term memory (TLSTM). Taking the water temperature of Xiamen Bay (source domain, with a large number of data) and Sansha Bay (target domain, with less data) as the research object, the DR model is established based on the data of monitoring water temperature and forecast temperature in the Sansha Bay, and the TLSTM model is established based on the long-term monitoring data of water temperature in the Xiamen Bay. The pure differential regression model, mixed differential regression model and TLSTM model are integrated into the DR-TLSTM model of Sansha Bay by using variable weight algorithm, and the performance of the model is evaluated, the results are compared with the LSTM model based on only a small amount of monitoring data in the Sansha Bay. The results show that: (1) the prediction accuracy of TLSTM model is better than that of LSTM model based on a small amount of data in the target domain; (2) the DR-TLSTM model has high prediction accuracy, and the root mean square error of prediction in the next 1−7 days is 0.13−0.77℃, and the root mean square error of prediction in the next 1−3 days is less than 0.4℃; (3) the DR-TLSTM model can effectively predict the sudden rise or fall trend of water temperature, and the root mean square error of predicting the sudden change point of water temperature is 0.29−1.09℃. Based on the DR-TLSTM model, the operational information service of water temperature early warning and forecast in the Sansha Bay is realized.

water temperature prediction  /  regression model  /  LSTM model  /  transfer learning  /  variable weight integration
赖晓倩, 余镒琦, 梁中耀, 陈火荣, 陈能汪. 基于差分回归模型和可迁移长短期记忆网络集成的三沙湾水温预测. 海洋学报, 2023 , 45 (4) : 165 -178 . DOI: 10.12284/hyxb2023027
Xiaoqian Lai, Yiqi Yu, Zhongyao Liang, Huorong Chen, Nengwang Chen. Water temperature prediction in the Sansha Bay based on the integration of differential regression model and transportable long short-term memory network[J]. Haiyang Xuebao, 2023 , 45 (4) : 165 -178 . DOI: 10.12284/hyxb2023027
海水养殖场一般选在沿海半封闭的内湾。自20世纪80年代起,沿海地区的海水养殖业迅猛发展[1]。渔业用海是我国最主要的海域使用类型。然而,很多海湾面临水质恶化和生态退化问题,危及海水养殖产业的可持续发展[2-3]。大量的生产实践和科学研究表明,海洋生物的生长和繁殖都与水温的变化有着密切关系[4-10]。当水温超过生物的生理极限时将导致生物死亡,水温过高还会造成饲料的发酵和微生物耗氧,最后导致养殖生物缺氧死亡[11]。此外,极端海温事件(热浪、寒潮)发生时易造成渔业经济损失。因此,海湾水温预测是保障渔业生产和环境安全的关键技术。
现有的水温预测主要采用数值模拟方法和数据驱动方法[12]。数值模拟方法通常需要建立复杂的多维机理模型,模型参数多且难获取,开发成本大,难以支撑业务化应用。数据驱动方法需建立水温和相关影响因子的经验关系,模型的准确性依赖于大量的监测数据。近年来,随着水质在线监测技术的发展,采用数据驱动方法进行水温的精确预测成为可能。长短期记忆网络(Long Short-Term Memory, LSTM)模型是一种高级循环神经网络(Recurrent Neural Network, RNN)模型,能够有效地克服传统RNN训练过程中梯度消失和爆炸的问题[13-14],在水温预测中展现出良好的效果[15-19]。例如,Zhang等[15]采用LSTM模型对海水表面温度进行预测,给出了未来1 d和3 d的短期预测结果与周和月均值的中长期预测结果。Hou等[16]提出了一种新的密集扩张卷积LSTM模型来预测海面温度。Yang等[17]结合时间和空间信息建立了一种端到端的、可训练的LSTM模型来预测海温。张昆[18]利用多层卷积LSTM模型对不同深度层的海水温度进行预测。Kim等[19]提出了一种基于LSTM模型的高海温预测方法。需要指出的是,大多数研究采用单一LSTM模型进行预测。然而,海温数据具有非线性和非平稳性的特点[20],且受到太阳辐射、海洋平流和海气热通量交换等多种扰动因素的影响[21],单一模型难以准确捕捉海温的变化规律,特别是水温骤升或骤降的趋势。此外,LSTM模型需要大量的训练数据才能获得较高的预测精度[22],但在很多实际应用场景中并没有足够多的实测数据。迁移学习方法的提出有效地解决了模型训练过程中样本数量不足问题,它能将在相似领域/区域(源域)学到的知识用于目标对象(目标域)[23]。迁移学习已广泛地应用于图像分析和自然语言处理等领域[24-25],但在海湾渔业水域环境预测的应用未见报道。
据此,本文提出了一种基于差分回归(Differential Regression, DR)和可迁移长短期记忆网络(Transferable LSTM, TLSTM)集成的方法(DR-TLSTM)用于历史监测数据较少的三沙湾的水温预测。首先,采用三沙湾水温数据和气温预报数据构建纯差分回归模型和混差分回归模型;其次,以历史监测数据较多的厦门湾为源域预训练LSTM模型,并将其迁移到三沙湾(冻结模型全部或部分权重参数,并以三沙湾数据进行模型再训练)构建TLSTM模型,获得水温预测结果;最后,采用变权算法将纯差分回归模型、混差分回归模型和TLSTM模型集成为DR-TLSTM模型,并进行三沙湾的水温等级预警预报业务化应用。
三沙湾位于福建省东北部沿海,由三都澳、卢门港、白马港、盐田港、东吾洋、官井洋、福鼎洋等组成,水域面积为570 km2,生物资源丰富,是我国珍稀物种中华白海豚的栖息地和我国著名的大黄鱼产卵场,在此大黄鱼网箱养殖规模和产值居全国第一位。三沙湾属于亚热带海洋性气候,年平均降水量约为1 631 mm,平均气温约为18.5℃,水温年变化范围为13.0~29.9℃[26]
厦门湾位于福建省南部,由九龙江口、西海域、南部海域(外港)、同安湾(包括浔江和东咀港)和东部海域等组成,水域面积为154.18 km2。厦门湾属于南亚热带季风海洋性气候,年平均降水量约为1 100 mm,年平均气温约为21℃,水温年变化范围为13.9~29.0℃[27]。厦门湾与三沙湾同属亚热带海洋性气候[28],且都受到闽浙沿岸流的影响[26, 29],两者的气候和水文条件较为接近,本文分别选择厦门湾和三沙湾作为源域和目标域进行研究。
本文采用的厦门湾水温数据由厦门大学在厦门湾布放的水质浮标测定,时间范围为2015年1月1日至2018年12月31日;三沙湾水温数据由福建省渔业监测中心在三沙湾布放的水质浮标测定,时间范围为2021年9月17日至2022年3月16日;水温数据的时间频率均为1次/(30 min)。气温预报数据来源于中国天气网(www.weather.com.cn)每日发布的宁德7日天气预报数据,时间范围为2021年9月17日至2022年3月16日,时间频率为1次/d。采用三沙湾2021年12月25日至2022年3月16日的水温数据进行模型的测试评估,其余数据均用于模型构建。
收集到的厦门湾、三沙湾的水温数据和气温预报数据,剔除非数值型数据和重复数据后,用阈值检验、均方差检验和尖峰检验剔除异常值,并用线性插值算法进行缺失值插补。处理完毕后,将数据重采样为日频率数据。
本文构建的DR-TLSTM集成模型流程见图1,包括以下4个步骤:(1)采用三沙湾(目标域)水温数据和三沙湾气温预报数据构建差分回归模型(包括纯差分和混差分回归模型);(2)采用厦门湾(源域)水温数据预训练LSTM模型;(3) 采用三沙湾(目标域)水温数据对LSTM模型进行迁移;(4) 采用变权算法将纯差分回归模型、混差分回归模型和TLSTM模型集成为DR-TLSTM模型。
LSTM模型的基本结构单元由遗忘门、输入门、输出门、记忆单元、临时记忆单元、隐藏单元等结构组成,各结构的数学表达式见表1。LSTM结构单元的计算过程包括以下几个步骤:
(1) 根据t时刻的输入数据$ {x}_{t} $和上一时刻的隐藏单元状态$ {h}_{t-1} $,计算遗忘门状态$ {f}_{t} $
(2) 根据t时刻的输入数据$ {x}_{t} $和上一时刻的隐藏单元状态$ {h}_{t-1} $,计算输入门状态$ {i}_{t} $
(3) 根据t时刻的输入数据$ {x}_{t} $和上一时刻的隐藏单元状态$ {h}_{t-1} $,计算临时记忆单元状态$ {\widehat{C}}_{t} $
(4) 根据遗忘门状态$ {f}_{t} $和上一时刻记忆单元状态$ {C}_{t-1} $来确定上一时刻信息的遗忘或保留;
(5) 根据输入门状态$ {i}_{t} $和临时记忆单元状态$ {\widehat{C}}_{t} $来确定当前时刻信息的遗忘或保留;
(6) 根据上一时刻信息和当前时刻信息的遗忘或保留情况,计算当前时刻的记忆单元状态$ {C}_{t} $
(7) 根据t时刻的输入数据$ {x}_{t} $和上一时刻的隐藏单元状态$ {h}_{t-1} $,计算输出门状态$ {o}_{t} $
(8) 根据输出门状态$ {o}_{t} $和当前时刻的记忆单元状态$ {C}_{t} $,计算当前时刻的隐藏单元状态$ {h}_{t} $$ {h}_{t} $即为当前时刻LSTM结构单元的输出;
(9) 将当前时刻的记忆单元状态$ {C}_{t} $和隐藏单元状态$ {h}_{t} $输入至下一LSTM结构单元,并重复以上步骤。
单层LSTM模型提取数据信息的能力是有限的,可以将多个LSTM模型组合成特定架构以提高预测能力。堆叠LSTM模型是最常用的组合模型之一。与单层LSTM模型相比,堆叠LSTM模型的网络结构更加复杂,特征提取能力更强[30]。因此,本文选用堆叠LSTM模型来进行水温预测。
本文用Z-Score标准化方法将厦门湾(源域)2015年1月1日至2018年12月31日的水温数据转化为平均值为0、标准差为1的数据集,按4∶1的比例将数据集划分为训练集和测试集。经过网络调参和模型测试,根据预测精度选出包含两个LSTM网络层的堆叠LSTM模型,模型的主要参数和设定值见表2
迁移学习的定义中包含域和任务这两个基本概念。域D由输入数据的所有参数组成的参数空间X和每一维参数的概率分布Px)构成,其中x={${x}_{1}, $ ${x}_{2}, $ $\cdots,{x}_{n}$}∈X。任务T由标签空间Y和目标函数f(·)构成。对于给定的源域DS和对应的源任务TS、目标域DT和对应的目标任务TT,迁移学习的目标是通过DSTS中的知识提高DT中目标函数fT(·)的学习效果[31]
微调是一种用目标域数据对预训练模型未冻结层的权重参数进行再训练的迁移学习方法。本文采用两种方式对预训练的LSTM模型进行迁移:一种是不进行微调,采用全部模型参数,将预训练的LSTM模型直接应用于三沙湾(目标域)的水温预测,将其记为TLSTM模型(未微调);另一种是进行微调,冻结预训练的LSTM模型的首个网络层的参数,用三沙湾2021年9月17日至12月24日的水温数据对剩余参数进行微调,将其记为TLSTM模型。模型微调的主要参数包括损失函数、优化器、批处理大小和迭代轮次,其设定值分别为MSE、Adam、32和150。
水温变化和气温变化有明显的相关性,本文根据三沙湾在线监测水温和预报气温数据,构建纯差分回归模型和混差分回归模型。
(1) 构建纯差分回归模型
按式(1)、式(2)构建线性回归方程并计算$ {w}_{p1} $$ {w}_{p2} $$ {w}_{p3} $$ {w}_{p4}^{n} $ (2≤n≤7),按式(3)、式(4)构建纯差分回归模型。
$ {x}_{t}={x}_{t-1}+\sum _{i=1}^{3}\left[{w}_{pi}\cdot \left({f}_{t+1-i}^{1}-{f}_{t-i}^{1}\right)\right], $
$ {x}_{t}={x}_{t+1-n}+{w}_{p4}^{n}\cdot \left({f}_{t}^{n}-{f}_{t}^{1}\right), $
$ {p}_{t}^{1}={x}_{t-1}+\sum _{i=1}^{3}\left[{w}_{pi}\cdot \left({f}_{t+1-i}^{1}-{f}_{t-i}^{1}\right)\right], $
$ {p}_{t}^{n}={p}_{t}^{1}+{w}_{p4}^{n}\cdot \left({f}_{t}^{n}-{f}_{t}^{1}\right), $
式中,$ {x}_{t} $表示第t天的水温实测值;${f}_{t}^{1},$ ${f}_{t}^{2},$$\cdots, {f}_{t}^{7}$表示第t天发布的未来1~7 d气温预报数据;${p}_{t}^{1},$${p}_{t}^{2},$$\cdots,{p}_{t}^{7}$表示模型在第t天的未来1~7 d水温预测值;$ {w}_{p1} $$ {w}_{p2} $$ {w}_{p3} $$ {w}_{p4}^{n} $(2≤n≤7)为权重系数。
(2) 构建混差分回归模型
按式(5)、式(6)构建线性回归方程并计算$ {w}_{m1} $$ {w}_{m2} $$ {w}_{m3}^{n} $ (2≤n≤7),按式(7)、式(8)构建混差分回归模型。
$ {x}_{t}={x}_{t-1}+{w}_{m1}\cdot \left({x}_{t-1}-{x}_{t-2}\right)+ {w}_{m2}\cdot \left({f}_{t}^{1}-{f}_{t-1}^{1}\right), $
$ {x}_{t}={x}_{t+1-n}+{w}_{m3}^{n}\cdot \left({f}_{t}^{n}-{f}_{t}^{1}\right), $
$ {m}_{t}^{1}={x}_{t-1}+{w}_{m1}\cdot \left({x}_{t-1}-{x}_{t-2}\right)+{w}_{m2}\cdot \left({f}_{t}^{1}-{f}_{t-1}^{1}\right), $
$ {m}_{t}^{n}={m}_{t}^{1}+{w}_{m3}^{n}\cdot \left({f}_{t}^{n}-{f}_{t}^{1}\right), $
式中,$ {x}_{t} $表示第t天的水温实测值;${f}_{t}^{1},$${f}_{t}^{2},$$\cdots, {f}_{t}^{7}$表示第t天发布的未来1~7 d气温预报数据;${m}_{t}^{1},$${m}_{t}^{2},$$\cdots, {m}_{t}^{7}$表示模型在第t天的未来1~7 d水温预测值;$ {w}_{m1} $$ {w}_{m2} $$ {w}_{m3}^{n} $(2≤n≤7)为权重系数。
变权算法原理见式(9)至式(12)。
$ {R}_{i}\left(t\right)=\left|{f}_{i}\left(t\right)-T\left(t\right)\right|, $
$ {D}_{i}\left(t\right) = \left\{ \begin{array}{cc}1, &第1天\text{,}\\ {D}_{i}\left(t - 1\right), & 第t天且{R}_{i}\left(t\right) \ne {\rm{min}}\left\{{R}_{1}\left(t\right),{R}_{2}\left(t\right),{R}_{3}\left(t\right)\right\},\\ {D}_{i}\left(t - 1\right) + 1, & 第t天且{R}_{i}\left(t\right) = {\rm{min}}\left\{{R}_{1}\left(t\right),{R}_{2}\left(t\right),{R}_{3}\left(t\right)\right\}, \end{array}\right. $
$ {W}_{i}\left(t+1\right)=\frac{{D}_{i}\left(t\right)}{{D}_{1}\left(t\right)+{D}_{2}\left(t\right)+{D}_{3}\left(t\right)}, $
$ Y\left(t+1\right)=\sum _{i=1}^{3}{W}_{i}\left(t+1\right)·{F}_{i}\left(t+1\right), $
式中,$ {R}_{i}\left(t\right) $表示第i个模型在第t天的预测误差;$ {f}_{i}\left(t\right) $表示第i个模型在第t天的当天水温预测值;$ T\left(t\right) $表示第t天的水温实测值;$ {D}_{i}\left(t\right) $表示第i个模型在第t天的误差最小累积天数;$ {W}_{i}\left(t+1\right) $表示第i个模型在第t+1天的权重;$ Y\left(t+1\right) $表示DR-TLSTM集成模型在第t+1天的预测结果;$ {F}_{i}\left(t+1\right) $表示第i个模型在第t+1天的预测结果(1≤i≤3)。
本文进行的模型测试包括迁移学习测试、变权组合测试和突变点测试。
(1) 迁移学习测试:基于三沙湾2021年9月17日至12月24日的水温数据构建目标域LSTM模型,与TLSTM模型(未微调)和TLSTM模型进行预测效果对比。
(2) 变权组合测试:将TLSTM模型、纯差分回归模型、混差分回归模型和DR-TLSTM集成模型进行预测效果对比。
(3) 突变点测试:对三沙湾2021年9月17日至12月24日的水温逐日变化量进行递减排序,将百分位为10%的数值定义为水温突变变化量阈值,将三沙湾2021年12月25日至2022年3月16日的水温逐日变化量高于该阈值的的数据定义为水温突变点。对比TLSTM模型、纯差分回归模型、混差分回归模型和DR-TLSTM集成模型在水温突变点的预测效果。
采用平均绝对误差(MAE)、均方根误差(RMSE)和拟合优度(R2)作为模型预测性能的评估指标,具体见式(13)至式(15)。
${\rm{ MAE}}=\frac{1}{n}\sum _{i=1}^{n}\left|{F}_{i}-{O}_{i}\right|, $
$ {\rm{RMSE}}=\sqrt{\frac{1}{n}\sum _{i=1}^{n}{\left({F}_{i}-{O}_{i}\right)}^{2}}, $
$ {R}^{2}=1-\frac{{\displaystyle\sum _{i=1}^{n}}{\left({F}_{i}-{O}_{i}\right)}^{2}}{{\displaystyle\sum _{i=1}^{n}}{\left({O}_{i}-\stackrel—{O}\right)}^{2}}, $
式中,$ {F}_{i} $$ {O}_{i} $$ \overline{O} $分别代表第i个预测值、第i个实测值和所有实测值的平均值;n为预测天数。
基于神经网络算法的预测结果具有一定的随机性,用同样的数据训练同一个网络会得到不同的结果,本文将目标域LSTM模型、TLSTM模型(未微调)、TLSTM模型和DR-TLSTM模型重复运行100次,取相关评估指标的平均值和标准偏差,用于分析模型的预测性能和稳定性。
此外,为满足三沙湾水温预警预报业务化应用需求,依据大黄鱼的适宜生产温度[32],结合现场调查,将水温分为7个等级,见表3。利用三沙湾3个渔排基水质浮标(分别位于宁海、三都澳和霞浦)数据,对DR-TLSTM模型进行未来1~7 d水温预测效果的业务化测试,采用水温等级预测准确率(预测正确天数占比)进行模型的等级预测效果评估。
选取目标域LSTM模型、TLSTM模型(未微调)和TLSTM模型,进行未来1~7 d日最高水温和日最低水温的预测,测试结果见表4图2。与实测值相比,各模型的预测精度均随着预测天数的增加而降低。其中,目标域LSTM模型未来1~7 d水温预测值的RMSE为5.42~6.70℃,MAE为5.30~6.54℃,R2为0.653~0.858;TLSTM模型(未微调)未来1~7 d水温预测值的RMSE为0.31~0.97℃,MAE为0.28~0.80℃,R2为0.737~0.994;TLSTM模型未来1~7 d水温预测值的RMSE为0.28~0.88℃,MAE为0.25~0.81℃,R2为0.759~0.995。
以上结果表明,当源域数据量远大于目标域数据量时,TLSTM模型和TLSTM模型(未微调)的预测精度显著高于目标域LSTM模型,100次测试相关评估指标的标准差均显著低于目标域LSTM模型。已有研究表明,目标域LSTM模型由少量数据训练而成,容易出现过拟合现象[22]。本文基于厦门湾源域大量数据迁移学习的TLSTM模型和TLSTM模型(未微调),能较好地提取到水温数据的通用特征,其预测精度和稳定性均优于三沙湾目标域模型。
在两个迁移模型中,TLSTM模型的精度和稳定性略优于TLSTM模型(未微调),说明加入目标域数据对模型进行微调,有助于模型更充分地提取目标域数据的特异性特征,使其更适应目标域的任务,从而提升预测精度。也有研究指出当目标域数据量过少或模型参数数量过多时,微调可能会导致过拟合,使得模型性能变差[33]
选取TLSTM模型、纯差分回归模型、混差分回归模型和DR-TLSTM集成模型,进行未来1~7 d的日最高水温和日最低水温的预测,测试结果见表5图3。TLSTM模型未来1~7 d水温预测值的RMSE为0.28~0.88℃,MAE为0.25~0.81℃,R2为0.759~0.995;纯差分回归模型未来1~7 d水温预测值的RMSE为0.14~0.93℃,MAE为0.11~0.77℃,R2为0.775~0.996;混差分回归模型未来1~7 d水温预测值的RMSE为0.15~0.95℃,MAE为0.11~0.74℃,R2为0.793~0.995;DR-TLSTM集成模型未来1~7 d水温预测值的RMSE为0.13~0.77℃,MAE为0.11~0.71℃,R2为0.791~0.996。
以上结果表明,相较于3种单一模型,DR-TLSTM集成模型的预测精度最高,且100次测试相关评估指标的标准差低于TLSTM模型。考虑到TLSTM模型和回归模型在不同时刻的预测误差不同,集成模型每日采用变权算法对模型的组合权重进行更新,赋予最佳单一模型最大权重,使得预测精度和稳定性均得到有效提升。将TLSTM模型和回归模型进行集成应用,既能保留LSTM模型处理非线性数据的优势,又能发挥回归模型对预报气温数据和水温数据的拟合优势。在刘明等[34]和康俊锋等[35]的变权组合测试中,也发现组合模型的预测效果优于单一模型。
选取TLSTM模型、纯差分回归模型、混差分回归模型和DR-TLSTM集成模型,对水温突变点进行未来1~7 d的日最高水温和日最低水温的预测,测试结果见表6图4。TLSTM模型未来1~7 d水温突变点预测值的RMSE为0.29~1.33℃,MAE为0.39~1.13℃,R2为0.450~0.992;纯差分回归模型未来1~7 d水温突变点预测值的RMSE为0.32~1.31℃,MAE为0.30~1.07℃,R2为0.526~0.995;混差分回归模型未来1~7 d水温突变点预测值的RMSE为0.34~1.34℃,MAE为0.29~1.00℃,R2为0.529~0.996;DR-TLSTM集成模型未来1~7 d水温突变点预测值的RMSE为0.29~1.09℃,MAE为0.29~0.98℃,R2为0.530~0.996。
近年来,全球极端海温(如热浪、寒潮)的出现频次和强度均呈上升趋势[36],受气候变化和区域地理特征等因素的影响[37],现有的模型方法难以准确预测极端海温。海温突变(骤升和骤降)能较好地指示极端海温事件的出现,对海温突变点进行预测有助于进行海水养殖的应急防范。相较于单一模型,本文构建的DR-TLSTM集成模型对水温突变点的预测精度最高,且100次测试相关评估指标的标准差最低,说明该模型能更加稳定有效地预测水温突变趋势,可为极端海温灾害预警提供技术支撑。
模型每日得到包括当日在内的未来1~7 d水温预测值,以三沙湾2021年12月25日至2022年3月10日水温实测数据为研究对象,与TLSTM模型、纯差分回归模型、混差分回归模型和DR-TLSTM模型得到的第1天至第7天预测值进行对比,对比结果见图5
与已有的研究报道相比,本文构建的DR-TLSTM模型在目标域实测数据量较少的情况下,水温预测精度高于其他基于大量目标域数据构建的LSTM模型。例如,取各研究模型的第1天预测值的RMSE进行对比,本文构建的DR-TLSTM模型的RMSE可达0.13℃,而Hou等[16]构建的密集扩张卷积LSTM模型、Yang等[17]构建的全连接卷积LSTM模型和Kim等[19]构建的LSTM模型的RMSE分别为0.39℃、0.15℃和0.40℃。因此,DR-TLSTM模型的水温预测性能佳,对于监测资料较少的区域有良好的推广应用潜力。
模型集成于“福建省智慧渔业水质监测与预警系统(三沙湾)”,基于湾内3个渔排基在线监测水温数据和天气预报数据进行水温预测,通过GIS可视化提供水温等级预警预报信息服务(图6)。以水温等级预测准确率为评价指标,统计DR-TLSTM模型在2021年12月16日至2022年6月1日每日发布的未来1~7 d预警结果。表7表明,模型对未来1~3 d的水温等级预测准确率高于91%,未来1~7 d水温等级预测准确率高于79%,可为渔业生产和管理用户提供重要的信息参考。
本文提出了一种DR-TLSTM集成模型,通过厦门湾源域(数据多)和三沙湾目标域(数据少)之间的迁移学习,采用变权算法将纯差分回归模型、混差分回归模型和TLSTM模型集成为三沙湾DR-TLSTM模型,有效解决了海湾水温监测数据不足和随机扰动(水温骤升或骤降)预测困难的问题,显著提升了水温短期预测的效果,具有预测精度高、稳定性强的优点。该模型已集成于“福建省智慧渔业水质监测与预警系统(三沙湾)”,并于2021年9月起开始持续运行,每日提供三沙湾实时水质动态和未来7日的海温预警预报单,为渔业生产和管理提供技术支撑与信息服务。本文构建的DR-TLSTM集成模型对于缺乏历史监测数据区域的水温预测具有重要的参考价值,也有良好的推广应用潜力。
  • 福建省海洋经济发展补助资金项目(ZHHY-2019-1);国家重点研发计划(2016YFC0502901)
参考文献 引证文献
排序方式:
1
田梓杨, 黄洪辉, 齐占会, 等. 大鹏澳海湾养殖生态环境的化学计量分析[J]. 中国水产科学, 2012, 19(5): 881−888.
Tian Ziyang, Huang Honghui, Qi Zhanhui, et al. Using chemometric methods to study the mariculture eco-environment in Dapeng Cove, China[J]. Journal of Fishery Sciences of China, 2012, 19(5): 881−888.
2
程建新, 肖佳媚, 陈明茹, 等. 兴化湾海湾生态系统退化评价[J]. 厦门大学学报(自然科学版), 2012, 51(5): 944−950.
Cheng Jianxin, Xiao Jiamei, Chen Mingru, et al. Ecosystem degradation assessment of Xinghua Bay[J]. Journal of Xiamen University (Natural Science), 2012, 51(5): 944−950.
3
Rabalais N N. Human impacts on fisheries across the land-sea interface[J]. Proceedings of the National Academy of Sciences of the United States of America, 2015, 112(26): 7892−7893.
4
肖应凯, 刘卫国, 马云麒, 等. 鹿角珊瑚δ18O, Sr/Ca和Mg/Ca比值与海水表面温度的相关性研究——不同海水温度下的珊瑚养殖实验[J]. 中国科学: 地球科学, 2014, 57(5): 1048−1060.
Xiao Yingkai, Liu Weiguo, Ma Yunqi, et al. Correlation between δ18O, Sr/Ca and Mg/Ca of coral Acropora and seawater temperature from coral culture experiments[J]. Science China Earth Sciences, 2014, 57(5): 1048−1060.
5
胡石建, 李诗翰. 海洋热浪研究进展与展望[J]. 地球科学进展, 2022, 37(1): 51−64.
Hu Shijian, Li Shihan. Progress and prospect of marine heatwave study[J]. Advances in Earth Science, 2022, 37(1): 51−64.
6
Biswas B K, Svirezhev Y M, Bala B K. A model to predict climate-change impact on fish catch in the world oceans[J]. IEEE Transactions on Systems, Man, and Cybernetics-Part A: Systems and Humans, 2005, 35(6): 773−783.
7
Gianelli I, Ortega L, Marín Y, et al. Evidence of ocean warming in Uruguay's fisheries landings: the mean temperature of the catch approach[J]. Marine Ecology Progress Series, 2019, 625: 115−125.
8
Wang Wenyu, Zhou Chenghu, Shao Quanqin, et al. Remote sensing of sea surface temperature and chlorophyll-a: implications for squid fisheries in the north-west Pacific Ocean[J]. International Journal of Remote Sensing, 2010, 31(17/18): 4515−4530.
9
Wheeland L J, Morgan M J. Age-specific shifts in Greenland halibut (Reinhardtius hippoglossoides) distribution in response to changing ocean climate[J]. ICES Journal of Marine Science, 2020, 77(1): 230−240.
10
Zisserson B, Cook A. Impact of bottom water temperature change on the southernmost snow crab fishery in the Atlantic Ocean[J]. Fisheries Research, 2017, 195: 12−18.
11
卢峰本, 黄滢, 周启强, 等. 海水养殖的气象风险分析及预报[J]. 气象, 2006, 32(11): 113−117.
Lu Fengben, Huang Ying, Zhou Qiqiang, et al. Meteorological risk analysis and forecast to mariculture production[J]. Meteorological Monthly, 2006, 32(11): 113−117.
12
Xie Jiang, Zhang Jiyuan, Yu Jie, et al. An adaptive scale sea surface temperature predicting method based on deep learning with attention mechanism[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 17(5): 740−744.
13
Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735−1780.
14
刘建伟, 宋志妍. 循环神经网络研究综述[J]. 控制与决策, 2022, 37(11): 2753−2768.
Liu Jianwei, Song Zhiyan. Overview of recurrent neural networks[J]. Control and Decision, 2022, 37(11): 2753−2768.
15
Zhang Qin, Wang Hui, Dong Junyu, et al. Prediction of sea surface temperature using long short-term memory[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(10): 1745−1749.
16
Hou Siyun, Li Wen’gen, Liu Tianying, et al. D2CL: a dense dilated convolutional LSTM model for sea surface temperature prediction[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 12514−12523.
17
Yang Yuting, Dong Junyu, Sun Xin, et al. A CFCC-LSTM model for sea surface temperature prediction[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15(2): 207−211.
18
张昆. 基于深度学习的深海遥感技术研究[D]. 厦门: 厦门大学, 2019.
Zhang Kun. Research on deep-ocean remote sensing by deep learning[D]. Xiamen: Xiamen University, 2019.
19
Kim M, Yang H, Kim J. Sea surface temperature and high water temperature occurrence prediction using a long short-term memory model[J]. Remote Sensing, 2020, 12(21): 3654.
20
李嘉康, 李其杰, 赵颖, 等. 基于CEEMD-BP神经网络的海温异常预测研究[J]. 数学的实践与认识, 2017, 47(24): 163−171.
Li Jiakang, Li Qijie, Zhao Ying, et al. Sea surface temperature anomaly forecast research based on CEEMD-BP neural network[J]. Mathematics in Practice and Theory, 2017, 47(24): 163−171.
21
董璐, 周天军. 20世纪太平洋海温变化中人为因子与自然因子贡献的模拟研究[J]. 海洋学报, 2014, 36(3): 48−60.
Dong Lu, Zhou Tianjun. Contributions of natural and anthropogenic forcings to the twentieth-century Pacific sea surface temperature variability simulated by a climate system model[J]. Haiyang Xuebao, 2014, 36(3): 48−60.
22
田远洋, 徐显涛, 彭安帮, 等. 训练数据量对LSTM网络学习性能影响分析[J]. 水文, 2022, 42(1): 29−34, 22.
Tian Yuanyang, Xu Xiantao, Peng Anbang, et al. Effects of training data on the study performance of LSTM network[J]. Journal of China Hydrology, 2022, 42(1): 29−34, 22.
23
Zhuang Fuzhen, Qi Zhiyuan, Duan Keyu, et al. A comprehensive survey on transfer learning[J]. Proceedings of the IEEE, 2021, 109(1): 43−76.
24
黎英, 宋佩华. 迁移学习在医学图像分类中的研究进展[J]. 中国图象图形学报, 2022, 27(3): 672−686.
Li Ying, Song Peihua. Review of transfer learning in medical image classification[J]. Journal of Image and Graphics, 2022, 27(3): 672−686.
25
张晓, 李业刚, 王栋, 等. 基于迁移学习的社交评论命名实体识别[J]. 计算机应用与软件, 2022, 39(1): 143−150.
Zhang Xiao, Li Yegang, Wang Dong, et al. Social comment named entity recognition based on transfer learning[J]. Computer Applications and Software, 2022, 39(1): 143−150.
26
王义刚, 宋志尧, 姜翠玲, 等. 福建省海湾数模与环境研究: 三沙湾[M]. 北京: 海洋出版社, 2009.
Wang Yigang, Song Zhiyao, Jiang Cuiling, et al. Numerical Modeling and Environmental Study of Gulf in Fujian Province: Sansha Bay[M]. Beijing: China Ocean Press, 2009.
27
厦门市地方志编纂委员会. 厦门市志[M]. 北京: 方志出版社, 2004.
Xiamen Local Chronicle Compilation Committee. County Annals of Xiamen[M]. Beijing: Local Records Publishing House, 2004.
28
欧阳俊峰. 福建沿海滩涂微型真核生物群落结构及对生物修复的响应[D]. 厦门: 集美大学, 2016.
Ouyang Junfeng. Microeukaryotic community structure and response to bioremediation in intertidal mudflats of Fujian coast[D]. Xiamen: Jimei University, 2016.
29
王翠, 郭晓峰, 方婧, 等. 闽浙沿岸流扩展范围的季节特征及其对典型海湾的影响[J]. 应用海洋学学报, 2018, 37(1): 1−8.
Wang Cui, Guo Xiaofeng, Fang Jing, et al. Characteristics of seasonal spatial expansion of Fujian and Zhejiang Coastal Current and their bay effects[J]. Journal of Applied Oceanography, 2018, 37(1): 1−8.
30
Yu Yong, Si Xiaosheng, Hu Changhua, et al. A review of recurrent neural networks: LSTM cells and network architectures[J]. Neural Computation, 2019, 31(7): 1235−1270.
31
Pan S J, Yang Qiang. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345−1359.
32
福建省科学技术厅. 大黄鱼养殖[M]. 北京: 海洋出版社, 2004.
Province Department of Science and Technology of Fujian. Breeding and farming of Pseudosciaena Crocea[M]. Beijing: China Ocean Press, 2004.
33
Yosinski J, Clune J, Bengio Y, et al. How transferable are features in deep neural networks?[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal: MIT Press, 2014.
34
刘明, 王红蕾, 索良泽. 基于变权组合模型的中长期负荷概率密度预测[J]. 电力系统及其自动化学报, 2019, 31(7): 88−94.
Liu Ming, Wang Honglei, Suo Liangze. Medium- and long-term load probability density forecasting based on variable weight combination model[J]. Proceedings of the CSU-EPSA, 2019, 31(7): 88−94.
35
康俊锋, 谭建林, 方雷, 等. XGBoost-LSTM变权组合模型支持下短期PM2.5浓度预测—以上海为例[J]. 中国环境科学, 2021, 41(9): 4016−4025.
Kang Junfeng, Tan Jianlin, Fang Lei, et al. Short-term PM2.5 concentration prediction based on XGBoost and LSTM variable weight combination model: a case study of Shanghai[J]. China Environmental Science, 2021, 41(9): 4016−4025.
36
Benthuysen J A, Oliver E C J, Feng Ming, et al. Extreme marine warming across tropical Australia during austral summer 2015−2016[J]. Journal of Geophysical Research: Oceans, 2018, 123(2): 1301−1326.
37
齐庆华, 蔡榕硕. 中国近海海表温度变化的极端特性及其气候特征研究[J]. 海洋学报, 2019, 41(7): 36−51.
Qi Qinghua, Cai Rongshuo. Analysis on climate characteristics of sea surface temperature extremes in coastal China seas[J]. Haiyang Xuebao, 2019, 41(7): 36−51.
2023年第45卷第4期
PDF下载
152
62
引用本文
BibTeX
文章信息
doi: 10.12284/hyxb2023027
  • 接收时间:2022-07-06
  • 首发时间:2025-12-26
  • 出版时间:2023-03-31
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2022-07-06
  • 修回日期:2022-09-26
基金
福建省海洋经济发展补助资金项目(ZHHY-2019-1);国家重点研发计划(2016YFC0502901)
作者信息
    1 厦门大学 近海海洋环境科学国家重点实验室,福建 厦门 361102
    2 厦门大学 福建省海陆界面生态环境重点实验室,福建 厦门 361102
    3 福建省渔业资源监测中心,福建 福州 350003

通讯作者:

*陈能汪(1976-),男,教授,主要从事海陆界面生态环境研究。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/hyxb/CN/10.12284/hyxb2023027
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