Article(id=1194652707404354268, tenantId=1146029695717560320, journalId=1149651085930835976, issueId=1194652705147822651, articleNumber=null, orderNo=null, doi=10.12284/hyxb2025028, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1727452800000, receivedDateStr=2024-09-28, revisedDate=1734451200000, revisedDateStr=2024-12-18, acceptedDate=null, acceptedDateStr=null, onlineDate=1762757001019, onlineDateStr=2025-11-10, pubDate=1738252800000, pubDateStr=2025-01-31, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1762757001019, onlineIssueDateStr=2025-11-10, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1762757001019, creator=13701087609, updateTime=1762757001019, updator=13701087609, issue=Issue{id=1194652705147822651, tenantId=1146029695717560320, journalId=1149651085930835976, year='2025', volume='47', issue='1', pageStart='1', pageEnd='132', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1762757000481, creator=13701087609, updateTime=1762757000481, updator=13701087609, preIssue=null, nextIssue=null, ext=null, issueFiles=null}, startPage=104, endPage=116, ext={EN=ArticleExt(id=1194652707626652382, articleId=1194652707404354268, tenantId=1146029695717560320, journalId=1149651085930835976, language=EN, title=Prediction of seawater dissolved oxygen concentration based on multi-buoy spatial multi-feature fusion, columnId=1194652705852465724, journalTitle=Haiyang Xuebao, columnName=Article, runingTitle=null, highlight=null, articleAbstract=

Dissolved oxygen concentration is one of the important indexes to measure seawater quality. In order to grasp the change of seawater quality in time and reduce the risk and loss of seawater pollution, it is very important to establish the prediction mechanism of marine water quality parameters. Therefore, this paper proposes a prediction model of dissolved oxygen concentration in seawater based on temporal and spatial information fusion of buoy Networks and Generative Adversarial Networks (GAN), which aims to integrate topological information of buoy networks in the monitoring area and realize multi-feature fusion of buoy sensors. The model uses the Graph Attention Mechanism (GAT) to mine the influence of different nearest neighbor points on the target node and calculate the weights of the adjacent nodes, so as to capture the spatio-temporal characteristics of the buoy data. The two-head attention mechanism and the two-time-scale Update Rule (TTUR) were used to optimize the GAN prediction network and the network training process, improve the training speed balance of the generated adversarial network, and improve the fitting effect of the generator network. The mean squared error, root mean squared error, mean absolute error and R-Square are used as evaluation indexes to compare the model prediction performance. The results show that the evaluation indexes of the proposed model are superior to other models, and can effectively mine the spatial information of multiple buoys. It overcomes the shortcomings of traditional methods in the prediction of dissolved oxygen concentration in seawater, such as low accuracy, inability to flexibly use historical spatial data, poor training stability and slow speed, and can provide important technical support for marine water quality monitoring and prediction.

, correspAuthors=Ying Chen, authorNote=null, correspAuthorsNote=null, copyrightStatement=Haiyang Xuebao, 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=Qiguang Zhu, Zhen Shen, Xiang Li, Zhen Wei, Wenjing Qiao, Linsong Zhang, Ying Chen), CN=ArticleExt(id=1194653015459205357, articleId=1194652707404354268, tenantId=1146029695717560320, journalId=1149651085930835976, language=CN, title=基于多浮标空间多特征融合的海水溶解氧浓度预测, columnId=1149698756456657529, journalTitle=海洋学报, columnName=论文, runingTitle=null, highlight=null, articleAbstract=

溶解氧浓度是衡量海水水质的重要指标之一。为了及时掌握海水水质变化情况,降低海水污染风险及其带来的损失,建立海洋水质参数预测机制至关重要。为此,本文提出了一种基于浮标网络时空信息融合和改进生成对抗网络(Generative Adversarial Networks, GAN)的海水溶解氧浓度预测模型,旨在整合监测区域内浮标网络的拓扑信息并实现浮标传感器的多特征融合。该模型利用图注意力网络(Graph Attention Mechanism,GAT)挖掘不同近邻点对目标节点的影响,计算邻接节点的权重,从而捕获浮标数据的时空特征;通过双头注意力机制与双时间尺度更新规则(Two Time-Scale Update Rule, TTUR)优化GAN预测网络及网络训练过程,改善生成对抗网络的训练速度平衡问题,提高生成器网络的拟合效果。以均方误差、均方根误差、平均绝对误差与决定系数为评价指标进行模型预测性能对比,结果表明,所提出模型的各项评价指标均优于其他模型,能够有效挖掘多浮标的空间信息,克服了传统方法在海水溶解氧浓度预测中存在的精度低、无法灵活利用历史空间数据、训练稳定性差和速度慢等不足,可为海洋水质监测及预测提供重要的技术支撑。

, correspAuthors=陈颖, authorNote=null, correspAuthorsNote=
*陈颖,教授,主要从事海洋生态环境监测与大数据分析的研究。E-mail:
, copyrightStatement=版权所有©《海洋学报》编辑部 2025, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=CXgZgU9EFnyd+QAqB9xyXQ==, magXml=+z8IGAYxy8n205drjJUG3g==, pdfUrl=null, pdf=TYCITp2VmPrtcpSBZw/LLw==, pdfFileSize=6305597, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=KraZXaLi2I0YG/rKXAdJHQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=C8YKsPR51AMPqVpF0rBtaw==, mapNumber=null, authorCompany=null, fund=null, authors=

朱奇光(1978—),男,浙江省宁波市人,副教授,主要从事多传感器信息融合与大数据分析方面的研究。E-mail:

, authorsList=朱奇光, 申震, 李享, 魏祯, 乔文静, 张淋淞, 陈颖)}, authors=[Author(id=1194975389605938001, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=zhu7880@ysu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1194975389706601300, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975389605938001, language=EN, stringName=Qiguang Zhu, firstName=Qiguang, middleName=null, lastName=Zhu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194975389832430421, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975389605938001, language=CN, stringName=朱奇光, firstName=奇光, middleName=null, lastName=朱, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004, bio={"content":"

朱奇光(1978—),男,浙江省宁波市人,副教授,主要从事多传感器信息融合与大数据分析方面的研究。E-mail:

"}, bioImg=null, bioContent=

朱奇光(1978—),男,浙江省宁波市人,副教授,主要从事多传感器信息融合与大数据分析方面的研究。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194975389413000009, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, xref=null, ext=[AuthorCompanyExt(id=1194975389421388618, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China), AuthorCompanyExt(id=1194975389429777227, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004)])]), Author(id=1194975389937288023, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, 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=1194975390033757017, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975389937288023, language=EN, stringName=Zhen Shen, firstName=Zhen, middleName=null, lastName=Shen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194975390126031706, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975389937288023, language=CN, stringName=申震, firstName=震, middleName=null, lastName=申, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194975389413000009, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, xref=null, ext=[AuthorCompanyExt(id=1194975389421388618, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China), AuthorCompanyExt(id=1194975389429777227, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004)])]), Author(id=1194975390188946268, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, 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=1194975390264443742, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975390188946268, language=EN, stringName=Xiang Li, firstName=Xiang, middleName=null, lastName=Li, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194975390360912735, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975390188946268, language=CN, stringName=李享, firstName=享, middleName=null, lastName=李, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194975389413000009, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, xref=null, ext=[AuthorCompanyExt(id=1194975389421388618, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China), AuthorCompanyExt(id=1194975389429777227, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004)])]), Author(id=1194975390490936161, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, 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=1194975390553850723, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975390490936161, language=EN, stringName=Zhen Wei, firstName=Zhen, middleName=null, lastName=Wei, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2. Key Laboratory of Measurement Technology and Instrument of Hebei Province, School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194975390629348196, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975390490936161, language=CN, stringName=魏祯, firstName=祯, middleName=null, lastName=魏, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.燕山大学电气工程学院 河北省测试计量技术及仪器重点实验室,河北 秦皇岛 066004, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194975389484303180, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, xref=null, ext=[AuthorCompanyExt(id=1194975389496886093, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389484303180, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. Key Laboratory of Measurement Technology and Instrument of Hebei Province, School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, China), AuthorCompanyExt(id=1194975389505274702, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389484303180, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.燕山大学电气工程学院 河北省测试计量技术及仪器重点实验室,河北 秦皇岛 066004)])]), Author(id=1194975390688068454, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, orderNo=4, 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=1194975390767760232, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975390688068454, language=EN, stringName=Wenjing Qiao, firstName=Wenjing, middleName=null, lastName=Qiao, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194975390839063401, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975390688068454, language=CN, stringName=乔文静, firstName=文静, middleName=null, lastName=乔, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194975389413000009, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, xref=null, ext=[AuthorCompanyExt(id=1194975389421388618, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China), AuthorCompanyExt(id=1194975389429777227, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004)])]), Author(id=1194975390927143787, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, orderNo=5, 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=1194975391006835565, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975390927143787, language=EN, stringName=Linsong Zhang, firstName=Linsong, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194975391065555822, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975390927143787, language=CN, stringName=张淋淞, firstName=淋淞, middleName=null, lastName=张, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194975389413000009, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, xref=null, ext=[AuthorCompanyExt(id=1194975389421388618, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China), AuthorCompanyExt(id=1194975389429777227, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004)])]), Author(id=1194975391145247600, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, orderNo=6, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=chenying@ysu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1194975391212356466, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975391145247600, language=EN, stringName=Ying Chen, firstName=Ying, middleName=null, lastName=Chen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, *, address=2. Key Laboratory of Measurement Technology and Instrument of Hebei Province, School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1194975391271076723, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, authorId=1194975391145247600, language=CN, stringName=陈颖, firstName=颖, middleName=null, lastName=陈, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, *, address=2.燕山大学电气工程学院 河北省测试计量技术及仪器重点实验室,河北 秦皇岛 066004, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1194975389484303180, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, xref=null, ext=[AuthorCompanyExt(id=1194975389496886093, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389484303180, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. Key Laboratory of Measurement Technology and Instrument of Hebei Province, School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, China), AuthorCompanyExt(id=1194975389505274702, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389484303180, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.燕山大学电气工程学院 河北省测试计量技术及仪器重点实验室,河北 秦皇岛 066004)])])], keywords=[Keyword(id=1194975391401100148, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, orderNo=1, keyword=prediction of dissolved oxygen concentration), Keyword(id=1194975391464014709, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, orderNo=2, keyword=spatial multi-feature fusion), Keyword(id=1194975391526929270, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, orderNo=3, keyword=Graph Attention Mechanism), Keyword(id=1194975391602426743, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, orderNo=4, keyword=Generative Adversarial Networks), Keyword(id=1194975391686312824, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, orderNo=5, keyword=Two Time-Scale Update Rule), Keyword(id=1194975391745033081, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, orderNo=1, keyword=溶解氧浓度预测), Keyword(id=1194975391803753338, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, orderNo=2, keyword=空间多特征融合), Keyword(id=1194975391883445115, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, orderNo=3, keyword=GAT), Keyword(id=1194975391963136892, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, orderNo=4, keyword=GAN), Keyword(id=1194975392026051453, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, orderNo=5, keyword=TTUR)], refs=[Reference(id=1194975394433581980, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, 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=齐庆华, 蔡榕硕, 颜秀花. 气候变化与我国海洋灾害风险治理探讨[J]. 海洋通报, 2019, 38(4): 361−367., articleTitle=null, refAbstract=null), Reference(id=1194975394509079453, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, 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=Qi Qinghua, Cai Rongshuo, Yan Xiuhua. Discussion on climate change and marine disaster risk governance in the coastal China seas[J]. Marine Science Bulletin, 2019, 38(4): 361−367., articleTitle=null, refAbstract=null), Reference(id=1194975394571994014, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, 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=Kisi O, Alizamir M, Gorgij A D. Dissolved oxygen prediction using a new ensemble method[J]. Environmental Science and Pollution Research, 2020, 27(9): 9589−9603., articleTitle=null, refAbstract=null), Reference(id=1194975394634908575, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Yanni, Yan Shiqiang, Shi Hongda, et al. Hydrodynamic analysis of a novel multi-buoy wind-wave energy system[J]. Renewable Energy, 2023, 219: 119477., articleTitle=null, refAbstract=null), Reference(id=1194975394710406048, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Shixuan, Song Miaomiao, Chen Shizhe, et al. An intelligent modeling framework to optimize the spatial layout of ocean moored buoy observing networks[J]. Frontiers in Marine Science, 2023, 10: 1134418., articleTitle=null, refAbstract=null), Reference(id=1194975394777514913, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=Jin Junchen, Rong Dingding, Zhang Tong, et al. A GAN-based short-term link traffic prediction approach for urban road networks under a parallel learning framework[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(9): 16185−16196., articleTitle=null, refAbstract=null), Reference(id=1194975394848818082, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=杜先君, 李河. 基于集成GCN-Transformer网络的ENSO预测模型[J]. 海洋学报, 2023, 45(12): 156−165., articleTitle=null, refAbstract=null), Reference(id=1194975394915926947, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=Du Xianjun, Li He. ENSO prediction model based on integrated GCN-Transformer network[J]. Haiyang Xuebao, 2023, 45(12): 156−165., articleTitle=null, refAbstract=null), Reference(id=1194975394991424420, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=Veličković P, Cucurull G, Casanova A, et al. Graph Attention Networks[C]//International Conference on Learning Representations. 2018., articleTitle=null, refAbstract=null), Reference(id=1194975395050144677, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Wenjing, Fang Huaiyang, Qin Guangxiong, et al. Concentration estimation of dissolved oxygen in Pearl River Basin using input variable selection and machine learning techniques[J]. Science of the Total Environment, 2020, 731: 139099., articleTitle=null, refAbstract=null), Reference(id=1194975395129836454, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=Wu Junhao, Wang Zhaocai, Dong Jinghan, et al. Multi-step ahead dissolved oxygen concentration prediction based on knowledge guided ensemble learning and explainable artificial intelligence[J]. Journal of Hydrology, 2024, 636: 131297., articleTitle=null, refAbstract=null), Reference(id=1194975395201139623, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Yeqi, Gong Chuanyang, Yang Ling, et al. DSTP-RNN: a dual-stage two-phase attention-based recurrent neural network for long-term and multivariate time series prediction[J]. Expert Systems with Applications, 2020, 143: 113082., articleTitle=null, refAbstract=null), Reference(id=1194975395268248488, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=Yue Weihao, Xu Yongsheng, Xiang Liang, et al. Prediction of 3-D ocean temperature based on self-attention and predictive RNN[J]. IEEE Geoscience and Remote Sensing Letters, 2024, 21: 1501305., articleTitle=null, refAbstract=null), Reference(id=1194975395335357353, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=石绥祥, 王蕾, 余璇, 等. 长短期记忆神经网络在叶绿素a浓度预测中的应用[J]. 海洋学报, 2020, 42(2): 134−142., articleTitle=null, refAbstract=null), Reference(id=1194975395410854826, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=Shi Suixiang, Wang Lei, Yu Xuan, et al. Application of long term and short term memory neural network in prediction of chlorophyll a concentration[J]. Haiyang Xuebao, 2020, 42(2): 134−142., articleTitle=null, refAbstract=null), Reference(id=1194975395486352299, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=Jia Xiaoyan, Ji Qiyan, Han Lei, et al. Prediction of sea surface temperature in the East China Sea based on LSTM neural network[J]. Remote Sensing, 2022, 14(14): 3300., articleTitle=null, refAbstract=null), Reference(id=1194975395570238380, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=张宇, 许大志, 俞胜宾, 等. 基于多尺度深度学习对南海海表温度预报的研究[J]. 海洋学报, 2024, 46(5): 27−36., articleTitle=null, refAbstract=null), Reference(id=1194975395645735853, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Yu, Xu Dazhi, Yu Shengbin, et al. Forecast of sea surface temperature in the South China Sea based on multi-scale deep learning model[J]. Haiyang Xuebao, 2024, 46(5): 27−36., articleTitle=null, refAbstract=null), Reference(id=1194975395725427630, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=Goodfellow I J, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal: MIT Press, 2014: 2672−2680., articleTitle=null, refAbstract=null), Reference(id=1194975395784147887, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks[J]. Communications of the ACM, 2020, 63(11): 139−144., articleTitle=null, refAbstract=null), Reference(id=1194975395880616880, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=Deng Jiaxiang, Deng Yong, Cheong K H. Combining conflicting evidence based on Pearson correlation coefficient and weighted graph[J]. International Journal of Intelligent Systems, 2021, 36(12): 7443−7460., articleTitle=null, refAbstract=null), Reference(id=1194975395960308657, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=Ibrahim A, Ismail A, Juahir H, et al. Water quality modelling using principal component analysis and artificial neural network[J]. Marine Pollution Bulletin, 2023, 187: 114493., articleTitle=null, refAbstract=null), Reference(id=1194975396073554866, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=Mai Gengchen, Xuan Yao, Zuo Wenyun, et al. Sphere2Vec: a general-purpose location representation learning over a spherical surface for large-scale geospatial predictions[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2023, 202: 439−462., articleTitle=null, refAbstract=null), Reference(id=1194975396161635251, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=Scarselli F, Gori M, Tsoi A C, et al. The graph neural network model[J]. IEEE Transactions on Neural Networks, 2009, 20(1): 61−80., articleTitle=null, refAbstract=null), Reference(id=1194975396232938420, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Xiaoqi, Yen K, Hu Yifan, et al. DeepGD: a deep learning framework for graph drawing using GNN[J]. IEEE Computer Graphics and Applications, 2021, 41(5): 32−44., articleTitle=null, refAbstract=null), Reference(id=1194975397289903029, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=22, rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Dujuan, Zhu Jiacheng, Yin Yunqiang, et al. Dynamic travel time prediction with spatiotemporal features: using a GNN-based deep learning method[J]. Annals of Operations Research, 2024, 340(1): 571−591., articleTitle=null, refAbstract=null), Reference(id=1194975397365400502, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks[C]//International Conference on Learning Representations, 2017., articleTitle=null, refAbstract=null), Reference(id=1194975397432509367, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=Derr T, Ma Yao, Tang Jiliang. Signed graph convolutional networks[C]//2018 IEEE International Conference on Data Mining (ICDM). Singapore: IEEE, 2018: 929−934., articleTitle=null, refAbstract=null), Reference(id=1194975397495423928, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=25, rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=Polamuri S R, Srinivas D K, Mohan D A K. Multi-model generative adversarial network hybrid prediction algorithm (MMGAN-HPA) for stock market prices prediction[J]. Journal of King Saud University-Computer and Information Sciences, 2022, 34(9): 7433−7444., articleTitle=null, refAbstract=null), Reference(id=1194975397562532793, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Jianguang, Zhang Xuyang, Yang Jianfeng, et al. Deep LSTM and GAN based short-term load forecasting method at the zone level[C]//2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC). Fukuoka: IEEE, 2020: 613−618., articleTitle=null, refAbstract=null), Reference(id=1194975397642224570, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=Festag S, Spreckelsen C. Medical multivariate time series imputation and forecasting based on a recurrent conditional Wasserstein GAN and attention[J]. Journal of Biomedical Informatics, 2023, 139: 104320., articleTitle=null, refAbstract=null), Reference(id=1194975397705139131, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=Gulrajani I, Ahmed F, Arjovsky M, et al. Improved training of wasserstein GANS[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates Inc. , 2017: 5769−5779., articleTitle=null, refAbstract=null), Reference(id=1194975397805802428, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Chenhan, Yu J J Q, Liu Yi. Spatial-temporal graph attention networks: a deep learning approach for traffic forecasting[J]. IEEE Access, 2019, 7: 166246−166256., articleTitle=null, refAbstract=null), Reference(id=1194975397872911293, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=30, rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=Fang Fang, Zhang Pengpeng, Zhou Bo, et al. Atten-GAN: pedestrian trajectory prediction with GAN based on attention mechanism[J]. Cognitive Computation, 2022, 14(6): 2296−2305., articleTitle=null, refAbstract=null), Reference(id=1194975397948408766, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=31, rfOrder=34, authorNames=null, journalName=null, refType=null, unstructuredReference=Bynagari N B. GANs trained by a two time-scale update rule converge to a local Nash equilibrium[J]. Asian Journal of Applied Science and Engineering, 2019, 8(1): 25−34., articleTitle=null, refAbstract=null), Reference(id=1194975398011323327, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=32, rfOrder=35, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Kang, Zhong Guoqiang, Dong Junyu, et al. Stock market prediction based on generative adversarial network[J]. Procedia Computer Science, 2019, 147: 400−406., articleTitle=null, refAbstract=null), Reference(id=1194975398091015104, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=33, rfOrder=36, authorNames=null, journalName=null, refType=null, unstructuredReference=Zha Wenshu, Liu Yuping, Wan Yujin, et al. Forecasting monthly gas field production based on the CNN-LSTM model[J]. Energy, 2022, 260: 124889., articleTitle=null, refAbstract=null), Reference(id=1194975398174901185, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=34, rfOrder=37, authorNames=null, journalName=null, refType=null, unstructuredReference=Yu Jingxin, Zhang Xin, Xu Linlin, et al. A hybrid CNN-GRU model for predicting soil moisture in maize root zone[J]. Agricultural Water Management, 2021, 245: 106649., articleTitle=null, refAbstract=null)], funds=[Fund(id=1194975394089649048, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, awardId=62275228, language=CN, fundingSource=国家自然科学基金(62275228), fundOrder=null, country=null), Fund(id=1194975394165146521, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, awardId=19273901D, language=CN, fundingSource=河北省重点研发计划(19273901D), fundOrder=null, country=null), Fund(id=1194975394223866778, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, awardId=20373301D, language=CN, fundingSource=河北省重点研发计划(20373301D), fundOrder=null, country=null), Fund(id=1194975394290975643, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, awardId=D2024203002, language=CN, fundingSource=河北省自然科学基金(D2024203002), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1194975389413000009, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, xref=null, ext=[AuthorCompanyExt(id=1194975389421388618, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China), AuthorCompanyExt(id=1194975389429777227, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389413000009, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004)]), AuthorCompany(id=1194975389484303180, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, xref=null, ext=[AuthorCompanyExt(id=1194975389496886093, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389484303180, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2. Key Laboratory of Measurement Technology and Instrument of Hebei Province, School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, China), AuthorCompanyExt(id=1194975389505274702, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, companyId=1194975389484303180, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.燕山大学电气工程学院 河北省测试计量技术及仪器重点实验室,河北 秦皇岛 066004)])], figs=[ArticleFig(id=1194975392185435006, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Fig. 1, caption=Geographic location of buoy, figureFileSmall=sgWRwiMgfSQDUoK1BHfhTw==, figureFileBig=zcdVAELVuWnAYAs5uVzkfA==, tableContent=null), ArticleFig(id=1194975392256738175, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=图1, caption=浮标地理位置, figureFileSmall=sgWRwiMgfSQDUoK1BHfhTw==, figureFileBig=zcdVAELVuWnAYAs5uVzkfA==, tableContent=null), ArticleFig(id=1194975392332235648, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Fig. 2, caption=Dissolved oxygen data measured by each buoy, figureFileSmall=PktZqJHvPnhOUf0MHhKBZQ==, figureFileBig=/UtP/gaeNz81tBv6u+dU1Q==, tableContent=null), ArticleFig(id=1194975392395150209, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=图2, caption=各浮标测得的溶解氧数据, figureFileSmall=PktZqJHvPnhOUf0MHhKBZQ==, figureFileBig=/UtP/gaeNz81tBv6u+dU1Q==, tableContent=null), ArticleFig(id=1194975392466453378, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Fig. 3, caption=Correlation analysis results between parameters of various buoys, figureFileSmall=KVWJu3bI9Wq+uDip2XVgAA==, figureFileBig=MKC31qt8S47IDIbXqlQs7Q==, tableContent=null), ArticleFig(id=1194975392541950851, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=图3, caption=各浮标参数间相关性分析结果, figureFileSmall=KVWJu3bI9Wq+uDip2XVgAA==, figureFileBig=MKC31qt8S47IDIbXqlQs7Q==, tableContent=null), ArticleFig(id=1194975392613254020, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Fig. 4, caption=The feature vector obtained by fusing the features of each buoy, figureFileSmall=ZLPotisyTsBeQqJPtD1GRg==, figureFileBig=vszh5VkkKpVVJPjsZ3Cghg==, tableContent=null), ArticleFig(id=1194975392667779973, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=图4, caption=各浮标特征融合后的特征向量, figureFileSmall=ZLPotisyTsBeQqJPtD1GRg==, figureFileBig=vszh5VkkKpVVJPjsZ3Cghg==, tableContent=null), ArticleFig(id=1194975392730694534, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Fig. 5, caption=Topological relationship between buoys, figureFileSmall=julGCd+KTaAZXlUCMV56sA==, figureFileBig=zlS/APGpVmvLzgSd78Ts6A==, tableContent=null), ArticleFig(id=1194975392801997703, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=图5, caption=浮标间的拓扑关系, figureFileSmall=julGCd+KTaAZXlUCMV56sA==, figureFileBig=zlS/APGpVmvLzgSd78Ts6A==, tableContent=null), ArticleFig(id=1194975392873300872, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Fig. 6, caption=GAT module structure, figureFileSmall=jfUV56QbsfDU8CKc+cB+Ng==, figureFileBig=7zd4HVqdMt9ShpydDJfSqA==, tableContent=null), ArticleFig(id=1194975392952992649, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=图6, caption=GAT模块结构, figureFileSmall=jfUV56QbsfDU8CKc+cB+Ng==, figureFileBig=7zd4HVqdMt9ShpydDJfSqA==, tableContent=null), ArticleFig(id=1194975393041073034, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Fig. 7, caption=Adjacency matrix after GAT adaptive spatial feature extraction, figureFileSmall=U9lZlwzIK35J38X0bVR6Jg==, figureFileBig=qiKsL5p4pW6IE36e0By5IA==, tableContent=null), ArticleFig(id=1194975393099793291, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=图7, caption=经GAT自适应空间特征提取后的邻接矩阵, figureFileSmall=U9lZlwzIK35J38X0bVR6Jg==, figureFileBig=qiKsL5p4pW6IE36e0By5IA==, tableContent=null), ArticleFig(id=1194975393166902156, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Fig. 8, caption=SA-WGAN_GP model, figureFileSmall=V3vpy81GTydCtogVQU/QzQ==, figureFileBig=DE1EhXltKynFq3UTAUi2vQ==, tableContent=null), ArticleFig(id=1194975393250788237, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=图8, caption=SA-WGAN_GP模型, figureFileSmall=V3vpy81GTydCtogVQU/QzQ==, figureFileBig=DE1EhXltKynFq3UTAUi2vQ==, tableContent=null), ArticleFig(id=1194975393305314190, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Fig. 9, caption=The predictive performance of different prediction models, figureFileSmall=dvMgCe8TprJ4SyK2rFy6Qw==, figureFileBig=W2zJyrPDqPU7hFUSALPK5A==, tableContent=null), ArticleFig(id=1194975393376617359, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=图9, caption=不同预测模型的预测性能, figureFileSmall=dvMgCe8TprJ4SyK2rFy6Qw==, figureFileBig=W2zJyrPDqPU7hFUSALPK5A==, tableContent=null), ArticleFig(id=1194975393431143312, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Table 1, caption=

Monitoring parameters of buoys equipped with sensors

, figureFileSmall=null, figureFileBig=null, tableContent=
参数英文符号单位
温度Temp
电导率CondmS/cm
盐度Sal
溶解氧质量浓度DOmg/L
酸碱度pH1
浊度TurbNTU
叶绿素Chlμg/L
藻红蛋白PEμg/L
), ArticleFig(id=1194975393519223697, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=表1, caption=

浮标搭载传感器监测参数

, figureFileSmall=null, figureFileBig=null, tableContent=
参数英文符号单位
温度Temp
电导率CondmS/cm
盐度Sal
溶解氧质量浓度DOmg/L
酸碱度pH1
浊度TurbNTU
叶绿素Chlμg/L
藻红蛋白PEμg/L
), ArticleFig(id=1194975393586332562, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Table 2, caption=

Features selected on different buoys

, figureFileSmall=null, figureFileBig=null, tableContent=
浮标参数1参数2参数3参数4参数5
GX-02ChlpHTempCondSal
GX-14pHTempChlCondSal
GX-15TempChlPECondSal
GX-17TempCondpHSalTurb
GX-03pHCondChlTempPE
GX-19TempCondpHSalPE
), ArticleFig(id=1194975393661830035, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=表2, caption=

不同浮标上所选择的特征

, figureFileSmall=null, figureFileBig=null, tableContent=
浮标参数1参数2参数3参数4参数5
GX-02ChlpHTempCondSal
GX-14pHTempChlCondSal
GX-15TempChlPECondSal
GX-17TempCondpHSalTurb
GX-03pHCondChlTempPE
GX-19TempCondpHSalPE
), ArticleFig(id=1194975393728938900, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Table 3, caption=

Distance relationship between buoys (distance/km)

, figureFileSmall=null, figureFileBig=null, tableContent=
浮标GX-02GX-14GX-15GX-17GX-03GX-19
GX-0203.3451.8059.2261.1671.48
GX-143.34048.7356.2258.1068.36
GX-1551.8048.7307.989.3919.80
GX-1759.2256.227.9803.0313.61
GX-0361.1658.109.393.03010.88
GX-1971.4868.3619.8013.6110.880
), ArticleFig(id=1194975393804436373, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=表3, caption=

浮标间的距离关系 (距离/km)

, figureFileSmall=null, figureFileBig=null, tableContent=
浮标GX-02GX-14GX-15GX-17GX-03GX-19
GX-0203.3451.8059.2261.1671.48
GX-143.34048.7356.2258.1068.36
GX-1551.8048.7307.989.3919.80
GX-1759.2256.227.9803.0313.61
GX-0361.1658.109.393.03010.88
GX-1971.4868.3619.8013.6110.880
), ArticleFig(id=1194975393879933846, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=EN, label=Table 4, caption=

Comparison of evaluation indicators for different models

, figureFileSmall=null, figureFileBig=null, tableContent=
模型MSERMSEMAE$ {{{R}}}^{2} $
CNN-LSTM0.5960.7720.5020.138
CNN-GRU0.4070.6370.2920.312
GCN-WGAN_GP0.3130.5590.2780.639
GAT-WGAN_GP0.1980.4440.1240.804
), ArticleFig(id=1194975393959625623, tenantId=1146029695717560320, journalId=1149651085930835976, articleId=1194652707404354268, language=CN, label=表4, caption=

不同模型的评价指标对比

, figureFileSmall=null, figureFileBig=null, tableContent=
模型MSERMSEMAE$ {{{R}}}^{2} $
CNN-LSTM0.5960.7720.5020.138
CNN-GRU0.4070.6370.2920.312
GCN-WGAN_GP0.3130.5590.2780.639
GAT-WGAN_GP0.1980.4440.1240.804
)], 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/hyxb2025028, detailUrlEn=https://castjournals.cast.org.cn/joweb/hyxb/EN/10.12284/hyxb2025028, pdfUrlCn=https://castjournals.cast.org.cn/joweb/hyxb/CN/PDF/10.12284/hyxb2025028, pdfUrlEn=https://castjournals.cast.org.cn/joweb/hyxb/EN/PDF/10.12284/hyxb2025028, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于多浮标空间多特征融合的海水溶解氧浓度预测
收藏切换
PDF下载
朱奇光 1 , 申震 1 , 李享 1 , 魏祯 2 , 乔文静 1 , 张淋淞 1 , 陈颖 2, *
海洋学报 | 论文 2025,47(1): 104-116
收起
收藏切换
海洋学报 | 论文 2025, 47(1): 104-116
基于多浮标空间多特征融合的海水溶解氧浓度预测
全屏
朱奇光1 , 申震1, 李享1, 魏祯2, 乔文静1, 张淋淞1, 陈颖2, *
作者信息
  • 1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004
  • 2.燕山大学电气工程学院 河北省测试计量技术及仪器重点实验室,河北 秦皇岛 066004
  • 朱奇光(1978—),男,浙江省宁波市人,副教授,主要从事多传感器信息融合与大数据分析方面的研究。E-mail:

通讯作者:

*陈颖,教授,主要从事海洋生态环境监测与大数据分析的研究。E-mail:
Prediction of seawater dissolved oxygen concentration based on multi-buoy spatial multi-feature fusion
Qiguang Zhu1 , Zhen Shen1, Xiang Li1, Zhen Wei2, Wenjing Qiao1, Linsong Zhang1, Ying Chen2, *
Affiliations
  • 1. Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, School of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China
  • 2. Key Laboratory of Measurement Technology and Instrument of Hebei Province, School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, China
出版时间: 2025-01-31 doi: 10.12284/hyxb2025028
文章导航
收藏切换

溶解氧浓度是衡量海水水质的重要指标之一。为了及时掌握海水水质变化情况,降低海水污染风险及其带来的损失,建立海洋水质参数预测机制至关重要。为此,本文提出了一种基于浮标网络时空信息融合和改进生成对抗网络(Generative Adversarial Networks, GAN)的海水溶解氧浓度预测模型,旨在整合监测区域内浮标网络的拓扑信息并实现浮标传感器的多特征融合。该模型利用图注意力网络(Graph Attention Mechanism,GAT)挖掘不同近邻点对目标节点的影响,计算邻接节点的权重,从而捕获浮标数据的时空特征;通过双头注意力机制与双时间尺度更新规则(Two Time-Scale Update Rule, TTUR)优化GAN预测网络及网络训练过程,改善生成对抗网络的训练速度平衡问题,提高生成器网络的拟合效果。以均方误差、均方根误差、平均绝对误差与决定系数为评价指标进行模型预测性能对比,结果表明,所提出模型的各项评价指标均优于其他模型,能够有效挖掘多浮标的空间信息,克服了传统方法在海水溶解氧浓度预测中存在的精度低、无法灵活利用历史空间数据、训练稳定性差和速度慢等不足,可为海洋水质监测及预测提供重要的技术支撑。

溶解氧浓度预测  /  空间多特征融合  /  GAT  /  GAN  /  TTUR

Dissolved oxygen concentration is one of the important indexes to measure seawater quality. In order to grasp the change of seawater quality in time and reduce the risk and loss of seawater pollution, it is very important to establish the prediction mechanism of marine water quality parameters. Therefore, this paper proposes a prediction model of dissolved oxygen concentration in seawater based on temporal and spatial information fusion of buoy Networks and Generative Adversarial Networks (GAN), which aims to integrate topological information of buoy networks in the monitoring area and realize multi-feature fusion of buoy sensors. The model uses the Graph Attention Mechanism (GAT) to mine the influence of different nearest neighbor points on the target node and calculate the weights of the adjacent nodes, so as to capture the spatio-temporal characteristics of the buoy data. The two-head attention mechanism and the two-time-scale Update Rule (TTUR) were used to optimize the GAN prediction network and the network training process, improve the training speed balance of the generated adversarial network, and improve the fitting effect of the generator network. The mean squared error, root mean squared error, mean absolute error and R-Square are used as evaluation indexes to compare the model prediction performance. The results show that the evaluation indexes of the proposed model are superior to other models, and can effectively mine the spatial information of multiple buoys. It overcomes the shortcomings of traditional methods in the prediction of dissolved oxygen concentration in seawater, such as low accuracy, inability to flexibly use historical spatial data, poor training stability and slow speed, and can provide important technical support for marine water quality monitoring and prediction.

prediction of dissolved oxygen concentration  /  spatial multi-feature fusion  /  Graph Attention Mechanism  /  Generative Adversarial Networks  /  Two Time-Scale Update Rule
朱奇光, 申震, 李享, 魏祯, 乔文静, 张淋淞, 陈颖. 基于多浮标空间多特征融合的海水溶解氧浓度预测. 海洋学报, 2025 , 47 (1) : 104 -116 . DOI: 10.12284/hyxb2025028
Qiguang Zhu, Zhen Shen, Xiang Li, Zhen Wei, Wenjing Qiao, Linsong Zhang, Ying Chen. Prediction of seawater dissolved oxygen concentration based on multi-buoy spatial multi-feature fusion[J]. Haiyang Xuebao, 2025 , 47 (1) : 104 -116 . DOI: 10.12284/hyxb2025028
及时了解海洋生态系统的动态变化,深入挖掘海洋环境数据并建立海洋水质参数的数据预测分析机制,并在污染发生的萌芽期进行有效防治,可以有效降低因海水污染带来的损失,对保护海洋生态系统和人类健康具有重要意义[1]
溶解氧(dissolved oxygen, DO)被广泛认为是评估水体健康状况的重要指标之一[2]。由于数据类型及技术限制等多种因素,目前的海洋溶解氧浓度预测大多是基于单一浮标数据来进行的,然而海洋作为一个多要素、多尺度的复杂系统,单个浮标所能覆盖的监测区域十分有限,因此,多浮标监测及预测已经逐渐成为研究学者关注的热点[34]。不同浮标获取的监测数据各不相同,如何将各浮标数据有机的结合起来并分析、挖掘多浮标数据中暗含着空间信息,充分地利用多浮标的范围性数据来提高溶解氧浓度预测的精度是一件极具挑战性的任务。
为了充分提取融合不同浮标的参数数据及空间信息,Jin等[5]通过堆叠图卷积网络(Graph Convolutional Network, GCN)、循环神经网络(Recurrent Neural Network, RNN)和注意力机制来更好地捕获特征之间的关系。杜先君和李河[6]基于GCN网络结合Transformer网络的全局特征聚焦能力,融合两个网络提取的数据特征实现厄尔尼诺现象的精准预测。然而,构建图卷积神经网络需要深入理解图的复杂结构,并涉及复杂的矩阵运算以解决相邻节点之间的权重问题。图注意力网络(Graph Attention Mechanism,GAT)[7]在很大程度上缓解了这一问题。GAT不仅继承了图神经网络处理结构性数据的能力,对相邻节点的特征进行加权求和来更新节点的特征向量的方式还使得它保留了注意力机制中自适应的学习参数间关系的能力,能够自适应的学习参数间的相关关系。在处理具有复杂结构和非局部关系的图数据时, GAT相比于 GCN表现出更好的灵活性和性能。因此,利用GAT能够实现海水多参数数据的特征与不同浮标间的空间信息的提取与融合,使浮标网络的空间拓扑信息和浮标传感器的特征属性得到更充分的利用。
而对于溶解氧浓度预测,传统的方法主要是基于统计模型和经验公式,然而,随着计算能力的提高和数据科学的发展,在处理复杂的非线性关系和多变量相互作用时,机器学习和深度学习技术已被广泛应用于溶解氧浓度预测[89],这些方法可以捕捉复杂的非线性关系,并通过在大量历史数据上进行训练来提高预测精度。
RNN网络、长短时记忆神经网络(Long Short-Term Memory, LSTM)、门控循环网络(Gated Recurrent Unit, GRU)等广泛应用于时间序列预测模型的神经网络在海洋水质参数的预测领域都已经取得了良好的效果。这些深度学习网络模型学习序列数据中的长期依赖关系,获取长期时间依赖关系以便能够对未来的数据进行预测,Liu等[10]比较了5种RNN方法,实验结果表明,基于注意力机制的RNN在短期和长期溶解氧预测中都表现出更高的准确性。Yue等[11]提出了自注意力预测递归神经网络来捕捉全局位置相关性和时空特征,但它缺乏泛化能力和性能。石绥祥等[12]构建融合的LSTM叶绿素a浓度预测模型,分割不同要素与叶绿素a浓度的长短期依赖关系,与传统的LSTM模型相比,提高了特征提取效率,提高了预测精度,但该模型不能捕捉空间信息。Jia等[13]利用LSTM建立了东海海面温度预测模型,探索了不同参数对预测精度的影响,虽然他们的模型能较好地捕捉时空特征关系,并做出良好的预测,但不能灵活地从多维数据中提取特征权重。张宇等[14]基于卷积LSTM网络结合多元集合经验模态分解方法预测海面温度,可同时获取数据的空间相关性与时间的周期特性,但当空间分布特征较为复杂时,预测精度会下降。由研究报道可以看出,双向网络、自注意力机制等改进策略也进一步缓解了循环网络本身的梯度爆炸以及梯度消失现象。但随着海量数据的加入,网络捕捉数据本身分布特征的能力也有所下降,特别是在捕捉不同浮标数据之间的空间依赖性方面存在一定困难。
生成对抗网络(Generative Adversarial Networks, GAN)[1516]作为深度学习领域的前沿技术能在一定程度上缓解以上问题,GAN网络引入了对抗性训练的思想,在生成任务上和数据概率分布信息的提取方面具有极强的能力,它由生成器网络和鉴别器网络组成,生成器负责生成尽可能真实的样本,而判别器则负责区分生成的样本和真实样本,基于生成器和鉴别器的对抗性训练,GAN原则上可以捕获数据的分布特征。鉴于GAN具有良好的捕捉数据间隐含深层关系的能力,本文尝试将GAN网络引入到模型框架,通过对抗训练,捕捉并重现海水参数数据集中复杂的概率分布特征,提高模型的预测性能。
基于此,本文首先利用皮尔逊相关系数分析不同浮标上所测参数与溶解氧浓度之间的相关关系,对多浮标数据进行降维,采用主成分分析法完成降维后数据的融合,用图注意力机制模型来对数据中所蕴含的空间信息进行特征提取,最后利用GAN网络来实现对数据的时间信息、概率分布信息的提取,引入自注意力机制(Self-Attention Mechanism, SA)和双时间尺度更新规则(Two Time-Scale Update Rule, TTUR)来提高GAN的训练效率和生成效果,从而实现最终的预测。
本文实验数据采集自广西北海近岸水域,数据来源于先岛仪器设备有限公司提供的EM700浮标,浮标所搭载的所有光学探头的最大工作深度为61 m,工作温度范围为−5℃至+50℃,可存储最多150000个数据点,浮标的地理位置如图1所示。该浮标搭载了全球定位系统(Global Positioning System,GPS)和高精度水质传感器YSI-6600,能够满足海洋水质监测的需求。
传感器每30 min记录一次数据,主要测量参数包括温度、电导率、盐度、溶解氧、pH值、浊度、叶绿素以及藻红蛋白等水质参数,如表1所示。
数据采集时间范围为2019年1月1日至2019年12月31日,海洋水质参数尤其是溶解氧浓度受季节性变化、温度波动、浮游植物生长等因素的影响较大,由于实验所采用的监测数据覆盖全年,能够全面反映这些时空变化特征,确保预测模型能够适应不同季节和环境条件下的水质变化。此外,1年的数据能够提供足够的样本量,确保数据的完整性和稳定性,有助于提高模型的泛化能力和预测精度。
由于传感器误差、设备故障或外部环境干扰等因素,部分数据存在突变现象,因此,在数据预处理阶段,使用$3\sigma $原则对监测点采集的异常值进行了筛选和修正,以确保数据的连续性和一致性。各浮标记录的溶解氧浓度数据随时间变化的可视化结果如图2所示。
完成异常值处理后,为了充分挖掘浮标数据中的关键信息,利用皮尔逊相关系数法分析各浮标参数之间的相关性,筛选出与溶解氧浓度密切相关的特征,实现数据的初步降维。
皮尔逊相关系数[17]是最常用的相关关系评价方法之一,用于衡量两个变量之间相关程度,其取值范围从−1到1。通过计算皮尔逊相关系数,可以得出两个变量之间是否符合线性关系,进而衡量这种线性关系的强度。其计算公式为
$ {\rho _{X,Y}} = \frac{{{{\mathrm{cov}}} (X,Y)}}{{{\sigma _X}{\sigma _Y}}} = \frac{{E(XY) - E(X)E(Y)}}{{\sqrt {E({X^2}) - {E^2}(X)} \sqrt {E({Y^2}) - {E^2}(Y)} }}, $
式中:${{\mathrm{cov}}} (X,Y)$表示两个特征的协方差,${\rho _{X,Y}}$表示两个特征的标准差。不同浮标各参数间皮尔逊相关关系热力图,如图3所示。
利用皮尔逊相关系数对单个浮标的监测参数进行相关性分析后,筛选出与该浮标的溶解氧浓度相关性最高的五个特征,如表2所示。
利用皮尔逊相关系数法选择得到的与溶解氧相关性最高的5个特征,采用主成分分析(Principal Component Analysis, PCA)法[18]将其与溶解氧浓度进行多特征融合。PCA是一种常用的数据降维和压缩方法,采用该方法对单个浮标数据进行降维融合,能够得到表征单浮标溶解氧相关信息的综合特征。对各浮标的数据进行特征融合后,得到的综合特征向量,如图4所示。
为了提高计算效率并充分利用空间信息,本文采用GAT对各浮标多特征融合后的综合特征进行空间融合,以减少环境噪声的影响,从而提高数据的空间分辨率。GAT引入了注意力机制,通过高效的特征传递方式增强其可扩展性,具有强大的节点分类能力和特征表达能力。作为一种基于图结构的神经网络模型,GAT能高效训练与推理,相比于传统机器学习方法,可以更快地收敛,更好地利用数据集中的信息。
针对多浮标数据中蕴含的空间信息,首先利用球面距离公式[19]来获取各浮标之间的距离,并进行归一化处理,得到原始邻接矩阵,进而构建包含空间拓扑信息的图结构作为GAT的输入;再以各浮标的中间融合参数为节点,基于原始邻接矩阵构建图注意力网络模型,进一步提取浮标间的相关关系并融合节点信息,充分利用浮标网络的空间拓扑信息和传感器特征,将多个特征融合为新特征,形成代表整片海域水质参数变化的新特征矩阵。
从空间维度分析,同一海域的浮标之间存在显著的空间相关性,因此,这些浮标可在空间分布上形成一个图结构。首先利用球面距离公式结合浮标的经纬度计算不同浮标之间的距离。设浮标${b_i}$的纬度角为${\beta _i}$,经度角为${\alpha _i}$,浮标${b_j}$的纬度角为${\beta _j}$,经度角为${\alpha _j}$,用式(2)计算出两个浮标之间的距离${d_{i,j}}$
$ d_{i,j}=R\arccos[\cos\beta_i\cos\beta_j\cos(\alpha_i-\alpha_j)+\sin\beta_i\sin\beta_j], $
式中:$R$是球体的半径。浮标之间的距离关系如表3所示。
结合纬度、经度和浮标距离的关系构建的浮标之间的拓扑结构,如图5所示。
GAT的输入数据是基于浮标${b_i}$${b_j}$之间的空间位置拓扑关系的图结构,构造的邻接矩阵即等价于建模数据图结构。设置邻接矩阵${\boldsymbol{A}} = {R^{N \times N}}$对空间相关性进行建模,${d_r}$为距离阈值,如果两个浮标之间的距离小于${d_r}$,则建立邻接点关系,说明两个浮标之间存在较强的空间相关性,如式(3)所示:
$ A_{ij}=\left\{\begin{array}{*{20}{c}}1, & 0 < d_{i,j} < d_r \\0, & d_{i,j} > d_r\end{array}.\right. $
合理的距离阈值对GAT网络${d_r}$的特征提取至关重要,太小会导致邻接矩阵过于稀疏,不能有效地构建浮标之间的空间位置关系,${d_r}$太大又会构建太多的图边缘关系,这将增加模型的计算压力。本文根据实际情况分析了浮标的距离和位置后,将距离阈值设置为50 km并基于此来构造邻接矩阵。
得到浮标之间的距离和位置关系后,借助图神经网络(Graph Neural Network,GNN)[2021]从浮标中提取空间特征。通过不断更新邻接矩阵和特征矩阵提取其他相邻浮标对中心浮标特征的贡献,使得空间信息得到充分的利用。
然而,传统的GCN网络对不同浮标传感器数据的空间依赖性不敏感[22],因此我们采用GAT模型处理图结构。GAT的自注意力机制通对相邻节点的特征进行加权求和来解决相邻节点之间的权值问题,不需要理解图本身的结构,也不需要复杂的矩阵运算。与GCN相比,GAT具有更好的灵活性,尤其是在处理结构复杂和非局部关系的图数据时,GAT通常能得到更好的结果,GAT的结构如图6所示。
图注意力层的输入是海洋水质参数的邻接矩阵${\boldsymbol{A}}$和特征矩阵${\boldsymbol{X}}$
$ {\boldsymbol{X}} = \left\{ {{x_1},{x_2}, \cdots, {x_N}} \right\},{x_i} \in {R^{m \times n}}, $
式中:$N$为浮标编号,$m$为输入特征的长度,$n$为输入特征的维数。
经过一层GAT提取特征后,输出数据为更新后的浮标特征矩阵${\boldsymbol{X}}'$
$ {\boldsymbol{X}}' = \left\{ {{x_1}',{x_2}', \cdots ,{x_N}'} \right\},{x_i}' \in {R^{m \times n}}. $
图注意力层首先使用共享权重矩阵$W$,将每个浮标的特征映射到高维,以获得足够的特征表达能力。然后对每个浮标采用共享自注意力机制$a$来计算注意力系数${e_{ij}}$,即浮标$j$的特征值对浮标$i$的贡献[23]
$ {e_{ij}} = a(W{x_i},W{x_j}). $
由于上述公式允许每个浮标与其他浮标均进行计算,违背了构建图结构的初衷,因此在模型中加入掩码注意力机制,以确保只计算浮标$i$的邻近浮标$j,j \in {N_i}$,其中${N_i}$是浮标$i$的邻近浮标集合。本文使用的邻近浮标即与中心浮标直接相邻的节点,这样更容易比较不同浮标之间的关系。相邻浮标$j$的注意力系数使用Softmax函数计算:
$ a_{ij}=\mathrm{soft}\max_j(e_{ij})=\frac{\exp(e_{ij})}{\displaystyle\sum\limits_{k\in N_i}^{ }\exp(e_{ik})}. $
注意力机制$a$是一个带参数${\vec a^T} \in {R^{2F'}}$的单层前馈神经网络,采用Leaky ReLu作为激活函数,其公式如下:
$ e_{ij}=\mathrm{Leaky}\; \mathrm{Re}\mathrm{Lu}(\vec{a}^T[Wx_i \left\| Wx\right._j]), $
$ a_{ij}=soft\; \max_j(e_{ij})=\frac{\exp(\mathrm{Leaky}\; \mathrm{Re}\mathrm{Lu}(\vec{a}^T[Wx_i \left\| Wx\right._j]))}{\displaystyle\sum\limits_{k\in N_i}^{ }\exp(\mathrm{Leaky}\; \mathrm{Re}\mathrm{Lu}(\vec{a}^T[Wx_i \left\| Wx\right._k]))}. $
根据注意力系数${a_{ij}}$对输入特征的最终加权得到输出特征,利用激活函数$\sigma $得到浮标$i$的新特征:
$ x' = \sigma \left(\displaystyle\sum\limits_{j \in {N_i}} {{a_{ij}}W{x_j}} \right). $
为提高模型的稳定性,GAT还引入了一种多头注意力机制,利用该机制分别计算$m$个子空间,并通过求平均值对计算结果进行拼接,计算公式为
$ x' = \sigma \left(\frac{1}{m}\sum\limits_{m = 1}^m {\sum\limits_{j \in {N_i}} {a_{ij}^mW{x_j}} } \right), $
式中:$a_{ij}^m$为浮标$i$$j$$m$子空间内的注意力系数[24]
经GAT层更新后的邻接矩阵如图7所示,以热力图的形式呈现了浮标之间的空间结构,其中包含了整个海域内浮标的拓扑关系,表征了浮标之间的相互作用。
经过空间特征融合得到整片海域水质参数变化的新特征矩阵后,为提高网络的预测精度并减少预测时间[2526],本文提出了一种基于迭代惩罚系数的生成对抗网络(Gradient Penalty for Wasserstein GAN,WGAN_GP)[2728]。为进一步提高数据质量改善数据拟合效果,在WGAN_GP 生成器网络的基础上增加了自注意力机制(Self-Attention Mechanism, SA)[2930]。本文改进后的SA-WGAN_GP模型结构如图8所示。SA-WGAN_GP的生成网络G引入了基于双向门控循环网络(Bidirectional Gated Recurrent Unit, BiGRU)的注意力机制,使时间序列数据的信息得以完整地呈现,不仅提高了预测精度,而且缩短了预测时间。SA-WGAN_GP网络的鉴别器网络D采用了具有良好特征提取能力的卷积神经网络(Convolutional Neural Networks, CNN)。
生成器网络G,首先由BiGRU的第一层捕获输入数据序列中的时间依赖性和关键特征,并输出每个序列单元的隐藏状态。然后输入第二层的特征注意层,通过加权输入特征,使模型学习哪些特征在训练过程中对任务更关键,降低模型对噪声或无关特征的敏感性,从而提高模型的泛化能力,将隐藏层信息与不同时间的不同特征结合,形成新的序列数据,并将其作为第三层的输入,第三层为BiGRU模型,用于学习映射任务。最后,将BiGRU的输出数据输入到特征注意力层,将注意力分配到时间序列上,减少模型对所有时间步长的过度注意,降低过拟合的风险,使模型更加集中的关注关键信息,最终通过全连接层输出预测值。
GAN网络传统的训练过程中,生成器网络和鉴别器网络学习速率基本没有被特殊考量,但GAN网络本身的特点决定了鉴别器网络D更新速度不能达到生成器网络G的更新速度,两种训练速度不匹配,因此模型的稳定性受损,容易发生模式崩溃[27]。为了平衡生成器和判别器的更新速度,我们引入了双时间尺度更新规则(Two Time-Scale Update Rule, TTUR)以提高GAN网络的训练效果[3132]
将鉴别器网络D模型梯度$D(y\left| x \right.)$设为$h(d)$,生成器网络G模型梯度$G(y\left| x \right.)$设为$h(g)$。假设鉴别器网络D在每次训练中都有$m$个输入数据$x(t)(1 \leqslant t \leqslant m)$,则D网络的梯度$h(d)$
$ h(d) = {\nabla _{{\theta _d}}}\frac{1}{m}\sum\limits_{t = 1}^m {[D(y\left| {x(t)} \right.,y < k + 1) + D(y\left| {G(z(t)),y = k + 1} \right.)]}, $
式中:$x(t)$为第$t$个输入数据,${\theta _d}$为D网络的参数。
G网络的梯度$h(g)$
$ h(g) = {\nabla _{{\theta _g}}}\frac{1}{m}\sum\limits_{t = 1}^m {[D(y\left| {G(z(t)),y = k + 1} \right.)]} ,$
式中:$G(z(t))$为G网络生成的第$t$个数据,${\theta _g}$为G网络的参数。
当D网络更新速度较快时,G网络的学习时间减少,导致生成的数据缺乏多样性;当G网络更新速度较快时,D网络由于无法跟上更新速率而陷入局部最优解,从而导致模式崩溃。利用TTUR平衡这两个网络的更新速度的公式:
$ {h_{n + 1}} = \beta {l_d}{h_n}(d) + \alpha {l_g}{h_n}(g), $
式中:${l_d}$为D网络的学习速率,${l_g}$为G网络的学习速率,$n$为GAN网络的迭代次数。TTUR作为生成对抗网络的一种训练方法,旨在优化生成器网络G和鉴别器网络D的学习速率,令生成器网络G和鉴别器网络D具有一致的更新速度,提高了GAN网络的稳定性和训练效率,得到了良好的结果。
本文使用6个浮标的监测数据为输入数据,即GX-02、GX-14、GX-15、GX-17、GX-03、GX-19。以浮标GX-15的溶解氧浓度为目标预测数据。数据集的时间范围为2019年1月1日至2019年12月31日,其中2019年1月1日至2019年9月31日的数据作为训练集,2019年10月1日至2019年12月31日的数据为评估模型泛化能力的测试集,原始数据采样周期为30 min,为更有效地提取数据特征,减少计算量,将数据每2 h采集重构,后续实验过程均使用重构后的数据集。本文实验基于python3.7环境,结合TenserFlow框架建立、训练并测试本文模型与其他基线模型。
评价指标采用平均绝对误差(Mean Absolute Error, MAE)、均方误差(Mean Squared Error, MSE)、均方根误差(Root Mean Squared Error, RMSE)以及决定系数(R-Square, R2),具体定义如下:
$ {\mathrm{MAE}} = \frac{1}{{\text{n}}}\sum\limits_{i - 1}^n {\left| {{{\hat y}_i} - {y_i}} \right|}, $
$ {\mathrm{MSE}} = \frac{1}{n}{\sum\limits_{i - 1}^n {({{\hat y}_i} - {y_i})} ^2}, $
$ {\mathrm{RMSE}} = \sqrt {\frac{1}{n}{{\sum\limits_{i - 1}^n {({{\hat y}_i} - {y_i})}^2 }}} ,$
$ {{{R}}^2} = 1 - \frac{{{{\displaystyle\sum\limits_{i - 1}^n {({{\hat y}_i} - {y_i})}^2 }}}}{{{{\displaystyle\sum\limits_{i - 1}^n {({{\bar y}_i} - {y_i})}^2 }}}}, $
式中:${\hat y_i}$是预测值,${y_i}$是真实值,${\bar y_i}$是真实值的平均值,$n$是样本量。
为了验证GAT-WGAN_GP模型在时空特征学习、空间信息建模和数据分布拟合方面的优势,本文将经过特征提取融合后的数据输入到不同模型中进行预测,并建立了3种基线对比模型,分别是:CNN-LSTM[33]、CNN-GRU[34]、GCN-WGAN_GP。CNN-LSTM和CNN-GRU模型利用CNN提取空间特征,同时通过LSTM或GRU捕捉时间序列数据中的依赖性,适用于海水溶解氧浓度预测中的时间序列特性。通过与这两个模型的比较,能够验证GAT-WGAN_GP在时空特征融合与数据分布拟合方面的优势。GCN-WGAN_GP结合了图卷积网络和生成对抗网络,其中GCN能够捕捉浮标间的空间关系,选择GCN-WGAN_GP作为对比模型,旨在验证图卷积网络和生成对抗网络结合的空间建模能力,并探讨GAT-WGAN_GP在引入图注意力机制后,是否能在更精细的空间关系建模和权重调整方面提供进一步的改进。通过将本文提出的GAT-WGAN_GP模型与上述三种基准模型进行对比,以验证模型的预测性能,结果如图9表4所示。
根据图9所示的4个模型的预测结果可以看出:与CNN-LSTM与CNN-GRU对融合后的数据处理效果不够好,CNN网络无法充分提取空间信息。而GAT-WGAN_GP通过GAT引入的注意力机制,使模型能够自动学习浮标和相邻浮标的关系,即空间信息,相比于GCN-WGAN_GP模型能够自适应地调整浮标与临近浮标之间的权重关系。这种在计算浮标特征时考虑全局信息,建立具有全局结构的策略,使得GAT-WGAN_GP能实现对所有浮标特征的充分学习,进而得到更好的模型性能。观察表4所示的4种模型预测误差指标上的表现,本文所建立的模型在所有评价指标上均表现最好,进一步验证了上述结论。
本文在空间特征提取中,采用GAT获取不同浮标数据间的空间信息,增强了模型对数据的全面分析能力;在溶解氧浓度预测中,在WGAN_GP网络中引入了自注意力机制,使模型更精准地拟合数据分布,同时引入TTUR避免了模式崩溃。本论文的主要贡献在于结合图注意力网络和生成对抗网络技术,提出了一种基于时空特征融合的海水溶解氧浓度预测方法,能够有效提高预测精度。通过与其他基线模型的比较,可以看出,本文提出的模型在增强参数分布学习能力和空间特征提取能力方面表现优异,可为水质监测、海洋生态系统管理和环境保护等领域提供一种有效的技术手段。未来,随着数据集的不断扩大和模型进一步优化,该预测方法有望在海洋科学、环境监测及其他相关领域的参数预测预报中得到更广泛的应用,从而为管理者提供更加精确的数据支持。
  • 国家自然科学基金(62275228)
  • 河北省重点研发计划(19273901D)
  • 河北省重点研发计划(20373301D)
  • 河北省自然科学基金(D2024203002)
参考文献 引证文献
排序方式:
1
齐庆华, 蔡榕硕, 颜秀花. 气候变化与我国海洋灾害风险治理探讨[J]. 海洋通报, 2019, 38(4): 361−367.
Qi Qinghua, Cai Rongshuo, Yan Xiuhua. Discussion on climate change and marine disaster risk governance in the coastal China seas[J]. Marine Science Bulletin, 2019, 38(4): 361−367.
2
Kisi O, Alizamir M, Gorgij A D. Dissolved oxygen prediction using a new ensemble method[J]. Environmental Science and Pollution Research, 2020, 27(9): 9589−9603.
3
Li Yanni, Yan Shiqiang, Shi Hongda, et al. Hydrodynamic analysis of a novel multi-buoy wind-wave energy system[J]. Renewable Energy, 2023, 219: 119477.
4
Liu Shixuan, Song Miaomiao, Chen Shizhe, et al. An intelligent modeling framework to optimize the spatial layout of ocean moored buoy observing networks[J]. Frontiers in Marine Science, 2023, 10: 1134418.
5
Jin Junchen, Rong Dingding, Zhang Tong, et al. A GAN-based short-term link traffic prediction approach for urban road networks under a parallel learning framework[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(9): 16185−16196.
6
杜先君, 李河. 基于集成GCN-Transformer网络的ENSO预测模型[J]. 海洋学报, 2023, 45(12): 156−165.
Du Xianjun, Li He. ENSO prediction model based on integrated GCN-Transformer network[J]. Haiyang Xuebao, 2023, 45(12): 156−165.
7
Veličković P, Cucurull G, Casanova A, et al. Graph Attention Networks[C]//International Conference on Learning Representations. 2018.
8
Li Wenjing, Fang Huaiyang, Qin Guangxiong, et al. Concentration estimation of dissolved oxygen in Pearl River Basin using input variable selection and machine learning techniques[J]. Science of the Total Environment, 2020, 731: 139099.
9
Wu Junhao, Wang Zhaocai, Dong Jinghan, et al. Multi-step ahead dissolved oxygen concentration prediction based on knowledge guided ensemble learning and explainable artificial intelligence[J]. Journal of Hydrology, 2024, 636: 131297.
10
Liu Yeqi, Gong Chuanyang, Yang Ling, et al. DSTP-RNN: a dual-stage two-phase attention-based recurrent neural network for long-term and multivariate time series prediction[J]. Expert Systems with Applications, 2020, 143: 113082.
11
Yue Weihao, Xu Yongsheng, Xiang Liang, et al. Prediction of 3-D ocean temperature based on self-attention and predictive RNN[J]. IEEE Geoscience and Remote Sensing Letters, 2024, 21: 1501305.
12
石绥祥, 王蕾, 余璇, 等. 长短期记忆神经网络在叶绿素a浓度预测中的应用[J]. 海洋学报, 2020, 42(2): 134−142.
Shi Suixiang, Wang Lei, Yu Xuan, et al. Application of long term and short term memory neural network in prediction of chlorophyll a concentration[J]. Haiyang Xuebao, 2020, 42(2): 134−142.
13
Jia Xiaoyan, Ji Qiyan, Han Lei, et al. Prediction of sea surface temperature in the East China Sea based on LSTM neural network[J]. Remote Sensing, 2022, 14(14): 3300.
14
张宇, 许大志, 俞胜宾, 等. 基于多尺度深度学习对南海海表温度预报的研究[J]. 海洋学报, 2024, 46(5): 27−36.
Zhang Yu, Xu Dazhi, Yu Shengbin, et al. Forecast of sea surface temperature in the South China Sea based on multi-scale deep learning model[J]. Haiyang Xuebao, 2024, 46(5): 27−36.
15
Goodfellow I J, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal: MIT Press, 2014: 2672−2680.
16
Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks[J]. Communications of the ACM, 2020, 63(11): 139−144.
17
Deng Jiaxiang, Deng Yong, Cheong K H. Combining conflicting evidence based on Pearson correlation coefficient and weighted graph[J]. International Journal of Intelligent Systems, 2021, 36(12): 7443−7460.
18
Ibrahim A, Ismail A, Juahir H, et al. Water quality modelling using principal component analysis and artificial neural network[J]. Marine Pollution Bulletin, 2023, 187: 114493.
19
Mai Gengchen, Xuan Yao, Zuo Wenyun, et al. Sphere2Vec: a general-purpose location representation learning over a spherical surface for large-scale geospatial predictions[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2023, 202: 439−462.
20
Scarselli F, Gori M, Tsoi A C, et al. The graph neural network model[J]. IEEE Transactions on Neural Networks, 2009, 20(1): 61−80.
21
Wang Xiaoqi, Yen K, Hu Yifan, et al. DeepGD: a deep learning framework for graph drawing using GNN[J]. IEEE Computer Graphics and Applications, 2021, 41(5): 32−44.
22
Wang Dujuan, Zhu Jiacheng, Yin Yunqiang, et al. Dynamic travel time prediction with spatiotemporal features: using a GNN-based deep learning method[J]. Annals of Operations Research, 2024, 340(1): 571−591.
23
Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks[C]//International Conference on Learning Representations, 2017.
24
Derr T, Ma Yao, Tang Jiliang. Signed graph convolutional networks[C]//2018 IEEE International Conference on Data Mining (ICDM). Singapore: IEEE, 2018: 929−934.
25
Polamuri S R, Srinivas D K, Mohan D A K. Multi-model generative adversarial network hybrid prediction algorithm (MMGAN-HPA) for stock market prices prediction[J]. Journal of King Saud University-Computer and Information Sciences, 2022, 34(9): 7433−7444.
26
Zhang Jianguang, Zhang Xuyang, Yang Jianfeng, et al. Deep LSTM and GAN based short-term load forecasting method at the zone level[C]//2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC). Fukuoka: IEEE, 2020: 613−618.
27
Festag S, Spreckelsen C. Medical multivariate time series imputation and forecasting based on a recurrent conditional Wasserstein GAN and attention[J]. Journal of Biomedical Informatics, 2023, 139: 104320.
28
Gulrajani I, Ahmed F, Arjovsky M, et al. Improved training of wasserstein GANS[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates Inc. , 2017: 5769−5779.
29
Zhang Chenhan, Yu J J Q, Liu Yi. Spatial-temporal graph attention networks: a deep learning approach for traffic forecasting[J]. IEEE Access, 2019, 7: 166246−166256.
30
Fang Fang, Zhang Pengpeng, Zhou Bo, et al. Atten-GAN: pedestrian trajectory prediction with GAN based on attention mechanism[J]. Cognitive Computation, 2022, 14(6): 2296−2305.
31
Bynagari N B. GANs trained by a two time-scale update rule converge to a local Nash equilibrium[J]. Asian Journal of Applied Science and Engineering, 2019, 8(1): 25−34.
32
Zhang Kang, Zhong Guoqiang, Dong Junyu, et al. Stock market prediction based on generative adversarial network[J]. Procedia Computer Science, 2019, 147: 400−406.
33
Zha Wenshu, Liu Yuping, Wan Yujin, et al. Forecasting monthly gas field production based on the CNN-LSTM model[J]. Energy, 2022, 260: 124889.
34
Yu Jingxin, Zhang Xin, Xu Linlin, et al. A hybrid CNN-GRU model for predicting soil moisture in maize root zone[J]. Agricultural Water Management, 2021, 245: 106649.
2025年第47卷第1期
PDF下载
268
120
引用本文
BibTeX
文章信息
doi: 10.12284/hyxb2025028
  • 接收时间:2024-09-28
  • 首发时间:2025-11-10
  • 出版时间:2025-01-31
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-09-28
  • 修回日期:2024-12-18
基金
国家自然科学基金(62275228)
河北省重点研发计划(19273901D)
河北省重点研发计划(20373301D)
河北省自然科学基金(D2024203002)
作者信息
    1.燕山大学信息科学与工程学院 河北省特种光纤与光纤传感器重点实验室,河北 秦皇岛 066004
    2.燕山大学电气工程学院 河北省测试计量技术及仪器重点实验室,河北 秦皇岛 066004

通讯作者:

*陈颖,教授,主要从事海洋生态环境监测与大数据分析的研究。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/hyxb/CN/10.12284/hyxb2025028
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