Article(id=1276897174982094882, tenantId=1146029695717560320, journalId=1276576982599962646, issueId=1276896975568109838, articleNumber=null, orderNo=null, doi=10.3724/j.slxb.20250474, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1755273600000, receivedDateStr=2025-08-16, revisedDate=null, revisedDateStr=null, acceptedDate=null, acceptedDateStr=null, onlineDate=1782365610118, onlineDateStr=2026-06-25, pubDate=1779206400000, pubDateStr=2026-05-20, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1782365610118, onlineIssueDateStr=2026-06-25, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1782365610118, creator=13701087609, updateTime=1782365610118, updator=13701087609, issue=Issue{id=1276896975568109838, tenantId=1146029695717560320, journalId=1276576982599962646, year='2026', volume='57', issue='5', pageStart='651', pageEnd='808', issueExtLink='null', onlineDate='null', pubDate='1779206400000', pubDateStr='2026-05-20', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1782365562574, creator='13701087609', updateTime=1782367019422, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1276903086153142605, tenantId=1146029695717560320, journalId=1276576982599962646, issueId=1276896975568109838, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1276903086153142606, tenantId=1146029695717560320, journalId=1276576982599962646, issueId=1276896975568109838, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=704, endPage=715, ext={EN=ArticleExt(id=1276897175183421476, articleId=1276897174982094882, tenantId=1146029695717560320, journalId=1276576982599962646, language=EN, title=Efficient and intelligent forecasting of urban waterlogging based on UNet-KAN-SR modeling, columnId=null, journalTitle=Journal of Hydraulic Engineering, columnName=null, runingTitle=null, highlight=null, articleAbstract=

With global climate change and accelerating urbanization, urban waterlogging disasters have become increasingly frequent and severe, making rapid waterlogging forecasting a key research focus. Compared with traditional numerical simulation methods, deep-learning-based artificial intelligence (AI) models can significantly improve computational efficiency. However, they often encounter training bottlenecks due to limited GPU memory. To address this, this study proposes an efficient AI urban waterlogging forecasting model named as UNet-KAN-SR. This model first employs the UNet-KAN module to efficiently simulate the spatio-temporal evolution of waterlogging over low-resolution grids, and then leverages the SR (super-resolution) module, along with high-resolution surface information, to progressively map the low-resolution waterlogging distribution to high-resolution distribution. This spatiotemporal decoupling strategy can ensure simulation accuracy while substantially reducing the computational resources required for training AI models. Experimental results demonstrate that the UNet-KAN-SR model can simulate a 3-hour waterlogging distribution within 3 minutes, achieving a root mean square error (RMSE) of 9 cm and a probability of detection (POD) of 0.84, demonstrating high accuracy and computational efficiency. Further analysis reveals that the integration of the KAN module can significantly enhance the model’s capability to capture nonlinear flood dynamics when compared with common CNN modules, reducing RMSE by 10%. Furthermore, this study finds that incorporating high-resolution features, such as surface topography, building coverage ratio, and land use, can significantly improve the simulation performance but performance improvement is similar under different feature combinations. This indicates that by optimizing the combination of input features during AI model construction, training speed can be enhanced, modeling costs controlled, and efficient intelligent forecasting achieved.

, authors=null, authorsList=Yaoming CHEN, Ruidong LI, Ji CHEN, Guangheng NI, authorCompany=null, correspAuthors=null, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1276897177817444402, articleId=1276897174982094882, tenantId=1146029695717560320, journalId=1276576982599962646, language=CN, title=基于UNet-KAN-SR模型的城市积涝高效智能预报, columnId=0, journalTitle=水利学报, columnName=, runingTitle=null, highlight=null, articleAbstract=

伴随着全球气候变化,在经历高速城镇化进程之后,城市积涝灾害趋多趋强,使得快速积涝预测成为研究热点问题。相较于传统数值模拟,基于深度学习的人工智能模拟方法能有效提升计算效率,但因GPU计算显存有限而易遭遇模型训练瓶颈。鉴于此,本研究提出了一种基于UNet-KAN-SR模型的城市积涝高效智能预报方法,先利用UNet-KAN模块学习低分辨率积涝分布的时空演变规律,再利用超分辨率SR模块与高分辨率下垫面信息,逐步将低分辨率积涝分布映射为高分辨率积涝分布,从而利用时空解耦在保障时空模拟精度的同时有效降低智能模型训练的计算资源需求。测试降雨情景的验证结果表明,UNet-KAN-SR模型能在3 min内完成未来3 h积涝分布预报,均方根误差为9 cm、命中率达0.86,具备较高的计算精度与效率。相较于已有智能模型常用的卷积层,UNet-KAN-SR模型引入的KAN层可显著提升模型对积涝时空演变过程的非线性建模能力,使均方根误差降低10%。此外,本研究发现高分辨率下垫面信息能显著提升智能模型对积涝分布的预测能力且不同类型的下垫面信息所能取得的提升幅度相近,由此说明在构建智能模型时,可通过优选输入下垫面特征组合,有效提升模型训练速度并控制建模成本,实现高效智能预报。

, authors=

陈耀明(2003—),博士生,主要从事水文水资源研究。E-mail:

, authorsList=陈耀明, 李瑞栋, 陈骥, 倪广恒, authorCompany=null, correspAuthors=null, authorNote=null, correspAuthorsNote=
李瑞栋(1997—),博士,助理研究员,主要从事城市洪涝研究。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=8XaVWGCDbtEhG6ylK1RmIw==, magXml=kjNNTpjGq5q0/SAfEIOl+Q==, pdfUrl=null, pdf=qdjZvx74f/IjKOuc9tfG8A==, pdfFileSize=14651740, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=Ll1hPKRaM3WWhDKDnVcNfQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=XA4lkDuYYnO2H4UDgJdquQ==, mapNumber=null, fund=null)}, authors=[Author(id=1277261249536721680, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=chenyaom@outlook.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1277261249855488788, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261249536721680, language=EN, stringName=Yaoming CHEN, firstName=Yaoming, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China
2.Department of Civil Engineering,The University of Hong Kong,Hong Kong 999077,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1277261251503850261, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261249536721680, language=CN, stringName=陈耀明, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084
2.香港大学 土木工程系,香港 999077, bio={"content":"

陈耀明(2003—),博士生,主要从事水文水资源研究。E-mail:

"}, bioImg=null, bioContent=

陈耀明(2003—),博士生,主要从事水文水资源研究。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1277261249184400136, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=1., ext=[AuthorCompanyExt(id=1277261249196983049, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China), AuthorCompanyExt(id=1277261249205371658, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084)]), AuthorCompany(id=1277261249440252683, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=2., ext=[AuthorCompanyExt(id=1277261249448641292, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249440252683, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Department of Civil Engineering,The University of Hong Kong,Hong Kong 999077,China), AuthorCompanyExt(id=1277261249457029901, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249440252683, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.香港大学 土木工程系,香港 999077)])]), Author(id=1277261251591930647, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=lyy0744@mail.tsinghua.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1277261251671622425, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261251591930647, language=EN, stringName=Ruidong LI, firstName=Ruidong, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1277261251885531930, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261251591930647, language=CN, stringName=李瑞栋, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1277261249184400136, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=1., ext=[AuthorCompanyExt(id=1277261249196983049, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China), AuthorCompanyExt(id=1277261249205371658, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084)])]), Author(id=1277261251973612316, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=1277261252359488286, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261251973612316, language=EN, stringName=Ji CHEN, firstName=Ji, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.Department of Civil Engineering,The University of Hong Kong,Hong Kong 999077,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1277261252732781343, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261251973612316, language=CN, stringName=陈骥, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.香港大学 土木工程系,香港 999077, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1277261249440252683, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=2., ext=[AuthorCompanyExt(id=1277261249448641292, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249440252683, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Department of Civil Engineering,The University of Hong Kong,Hong Kong 999077,China), AuthorCompanyExt(id=1277261249457029901, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249440252683, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.香港大学 土木工程系,香港 999077)])]), Author(id=1277261252825056033, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=1277261253168988963, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261252825056033, language=EN, stringName=Guangheng NI, firstName=Guangheng, middleName=null, lastName=NI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1277261253567447844, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261252825056033, language=CN, stringName=倪广恒, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1277261249184400136, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=1., ext=[AuthorCompanyExt(id=1277261249196983049, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China), AuthorCompanyExt(id=1277261249205371658, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084)])])], keywords=[Keyword(id=1277261254028821285, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, orderNo=1, keyword=urban waterlogging forecasting), Keyword(id=1277261254104318758, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, orderNo=2, keyword=deep learning), Keyword(id=1277261254414697255, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, orderNo=3, keyword=spatio-temporal prediction), Keyword(id=1277261255941423912, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, orderNo=4, keyword=super-resolution model), Keyword(id=1277261256113390377, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, orderNo=5, keyword=Beijing municipal administrative center), Keyword(id=1277261256499266346, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, orderNo=1, keyword=城市内涝预测), Keyword(id=1277261256914502443, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, orderNo=2, keyword=深度学习), Keyword(id=1277261256994194220, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, orderNo=3, keyword=时空预测), Keyword(id=1277261257338127149, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, orderNo=4, keyword=超分辨率模型), Keyword(id=1277261257761751854, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, orderNo=5, keyword=北京城市副中心)], refs=[Reference(id=1277261261578568511, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=苑希民,兰卓青,王丽娜,等. 极端天气城市特大暴雨洪涝灾害特征再分析及应对策略[J]. 水利学报,2024,55(11):1298-1308., articleTitle=苑希民,兰卓青,王丽娜,等. 极端天气城市特大暴雨洪涝灾害特征再分析及应对策略[J]. 水利学报,2024,55(11):1298-1308., refAbstract=null), Reference(id=1277261261645677376, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=YUAN X M,LAN Z Q,WANG L N,et al. Re-analysis of the characteristics and response strategies of extreme weather urban heavy rainstorm flood disasters[J]. Journal of Hydraulic Engineering,2024,55(11):1298-1308. (in Chinese), articleTitle=YUAN X M,LAN Z Q,WANG L N,et al. Re-analysis of the characteristics and response strategies of extreme weather urban heavy rainstorm flood disasters[J]. Journal of Hydraulic Engineering,2024,55(11):1298-1308. (in Chinese), refAbstract=null), Reference(id=1277261261733757761, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=水文科技热点[J]. 水文,2024,44(6):118., articleTitle=水文科技热点[J]. 水文,2024,44(6):118., refAbstract=null), Reference(id=1277261261805060930, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=Hydrological science and technology hotspots[J]. Hydrology,2024,44(6):118. (in Chinese), articleTitle=Hydrological science and technology hotspots[J]. Hydrology,2024,44(6):118. (in Chinese), refAbstract=null), Reference(id=1277261261867975491, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=侯精明,张兆安,马利平,等. 基于GPU加速技术的非结构流域雨洪数值模型[J]. 水科学进展,2021,32(4):567-576., articleTitle=侯精明,张兆安,马利平,等. 基于GPU加速技术的非结构流域雨洪数值模型[J]. 水科学进展,2021,32(4):567-576., refAbstract=null), Reference(id=1277261261926695748, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=HOU J M,ZHANG Z A,MA L P,et al. A GPU-accelerated unstructured basin rainfall-runoff numerical model[J]. Advances in Water Science,2021,32(4):567-576. (in Chinese), articleTitle=HOU J M,ZHANG Z A,MA L P,et al. A GPU-accelerated unstructured basin rainfall-runoff numerical model[J]. Advances in Water Science,2021,32(4):567-576. (in Chinese), refAbstract=null), Reference(id=1277261261989610309, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=张大伟,冯新政,林文青,等. 国内外不同GPU用于洪水模拟时加速性能对比分析[J]. 中国水利水电科学研究院学报(中英文),2024,22(5):530-538., articleTitle=张大伟,冯新政,林文青,等. 国内外不同GPU用于洪水模拟时加速性能对比分析[J]. 中国水利水电科学研究院学报(中英文),2024,22(5):530-538., refAbstract=null), Reference(id=1277261262069302086, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=ZHANG D W,FENG X Z,LIN W Q,et al. Comparative analysis of acceleration performance of different GPUs in flood simulation[J]. Journal of the Chinese Institute of Water Resources and Hydropower Research,2024,22(5):530-538. (in Chinese), articleTitle=ZHANG D W,FENG X Z,LIN W Q,et al. Comparative analysis of acceleration performance of different GPUs in flood simulation[J]. Journal of the Chinese Institute of Water Resources and Hydropower Research,2024,22(5):530-538. (in Chinese), refAbstract=null), Reference(id=1277261262132216647, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=AURELI F,PROST F,VACONDIO R,et al. A GPU-accelerated shallow-water scheme for surface runoff simulations[J]. Water,2020,12(3):637., articleTitle=AURELI F,PROST F,VACONDIO R,et al. A GPU-accelerated shallow-water scheme for surface runoff simulations[J]. Water,2020,12(3):637., refAbstract=null), Reference(id=1277261262199325512, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=BENTIVOGLIO R,ISUFI E,JONKMAN S N,et al. Deep learning methods for flood mapping:a review of existing applications and future research directions[J]. Hydrology and Earth System Sciences Discussions,2022,26(16):4345-4378, articleTitle=BENTIVOGLIO R,ISUFI E,JONKMAN S N,et al. Deep learning methods for flood mapping:a review of existing applications and future research directions[J]. Hydrology and Earth System Sciences Discussions,2022,26(16):4345-4378, refAbstract=null), Reference(id=1277261262262240073, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=林凯荣,欧阳佳娜,马旭民,等. 基于深度学习的城市内涝积水水深预测模型[J]. 水资源保护,2025,41(1):56-63., articleTitle=林凯荣,欧阳佳娜,马旭民,等. 基于深度学习的城市内涝积水水深预测模型[J]. 水资源保护,2025,41(1):56-63., refAbstract=null), Reference(id=1277261262329348938, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=LIN K R,OUYANG J N,MA X M,et al. Deep learning-based model for predicting urban waterlogging depth[J]. Water Resources Protection,2025,41(1):56-63. (in Chinese), articleTitle=LIN K R,OUYANG J N,MA X M,et al. Deep learning-based model for predicting urban waterlogging depth[J]. Water Resources Protection,2025,41(1):56-63. (in Chinese), refAbstract=null), Reference(id=1277261262396457803, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=李玉龙,梅超,刘家宏,等. 基于数值模型与机器学习的城市内涝智能模拟[J/OL]. 水利水电技术(中英文),1-15[2025-12-17]., articleTitle=李玉龙,梅超,刘家宏,等. 基于数值模型与机器学习的城市内涝智能模拟[J/OL]. 水利水电技术(中英文),1-15[2025-12-17]., refAbstract=null), Reference(id=1277261262463566668, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=LI Y L,MEI C,LIU J H,et al. Intelligent simulation of urban waterlogging based on numerical models and machine learning[J/OL]. Hydraulic and Hydropower Engineering,1-15[2025-12-17]. (in Chinese), articleTitle=LI Y L,MEI C,LIU J H,et al. Intelligent simulation of urban waterlogging based on numerical models and machine learning[J/OL]. Hydraulic and Hydropower Engineering,1-15[2025-12-17]. (in Chinese), refAbstract=null), Reference(id=1277261262526481229, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=BIAN W C,FANG J Y,WANG P,et al. Deep learning surrogate models for spatiotemporal prediction of coastal flooding inundations in Tianjin,China[J]. Journal of Hydrology:Regional Studies,2025,60:102593., articleTitle=BIAN W C,FANG J Y,WANG P,et al. Deep learning surrogate models for spatiotemporal prediction of coastal flooding inundations in Tianjin,China[J]. Journal of Hydrology:Regional Studies,2025,60:102593., refAbstract=null), Reference(id=1277261262610367310, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=康永德,康爱泽,侯精明,等. 基于GPU的洪涝模型计算效率提升方法研究[J]. 中国水利,2025(12):64-72., articleTitle=康永德,康爱泽,侯精明,等. 基于GPU的洪涝模型计算效率提升方法研究[J]. 中国水利,2025(12):64-72., refAbstract=null), Reference(id=1277261262685864783, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=KANG Y D,KANG A Z,HOU J M,et al. Research on methods to improve computational efficiency of flood models based on GPU[J]. China Water Resources,2025(12):64-72. (in Chinese), articleTitle=KANG Y D,KANG A Z,HOU J M,et al. Research on methods to improve computational efficiency of flood models based on GPU[J]. China Water Resources,2025(12):64-72. (in Chinese), refAbstract=null), Reference(id=1277261262744585040, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=HE J,ZHANG L M,XIAO T,et al. Deep learning enables super-resolution hydrodynamic flooding process modeling under spatiotemporally varying rainstorms[J]. Water Research,2023,239:120057., articleTitle=HE J,ZHANG L M,XIAO T,et al. Deep learning enables super-resolution hydrodynamic flooding process modeling under spatiotemporally varying rainstorms[J]. Water Research,2023,239:120057., refAbstract=null), Reference(id=1277261262807499601, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=WANG R K,LIAN J J,YUAN X M,et al. Rapid simulation of floods by considering the spatial and temporal characteristics of inundation[J]. International Journal of Disaster Risk Science,2025,16(3):1-15., articleTitle=WANG R K,LIAN J J,YUAN X M,et al. Rapid simulation of floods by considering the spatial and temporal characteristics of inundation[J]. International Journal of Disaster Risk Science,2025,16(3):1-15., refAbstract=null), Reference(id=1277261262882997074, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=GUO Z F,LEITAO J P,SIMÕES N E,et al. Data‐driven flood emulation:Speeding up urban flood predictions by deep convolutional neural networks[J]. Journal of Flood Risk Management,2021,14(1):e12684., articleTitle=GUO Z F,LEITAO J P,SIMÕES N E,et al. Data‐driven flood emulation:Speeding up urban flood predictions by deep convolutional neural networks[J]. Journal of Flood Risk Management,2021,14(1):e12684., refAbstract=null), Reference(id=1277261262941717331, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=HERATH H M V V,MARSHALL L,SAHA A,et al. Subgrid informed neural networks for high-resolution flood mapping[J]. Journal of Hydrology,2025,660:133329., articleTitle=HERATH H M V V,MARSHALL L,SAHA A,et al. Subgrid informed neural networks for high-resolution flood mapping[J]. Journal of Hydrology,2025,660:133329., refAbstract=null), Reference(id=1277261263017214804, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=LIU Z M,WANG Y X,VAIDYA S,et al. Kan:Kolmogorov-arnold networks[J]. arXiv,2024:2404.19756., articleTitle=LIU Z M,WANG Y X,VAIDYA S,et al. Kan:Kolmogorov-arnold networks[J]. arXiv,2024:2404.19756., refAbstract=null), Reference(id=1277261263075935061, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[16], rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=REN D F,HU Q,ZHANG T D. EKLT:Kolmogorov-Arnold attention-driven LSTM with Transformer model for river water level prediction[J]. Journal of Hydrology,2025,649:132430., articleTitle=REN D F,HU Q,ZHANG T D. EKLT:Kolmogorov-Arnold attention-driven LSTM with Transformer model for river water level prediction[J]. Journal of Hydrology,2025,649:132430., refAbstract=null), Reference(id=1277261263143043926, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=WANG C,ZHANG X H,LIU L W. FloodKAN:Integrating kolmogorov-arnold networks for efficient flood extent extraction[J]. Remote Sensing,2025,17(4):564., articleTitle=WANG C,ZHANG X H,LIU L W. FloodKAN:Integrating kolmogorov-arnold networks for efficient flood extent extraction[J]. Remote Sensing,2025,17(4):564., refAbstract=null), Reference(id=1277261263294038871, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=沈佳坤,曹婷. 北京城市副中心创新发展路径研究[J]. 时代经贸,2025,22(3):132-135., articleTitle=沈佳坤,曹婷. 北京城市副中心创新发展路径研究[J]. 时代经贸,2025,22(3):132-135., refAbstract=null), Reference(id=1277261264862708568, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=SHEN J K,CAO T. Research on the innovative development path of the sub-center of Beijing[J]. Times of Economy & Trade,2025,22(3):132-135. (in Chinese), articleTitle=SHEN J K,CAO T. Research on the innovative development path of the sub-center of Beijing[J]. Times of Economy & Trade,2025,22(3):132-135. (in Chinese), refAbstract=null), Reference(id=1277261264971760473, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=刘青云,王慧彦. 北京市通州区洪涝灾害风险评估及区划研究[J]. 水利规划与设计,2022(10):76-82., articleTitle=刘青云,王慧彦. 北京市通州区洪涝灾害风险评估及区划研究[J]. 水利规划与设计,2022(10):76-82., refAbstract=null), Reference(id=1277261265059840858, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=LIU Q Y,WANG H Y. Flood disaster risk assessment and zoning study in Tongzhou District,Beijing[J]. Water Resources Planning and Design,2022(10):76-82. (in Chinese), articleTitle=LIU Q Y,WANG H Y. Flood disaster risk assessment and zoning study in Tongzhou District,Beijing[J]. Water Resources Planning and Design,2022(10):76-82. (in Chinese), refAbstract=null), Reference(id=1277261265139532635, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=周正正,刘曙光,WRIGHT D B. 基于随机暴雨移置方法的城市设计暴雨分析[J]. 水科学进展,2020,31(4):583-591., articleTitle=周正正,刘曙光,WRIGHT D B. 基于随机暴雨移置方法的城市设计暴雨分析[J]. 水科学进展,2020,31(4):583-591., refAbstract=null), Reference(id=1277261265227613020, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[20], rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=ZHOU Z Z,LIU S G,WRIGHT D B. Urban design storm analysis based on the stochastic storm transposition method[J]. Advances in Water Science,2020,31(4):583-591. (in Chinese), articleTitle=ZHOU Z Z,LIU S G,WRIGHT D B. Urban design storm analysis based on the stochastic storm transposition method[J]. Advances in Water Science,2020,31(4):583-591. (in Chinese), refAbstract=null), Reference(id=1277261265303110493, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=李瑞栋,倪广恒,陆科成,等. 北京“23·7”特大暴雨影响下城市交通应急响应时间动态分析:以北京市通州区为例[J]. 应用基础与工程科学学报,2024,32(2):365-377., articleTitle=李瑞栋,倪广恒,陆科成,等. 北京“23·7”特大暴雨影响下城市交通应急响应时间动态分析:以北京市通州区为例[J]. 应用基础与工程科学学报,2024,32(2):365-377., refAbstract=null), Reference(id=1277261265386996574, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[21], rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=LI R D,NI G H,LU K C,et al. Dynamic analysis of urban traffic emergency response time under the impact of the Beijing "23·7" extreme rainstorm:A case study of Tongzhou District,Beijing[J]. Journal of Basic Science and Engineering,2024,32(2):365-377. (in Chinese), articleTitle=LI R D,NI G H,LU K C,et al. Dynamic analysis of urban traffic emergency response time under the impact of the Beijing "23·7" extreme rainstorm:A case study of Tongzhou District,Beijing[J]. Journal of Basic Science and Engineering,2024,32(2):365-377. (in Chinese), refAbstract=null), Reference(id=1277261265462494047, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[22], rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=KARRA K,KONTGIS C,STATMAN-WEIL Z,et al. Global land use/land cover with Sentinel 2 and deep learning[C]// 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS. 2021, articleTitle=KARRA K,KONTGIS C,STATMAN-WEIL Z,et al. Global land use/land cover with Sentinel 2 and deep learning[C]// 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS. 2021, refAbstract=null), Reference(id=1277261265525408608, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[23], rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=SHI Q,ZHU J J,LIU Z Y,et al. The last puzzle of global building footprints—Mapping 280 million buildings in East Asia based on VHR images[J]. Journal of Remote Sensing,2024,4:138., articleTitle=SHI Q,ZHU J J,LIU Z Y,et al. The last puzzle of global building footprints—Mapping 280 million buildings in East Asia based on VHR images[J]. Journal of Remote Sensing,2024,4:138., refAbstract=null), Reference(id=1277261265596711777, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=杨志达. 基于连续性约束交叉相关(COTREC)的降水短临预报方法改进研究[D]. 兰州:兰州大学,2024., articleTitle=杨志达. 基于连续性约束交叉相关(COTREC)的降水短临预报方法改进研究[D]. 兰州:兰州大学,2024., refAbstract=null), Reference(id=1277261265672209250, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[24], rfOrder=35, authorNames=null, journalName=null, refType=null, unstructuredReference=YANG Z D. Research on the Improvement of Precipitation Nowcasting Method Based on Continuity-Constrained Cross-Correlation (COTREC)[D]. Lanzhou:Lanzhou University,2024. (in Chinese), articleTitle=YANG Z D. Research on the Improvement of Precipitation Nowcasting Method Based on Continuity-Constrained Cross-Correlation (COTREC)[D]. Lanzhou:Lanzhou University,2024. (in Chinese), refAbstract=null), Reference(id=1277261265756095331, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=BRACHT E,KINDRATENKO V,BRUNNER R J. Sparse spatio-temporal neural network for large-scale forecasting[C]// 2022 IEEE International Conference on Big Data (Big Data). Osaka,2022., articleTitle=BRACHT E,KINDRATENKO V,BRUNNER R J. Sparse spatio-temporal neural network for large-scale forecasting[C]// 2022 IEEE International Conference on Big Data (Big Data). Osaka,2022., refAbstract=null), Reference(id=1277261265831592804, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=HAN D,CHOO M,IM J,et al. Precipitation nowcasting using ground radar data and simpler yet better video prediction deep learning[J]. GIScience & Remote Sensing,2023,60(1):2203363., articleTitle=HAN D,CHOO M,IM J,et al. Precipitation nowcasting using ground radar data and simpler yet better video prediction deep learning[J]. GIScience & Remote Sensing,2023,60(1):2203363., refAbstract=null), Reference(id=1277261265894507365, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[27], rfOrder=38, authorNames=null, journalName=null, refType=null, unstructuredReference=RONNEBERGER O,FISCHER P,BROX T. U-net:Convolutional networks for biomedical image segmentation[C]// International Conference on Medical Image Computing and Computer-assisted Intervention. Cham:Springer international publishing,2015, articleTitle=RONNEBERGER O,FISCHER P,BROX T. U-net:Convolutional networks for biomedical image segmentation[C]// International Conference on Medical Image Computing and Computer-assisted Intervention. Cham:Springer international publishing,2015, refAbstract=null), Reference(id=1277261265965810534, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[28], rfOrder=39, authorNames=null, journalName=null, refType=null, unstructuredReference=LI C X,LIU X Y,LI W Y,et al. U-kan makes strong backbone for medical image segmentation and generation[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2025,39(5):4652-4660., articleTitle=LI C X,LIU X Y,LI W Y,et al. U-kan makes strong backbone for medical image segmentation and generation[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2025,39(5):4652-4660., refAbstract=null), Reference(id=1277261266037113703, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[29], rfOrder=40, authorNames=null, journalName=null, refType=null, unstructuredReference=王丽晶,潘兴瑶,赵飞,等. 北京城市积水内涝风险地图编制与应用[J]. 北京水务,2022(6):24-29,43., articleTitle=王丽晶,潘兴瑶,赵飞,等. 北京城市积水内涝风险地图编制与应用[J]. 北京水务,2022(6):24-29,43., refAbstract=null), Reference(id=1277261266116805480, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[29], rfOrder=41, authorNames=null, journalName=null, refType=null, unstructuredReference=WANG L J,PAN X Y,ZHAO F,et al. Compilation and application of the urban floodwater and waterlogging risk map for Beijing[J]. Beijing Water,2022(6):24-29,43. (in Chinese), articleTitle=WANG L J,PAN X Y,ZHAO F,et al. Compilation and application of the urban floodwater and waterlogging risk map for Beijing[J]. Beijing Water,2022(6):24-29,43. (in Chinese), refAbstract=null), Reference(id=1277261266183914345, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[30], rfOrder=42, authorNames=null, journalName=null, refType=null, unstructuredReference=徐宗学,叶陈雷. 城市暴雨洪涝模拟:原理、模型与展望[J]. 水利学报,2021,52(4):381-392., articleTitle=徐宗学,叶陈雷. 城市暴雨洪涝模拟:原理、模型与展望[J]. 水利学报,2021,52(4):381-392., refAbstract=null), Reference(id=1277261266246828906, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[30], rfOrder=43, authorNames=null, journalName=null, refType=null, unstructuredReference=XU Z X,YE C L. Urban rainstorm flood simulation:Principles,models,and prospects[J]. Journal of Hydraulic Engineering,2021,52(4):381-392. (in Chinese), articleTitle=XU Z X,YE C L. Urban rainstorm flood simulation:Principles,models,and prospects[J]. Journal of Hydraulic Engineering,2021,52(4):381-392. (in Chinese), refAbstract=null), Reference(id=1277261266318132075, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[31], rfOrder=44, authorNames=null, journalName=null, refType=null, unstructuredReference=杨燈,许健,陈德清,等. 激光雷达技术流域地形构建应用现状与展望[J]. 南水北调与水利科技(中英文),2024,22(S1):127-131,135., articleTitle=杨燈,许健,陈德清,等. 激光雷达技术流域地形构建应用现状与展望[J]. 南水北调与水利科技(中英文),2024,22(S1):127-131,135., refAbstract=null), Reference(id=1277261266389435244, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[31], rfOrder=45, authorNames=null, journalName=null, refType=null, unstructuredReference=YANG D,XU J,CHEN D Q,et al. Application status and prospects of LiDAR technology in basin topography construction[J]. South-to-North Water Transfers and Water Science & Technology,2024,22(S1):127-131,135. (in Chinese), articleTitle=YANG D,XU J,CHEN D Q,et al. Application status and prospects of LiDAR technology in basin topography construction[J]. South-to-North Water Transfers and Water Science & Technology,2024,22(S1):127-131,135. (in Chinese), refAbstract=null), Reference(id=1277261266448155501, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[32], rfOrder=46, authorNames=null, journalName=null, refType=null, unstructuredReference=LI Z H,HE W,CHENG M F,et al. SinoLC-1:The first 1-meter resolution national-scale land-cover map of China created with the deep learning framework and open-access data[J]. Earth System Science Data Discussions,2023,2023:1-38., articleTitle=LI Z H,HE W,CHENG M F,et al. SinoLC-1:The first 1-meter resolution national-scale land-cover map of China created with the deep learning framework and open-access data[J]. Earth System Science Data Discussions,2023,2023:1-38., refAbstract=null), Reference(id=1277261266515264366, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[33], rfOrder=47, authorNames=null, journalName=null, refType=null, unstructuredReference=李国一,刘家宏. 基于TELEMAC-2D模型的深圳洪涝风险评估[J]. 水资源保护,2022,38(5):58-64., articleTitle=李国一,刘家宏. 基于TELEMAC-2D模型的深圳洪涝风险评估[J]. 水资源保护,2022,38(5):58-64., refAbstract=null), Reference(id=1277261266590761839, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[33], rfOrder=48, authorNames=null, journalName=null, refType=null, unstructuredReference=LI G Y,LIU J H. Flood risk assessment of Shenzhen based on TELEMAC-2D model[J]. Water Resources Protection,2022,38(5):58-64. (in Chinese), articleTitle=LI G Y,LIU J H. Flood risk assessment of Shenzhen based on TELEMAC-2D model[J]. Water Resources Protection,2022,38(5):58-64. (in Chinese), refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1277261249184400136, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=1., ext=[AuthorCompanyExt(id=1277261249196983049, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China), AuthorCompanyExt(id=1277261249205371658, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084)]), AuthorCompany(id=1277261249440252683, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=2., ext=[AuthorCompanyExt(id=1277261249448641292, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249440252683, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Department of Civil Engineering,The University of Hong Kong,Hong Kong 999077,China), AuthorCompanyExt(id=1277261249457029901, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249440252683, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.香港大学 土木工程系,香港 999077)])], figs=[ArticleFig(id=1277261258613195567, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, label=Fig.1, caption=Surface data and example rainfall-inundation distribution in the beijing municipal administrative center, figureFileSmall=FI4lCQz/AK0lvD5Y2XBTCA==, figureFileBig=uGZR2XluBHcbd4/8H8OPJw==, tableContent=null), ArticleFig(id=1277261258671915824, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, label=图1, caption=北京城市副中心下垫面数据与降雨积涝分布示例, figureFileSmall=FI4lCQz/AK0lvD5Y2XBTCA==, figureFileBig=uGZR2XluBHcbd4/8H8OPJw==, tableContent=null), ArticleFig(id=1277261258751607601, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, label=Fig.2, caption=Total rainfall and hourly rainfall peak values for 32 rainfall events, figureFileSmall=1KQrJVzsZd9O7llRLvqxXg==, figureFileBig=nJMxbKdMjmS0fH3854ooeA==, tableContent=null), ArticleFig(id=1277261260320277298, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, label=图2, caption=32场降雨过程的总雨量和小时降雨量峰值, figureFileSmall=1KQrJVzsZd9O7llRLvqxXg==, figureFileBig=nJMxbKdMjmS0fH3854ooeA==, tableContent=null), ArticleFig(id=1277261260404163379, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, label=Fig.3, caption=Overview of UNet-KAN-SR, figureFileSmall=UhJxdbymWkVn+Nnj55UaIg==, figureFileBig=H+VilQkwK+Nmckm9/6vSAw==, tableContent=null), ArticleFig(id=1277261260475466548, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, label=图3, caption=UNet-KAN-SR模型图

注:KAN/CNN/ResUnit(CinCoutHW),Cin表示输入通道数,Cout表示输出通道数,HW分别表示输入淹没分布图的长宽尺寸(按格点数计);NSR表示SR模块的输入通道数;UNet-KAN模块以替换3层KAN为例。

, figureFileSmall=UhJxdbymWkVn+Nnj55UaIg==, figureFileBig=H+VilQkwK+Nmckm9/6vSAw==, tableContent=null), ArticleFig(id=1277261260697764661, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, label=Fig.4, caption=Predicted 3-hour regional inundation distribution based on the UNet-KAN-SR model(Areas with water depth less than 1 cm are not shown), figureFileSmall=Q2LL0uq1OaLCoxXFB9uviw==, figureFileBig=mysJ/XGXMqS5moHPe/DhFQ==, tableContent=null), ArticleFig(id=1277261260781650742, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, label=图4, caption=基于UNet-KAN-SR模型的3小时区域积涝分布预测(水深小于1 cm的区域未绘出), figureFileSmall=Q2LL0uq1OaLCoxXFB9uviw==, figureFileBig=mysJ/XGXMqS5moHPe/DhFQ==, tableContent=null), ArticleFig(id=1277261260886508343, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, label=Fig.5, caption=Predicted 1-24 hour inundation process at historical flood-prone locations based on the UNet-KAN-SR model, figureFileSmall=Ew8ePu9w6DfzjSpVfX3deQ==, figureFileBig=ruyCl4VJdsyBuSX1+7znYg==, tableContent=null), ArticleFig(id=1277261260966200120, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, label=图5, caption=基于UNet-KAN-SR模型的历史易涝点处1~24 h积涝过程预测, figureFileSmall=Ew8ePu9w6DfzjSpVfX3deQ==, figureFileBig=ruyCl4VJdsyBuSX1+7znYg==, tableContent=null), ArticleFig(id=1277261261037503289, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, label=Fig.6, caption=Root mean square error distribution of the UNet-KAN-SR model based on different combinations of surface characteristics, figureFileSmall=gFQwQun9NcAlh45kayyGeQ==, figureFileBig=2nkHK95C5Z5JayfCJVgz/A==, tableContent=null), ArticleFig(id=1277261261125583674, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, label=图6, caption=基于不同下垫面信息组合的UNet-KAN-SR模型均方根误差分布, figureFileSmall=gFQwQun9NcAlh45kayyGeQ==, figureFileBig=2nkHK95C5Z5JayfCJVgz/A==, tableContent=null), ArticleFig(id=1277261261205275451, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, label=Table 1, caption=

The influence of replaced KAN layer numbers on the UNet-KAN prediction performance

, figureFileSmall=null, figureFileBig=null, tableContent=
KAN替换层数RMSE/cmMAE/cmPODFARCSI
01.450.220.790.050.76
11.300.170.810.030.79
21.320.180.820.030.79
31.300.180.820.030.79
41.300.130.830.030.80
), ArticleFig(id=1277261261263995708, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, label=表1, caption=

不同KAN替换层数对UNet-KAN预报性能的影响

, figureFileSmall=null, figureFileBig=null, tableContent=
KAN替换层数RMSE/cmMAE/cmPODFARCSI
01.450.220.790.050.76
11.300.170.810.030.79
21.320.180.820.030.79
31.300.180.820.030.79
41.300.130.830.030.80
), ArticleFig(id=1277261261352076093, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=EN, label=Table 2, caption=

Performance comparison of UNet-KAN-SR models based on different combinations of surface characteristics

, figureFileSmall=null, figureFileBig=null, tableContent=
下垫面组合形式初始ZLBZBBLZLZBL
RMSE/cm34.89.29.512.59.39.517.29.5
训练时间/h

66535985606988
), ArticleFig(id=1277261261419184958, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, language=CN, label=表2, caption=

基于不同下垫面输入信息组合的UNet-KAN-SR模型性能比较

, figureFileSmall=null, figureFileBig=null, tableContent=
下垫面组合形式初始ZLBZBBLZLZBL
RMSE/cm34.89.29.512.59.39.517.29.5
训练时间/h

66535985606988
)], attaches=null, journal=Journal(id=1276576005759143953, delFlag=0, nameCn=水利学报, nameEn=Journal of Hydraulic Engineering, nameHistory1=null, nameHistory2=null, issn=0559-9350, eissn=null, cn=11-1882/TV, coden=null, periodic=0, language=CN, oaType=null, 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=VQovbL6eK2UmIe7CAOCGSg==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Hydraulic Engineering, journalRemark=null, publicationField=null, createdTime=1782289037409, updatedTime=1784021719922, createdBy=18614031015, updatedBy=13041195026, firstLetterCn=S, firstLetterEn=S, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=VQovbL6eK2UmIe7CAOCGSg==, picEn=HLcP8PGWIFmFsvb9OpMK1g==, jcr=null, cjcr=null, exts=[JournalExt(id=1283843403062416007, language=CN, name=水利学报, 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=, createdTime=1784021719947, updatedTime=1784021719947, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=https://www.scicloudcenter.com/jhe, submissionEditorUrl=https://www.scicloudcenter.com/jhe, submissionReviewUrl=https://www.scicloudcenter.com/jhe, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1283843403108553352, language=EN, name=Journal of Hydraulic Engineering, 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=, createdTime=1784021719958, updatedTime=1784021719958, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=https://www.scicloudcenter.com/jhe, submissionEditorUrl=https://www.scicloudcenter.com/jhe, submissionReviewUrl=https://www.scicloudcenter.com/jhe, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1276576982599962646, websiteList=[Website(id=1276578024648736975, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1276576982599962646, 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/slxb/CN, language=CN, createTime=1782289518750, createBy=18614031015, updateTime=1782290003617, updateBy=18614031015, name=水利学报-中文, tplId=1146099689490845704, title=水利学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1276580224183702482, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024648736975, code=articleTextType, value=kx, createTime=1782290043159, updateTime=1782290043159, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580224158536655, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024648736975, code=banner, value=null, createTime=1782290043153, updateTime=1782290043153, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580224204674005, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024648736975, code=grayFlag, value=0, createTime=1782290043164, updateTime=1782290043164, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580224145953742, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024648736975, code=logo, value=https://castjournals.cast.org.cn/joweb/slxb/CN/file/pic?fileId=qMhrBHpC0+tO2zs+IdNg8g==, createTime=1782290043150, updateTime=1782290043150, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580224221451223, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024648736975, code=minRunFlag, value=0, createTime=1782290043168, updateTime=1782290043168, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580224175313873, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024648736975, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/slxb/CN/file/pic, createTime=1782290043157, updateTime=1782290043157, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580224213062614, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024648736975, code=silenceFlag, value=0, createTime=1782290043166, updateTime=1782290043166, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580224166925264, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024648736975, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1782290043155, updateTime=1782290043155, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580224192091091, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024648736975, code=themeColor, value=null, createTime=1782290043161, updateTime=1782290043161, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580224196285396, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578024648736975, code=themeStyle, value=null, createTime=1782290043162, updateTime=1782290043162, creator=18614031015, updator=18614031015)]), Website(id=1276578025089138904, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1276576982599962646, 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/slxb/EN, language=EN, createTime=1782289518854, createBy=18614031015, updateTime=1782290000347, updateBy=18614031015, name=水利学报-英文, tplId=1146101810881728533, title=Journal of Hydraulic Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1276580162732954555, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025089138904, code=articleTextType, value=kx, createTime=1782290028508, updateTime=1782290028508, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580162711983032, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025089138904, code=banner, value=null, createTime=1782290028503, updateTime=1782290028503, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580162753926078, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025089138904, code=grayFlag, value=0, createTime=1782290028513, updateTime=1782290028513, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580162703594423, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025089138904, code=logo, value=https://castjournals.cast.org.cn/joweb/slxb/EN/file/pic?fileId=qMhrBHpC0+tO2zs+IdNg8g==, createTime=1782290028501, updateTime=1782290028501, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580162766508992, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025089138904, code=minRunFlag, value=0, createTime=1782290028516, updateTime=1782290028516, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580162728760250, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025089138904, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/slxb/EN/file/pic, createTime=1782290028507, updateTime=1782290028507, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580162762314687, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025089138904, code=silenceFlag, value=0, createTime=1782290028515, updateTime=1782290028515, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580162720371641, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025089138904, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1782290028505, updateTime=1782290028505, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580162741343164, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025089138904, code=themeColor, value=null, createTime=1782290028510, updateTime=1782290028510, creator=18614031015, updator=18614031015), WebsiteProps(id=1276580162745537469, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1276578025089138904, code=themeStyle, value=null, createTime=1782290028511, updateTime=1782290028511, creator=18614031015, updator=18614031015)])], journalTitle=水利学报, weixinUrl=null, journalUrl=https://www.sciengine.com/JHE, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Hydraulic Engineering, journalPhotoCn=VQovbL6eK2UmIe7CAOCGSg==, journalPhotoEn=HLcP8PGWIFmFsvb9OpMK1g==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/slxb/CN/10.3724/j.slxb.20250474, detailUrlEn=https://castjournals.cast.org.cn/joweb/slxb/EN/10.3724/j.slxb.20250474, pdfUrlCn=https://castjournals.cast.org.cn/joweb/slxb/CN/PDF/10.3724/j.slxb.20250474, pdfUrlEn=https://castjournals.cast.org.cn/joweb/slxb/EN/PDF/10.3724/j.slxb.20250474, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1779206400000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于UNet-KAN-SR模型的城市积涝高效智能预报
收藏切换
PDF下载
陈耀明 1, 2 , 李瑞栋 1 , 陈骥 2 , 倪广恒 1
水利学报 | 2026,57(5): 704-715
收起
收藏切换
水利学报 | 2026 , 57 (5) : 704 -715
基于UNet-KAN-SR模型的城市积涝高效智能预报
全屏
[Author(id=1277261249536721680, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=chenyaom@outlook.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1277261249855488788, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261249536721680, language=EN, stringName=Yaoming CHEN, firstName=Yaoming, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China
2.Department of Civil Engineering,The University of Hong Kong,Hong Kong 999077,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1277261251503850261, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261249536721680, language=CN, stringName=陈耀明, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084
2.香港大学 土木工程系,香港 999077, bio={"content":"

陈耀明(2003—),博士生,主要从事水文水资源研究。E-mail:

"}, bioImg=null, bioContent=

陈耀明(2003—),博士生,主要从事水文水资源研究。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1277261249184400136, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=1., ext=[AuthorCompanyExt(id=1277261249196983049, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China), AuthorCompanyExt(id=1277261249205371658, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084)]), AuthorCompany(id=1277261249440252683, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=2., ext=[AuthorCompanyExt(id=1277261249448641292, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249440252683, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Department of Civil Engineering,The University of Hong Kong,Hong Kong 999077,China), AuthorCompanyExt(id=1277261249457029901, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249440252683, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.香港大学 土木工程系,香港 999077)])]), Author(id=1277261251591930647, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=lyy0744@mail.tsinghua.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1277261251671622425, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261251591930647, language=EN, stringName=Ruidong LI, firstName=Ruidong, middleName=null, lastName=LI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1277261251885531930, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261251591930647, language=CN, stringName=李瑞栋, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1277261249184400136, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=1., ext=[AuthorCompanyExt(id=1277261249196983049, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China), AuthorCompanyExt(id=1277261249205371658, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084)])]), Author(id=1277261251973612316, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=1277261252359488286, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261251973612316, language=EN, stringName=Ji CHEN, firstName=Ji, middleName=null, lastName=CHEN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.Department of Civil Engineering,The University of Hong Kong,Hong Kong 999077,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1277261252732781343, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261251973612316, language=CN, stringName=陈骥, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2.香港大学 土木工程系,香港 999077, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1277261249440252683, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=2., ext=[AuthorCompanyExt(id=1277261249448641292, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249440252683, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.Department of Civil Engineering,The University of Hong Kong,Hong Kong 999077,China), AuthorCompanyExt(id=1277261249457029901, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249440252683, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2.香港大学 土木工程系,香港 999077)])]), Author(id=1277261252825056033, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, 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=1277261253168988963, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261252825056033, language=EN, stringName=Guangheng NI, firstName=Guangheng, middleName=null, lastName=NI, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1277261253567447844, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, authorId=1277261252825056033, language=CN, stringName=倪广恒, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1277261249184400136, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, xref=1., ext=[AuthorCompanyExt(id=1277261249196983049, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China), AuthorCompanyExt(id=1277261249205371658, tenantId=1146029695717560320, journalId=1276576982599962646, articleId=1276897174982094882, companyId=1277261249184400136, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084)])])]
陈耀明1, 2 , 李瑞栋1 , 陈骥2, 倪广恒1
作者信息
  • 1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084
  • 2.香港大学 土木工程系,香港 999077
通讯作者:
李瑞栋(1997—),博士,助理研究员,主要从事城市洪涝研究。E-mail:
Efficient and intelligent forecasting of urban waterlogging based on UNet-KAN-SR modeling
Yaoming CHEN1, 2 , Ruidong LI1 , Ji CHEN2, Guangheng NI1
Affiliations
  • 1.State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China
  • 2.Department of Civil Engineering,The University of Hong Kong,Hong Kong 999077,China
出版时间: 2026-05-20 doi: 10.3724/j.slxb.20250474
文章导航
收藏切换

伴随着全球气候变化,在经历高速城镇化进程之后,城市积涝灾害趋多趋强,使得快速积涝预测成为研究热点问题。相较于传统数值模拟,基于深度学习的人工智能模拟方法能有效提升计算效率,但因GPU计算显存有限而易遭遇模型训练瓶颈。鉴于此,本研究提出了一种基于UNet-KAN-SR模型的城市积涝高效智能预报方法,先利用UNet-KAN模块学习低分辨率积涝分布的时空演变规律,再利用超分辨率SR模块与高分辨率下垫面信息,逐步将低分辨率积涝分布映射为高分辨率积涝分布,从而利用时空解耦在保障时空模拟精度的同时有效降低智能模型训练的计算资源需求。测试降雨情景的验证结果表明,UNet-KAN-SR模型能在3 min内完成未来3 h积涝分布预报,均方根误差为9 cm、命中率达0.86,具备较高的计算精度与效率。相较于已有智能模型常用的卷积层,UNet-KAN-SR模型引入的KAN层可显著提升模型对积涝时空演变过程的非线性建模能力,使均方根误差降低10%。此外,本研究发现高分辨率下垫面信息能显著提升智能模型对积涝分布的预测能力且不同类型的下垫面信息所能取得的提升幅度相近,由此说明在构建智能模型时,可通过优选输入下垫面特征组合,有效提升模型训练速度并控制建模成本,实现高效智能预报。

城市内涝预测  /  深度学习  /  时空预测  /  超分辨率模型  /  北京城市副中心

With global climate change and accelerating urbanization, urban waterlogging disasters have become increasingly frequent and severe, making rapid waterlogging forecasting a key research focus. Compared with traditional numerical simulation methods, deep-learning-based artificial intelligence (AI) models can significantly improve computational efficiency. However, they often encounter training bottlenecks due to limited GPU memory. To address this, this study proposes an efficient AI urban waterlogging forecasting model named as UNet-KAN-SR. This model first employs the UNet-KAN module to efficiently simulate the spatio-temporal evolution of waterlogging over low-resolution grids, and then leverages the SR (super-resolution) module, along with high-resolution surface information, to progressively map the low-resolution waterlogging distribution to high-resolution distribution. This spatiotemporal decoupling strategy can ensure simulation accuracy while substantially reducing the computational resources required for training AI models. Experimental results demonstrate that the UNet-KAN-SR model can simulate a 3-hour waterlogging distribution within 3 minutes, achieving a root mean square error (RMSE) of 9 cm and a probability of detection (POD) of 0.84, demonstrating high accuracy and computational efficiency. Further analysis reveals that the integration of the KAN module can significantly enhance the model’s capability to capture nonlinear flood dynamics when compared with common CNN modules, reducing RMSE by 10%. Furthermore, this study finds that incorporating high-resolution features, such as surface topography, building coverage ratio, and land use, can significantly improve the simulation performance but performance improvement is similar under different feature combinations. This indicates that by optimizing the combination of input features during AI model construction, training speed can be enhanced, modeling costs controlled, and efficient intelligent forecasting achieved.

urban waterlogging forecasting  /  deep learning  /  spatio-temporal prediction  /  super-resolution model  /  Beijing municipal administrative center
陈耀明, 李瑞栋, 陈骥, 倪广恒. 基于UNet-KAN-SR模型的城市积涝高效智能预报. 水利学报, 2026 , 57 (5) : 704 -715 . DOI: 10.3724/j.slxb.20250474
Yaoming CHEN, Ruidong LI, Ji CHEN, Guangheng NI. Efficient and intelligent forecasting of urban waterlogging based on UNet-KAN-SR modeling[J]. Journal of Hydraulic Engineering, 2026 , 57 (5) : 704 -715 . DOI: 10.3724/j.slxb.20250474
全球城市极端暴雨事件趋多趋强1,使得极端暴雨下洪涝非线性响应机制成为当前水文科学探索和技术应用领域的热点问题2。其中,精准高效的城市内涝模型可动态推演地表淹没过程,为灾害应急管理、公共安全保障提供科学依据。然而,现有积涝模型高度依赖于水文水动力方程的并行化数值求解3,使得模型计算受限于如图形处理单元(Graphics Processing Unit,GPU)等特定高性能计算设备4,仍然无法有效满足临灾快速预警要求5。因此,突破传统数值模拟的效能瓶颈、实现快速积涝推演仍然是城市洪涝研究的前沿问题。
近年来,以深度学习为代表的人工智能模拟方法为城市积涝高效预报带来了崭新思路6。相较于经典数值模型,该类方法常以完成参数验证的数值模型为参考目标,采用以卷积神经网络(Convolutional Neural Networks,CNN)、长短期记忆网络(Long Short Term Memory,LSTM)等为代表的深度神经网络,自动学习数值模型输入与输出间的映射关系,有效规避了数值建模的复杂计算逻辑与稳定性约束条件,能有效提升模拟效率。如林凯荣等7选取25.45 km²的深圳市大空港新城区,构建融合CNN、LSTM及注意力机制的智能模型,实现了18 s内完成3 h设计暴雨计算。李玉龙等8选取郑州市金水区西部片区,以多重现期、多雨型情景下城市水文水动力耦合模型计算结果为样本数据训练基于CNN的智能模型,实现了区域最大积水深度的高精度快速预测。Bian等9利用时空深度学习模型ConvLSTM,快速模拟沿海区域洪水动态淹没过程,为城市洪涝风险早期预警提供高效智能工具。然而,由于地表积涝发展过程复杂、模拟单元数量众多,为了学习大量单元上流场的时空演变特征,智能模型往往需要采用输出维度较高、隐藏层参数较多的深度神经网络,使得用于相应模型训练的GPU计算资源消耗极高10。因此,如何实现高效智能模拟仍是当前研究的热点问题之一。
考虑到低分辨率模拟占用计算资源相对较少并包含整体流场发展的必要信息,He等11提出以低分辨率流场为输入,利用深度学习模型进行超分辨率(Super-Resolution,SR)计算,形成满足目标应用需求的高分辨率流场。如Wang等12利用旋转经验正交函数提取洪水演进时空特征,构建了融合低精度洪水模型与稀疏高斯过程的混合智能模型,实现了11 min内完成120 h溃堤洪水模拟。在考虑低分辨率流场作为输入的同时,也可以使用图片分块方法来降低显存消耗13。图片分块方法将大的图像分解成多个小块来处理,以减少一次性加载到显存中的数据量,从而降低计算资源的需求。Herath等14提出了融合物理约束与深度学习的次网格UNet模型(Subgrid UNet,SGUnet),模型通过将大尺度洪水深度图按512×512像素进行图片分块处理,在降低显存与计算负担的同时保持对局部地形特征的敏感性。然而,分块方法在保持水文连通性方面存在潜在不足。由于每个分块的计算相互独立,相邻区域间的流动关系可能被割裂,导致积涝演进过程中的全局水动力特征无法被充分捕捉。虽然通过减小分块尺寸可以进一步降低显存占用,但目前尚缺乏系统性研究探讨分块大小变化对模型预测精度的影响。此外,改进用于非线性拟合的神经网络基础模块也是提升智能建模效率的重要方向。Liu等15提出了基于Kolmogorov-Arnold表示定理的KAN(Kolmogorov-Arnold Networks)模块,创新采用激活函数学习范式,显著提升了深度神经网络对复杂非线性关系的拟合效率。近年来,KAN已逐渐被应用于河流水位预测16、淹没范围提取17等任务,展现出优于经典深度神经网络架构的计算效能。
基于前述城市积涝智能预报研究进展,本论文以北京城市副中心为研究区域,提出了一种基于两阶段时空解耦学习的城市积涝时空预报模型UNet-KAN-SR,通过分别学习低分辨率积涝过程发展、低分辨率至高分辨率积涝场的映射规律,实现3 min快速准确预报3 h积涝过程。
北京城市副中心是承担北京市政府非首都功能疏解和京津冀协同发展的重要战略载体18。北京城市副中心地处北运河流域冲积平原,总面积为307 km2,地势相对平坦,平均坡度约为1.5%,下垫面不透水率与建筑物面积占比较高(图1(b)(c))。北京城市副中心属于温带大陆性半湿润季风气候,年平均降雨量为546.8 mm且夏季降水约占全年降水的70%以上。作为“九河下梢”,北京城市副中心处于北京市多条河流下游位置,地势相对低洼平坦,排水能力不足,存在较高内涝风险19。依据地方政府提供的内涝监测资料,该区域历史易涝点主要集中在下凹式立交桥、铁路桥等地(图1(a))。
本研究选用美国气候预测中心CMORPH卫星降水降尺度产品和随机暴雨移置方法20,随机生成32场时空分辨率分别为1 h、1 km的24 h降雨过程。这32场降雨过程的统计特征总结如图2所示,其中总降雨量分布在150至210 mm之间,最大小时雨强介于50至110 mm/h之间,涵盖了不同重现期的降雨情景。上述降雨过程被作为气象驱动数据,输入积涝数值模型,进行城市积涝数值模拟计算,从而获取高分辨率的地表淹没训练样本。积涝数值模型选用自主研发的城市水文水动力耦合模型,具体计算原理与建模过程详见李瑞栋等21。高程数据(图1(a))由大疆禅思L2系统搭载激光雷达测量采集,原始点云密度约为每平方米100点。土地利用数据(图1(b))使用ESRI公司发布的2020年10 m分辨率全球土地利用数据集22。建筑物面积数据(图1(c))选用Shi等23发布的东亚国家建筑物矢量边界数据集,并利用GIS渔网分析工具将其重采样至2 m分辨率。由于原数值模型采用非结构三角网格离散求解域,本研究利用GIS渔网分析工具,将三角网格存储的模拟水深结果重采样至30 和100 m分辨率(图1(e)(f)),得到图像大小分别为880×726264×218的高、低分辨率淹没分布序列作为参考数据集,用于后续智能模型训练。
城市积涝预报属于典型的时空序列预测问题,即针对一场按固定时间间隔划分的降雨积涝过程,以当前时刻t为起点,依次输入前序M个时间步的淹没分布图xt-ii=0, ,M-1与相关辅助信息A,利用计算模型输出未来时段N个时间步的淹没分布图x^t+ii=1, ,N,具体可表示为:
x^t+N, , x^t+1=(xt , ,xt-M+1,A)
考虑到常见降水短临预报的预见期约为0~2 h24,本研究设定淹没训练样本的时间间隔为10 min,并设定输入序列步数M=6与输出序列步数N=6,由此实现0~1 h积涝预报。具体而言,模型利用前1 h(6×10 min)的淹没分布序列作为初始驱动,预测后1 h的积涝演变,并采取滚动预报策略(即将首轮预报的末期结果更新至下一轮的输入数据),递推生成更长预见期的淹没分布图。
UNet-KAN-SR模型采用两阶段串联结构,分别由UNet-KAN模块与SR模块组成(图3(a))。针对已有时空序列学习方法因需联合考虑时空演变规律而面临的模型训练显存急剧上升这一问题25-26,UNet-KAN-SR模型首先采用UNet-KAN模块推演100 m分辨率淹没分布图的时空演变过程,即以历史6步淹没图与当前步降雨组合为输入,输出未来6步淹没图,由此规避因直接预测精细化淹没分布演变过程而产生的高昂计算开支;在此基础上,UNet-KAN-SR模型进一步采用SR模块,学习单时间步内淹没分布自100 m分辨率映射至30 m分辨率的统计降尺度规律,由此逐一计算得到各个目标时刻的30 m分辨率淹没分布。相较于经典时空深度学习模型,UNet-KAN-SR模型将积涝时空模拟分解为两阶段任务,使得UNet-KAN模块与SR模块可以进行解耦训练(即分别采用各自的输入-输出数据组合训练),实现在保障时空建模精度的同时有效降低模型训练所需的计算资源。
为了高效表征城市积涝过程的非线性发展规律,UNet-KAN-SR模型采用基于卷积神经网络UNet27架构的UNet-KAN模块,预测未来1 h、时间分辨率为10 min、空间分辨率为100 m的城市积涝演变过程。该模块采用典型的编码器-解码器结构,并通过跳跃连接机制,将编码器各层提取的多尺度特征图直接传递至解码器对应层,与上采样后的特征图进行特征拼接,进而联合进行卷积运算以实现高精度的空间信息还原。在此基础上,UNet-KAN模块将UNet结构内部分上、下采样阶段的CNN层替换为相应的KAN层(图3(b))。具体而言,CNN层通过考虑空间局部关系的多层卷积计算融合历史淹没与降雨图像蕴含的特征信息,并利用非线性激活函数提升对特征间非线性作用关系的拟合能力。在此基础上,KAN层28首先采用较大空间感受野的深度可分离卷积计算提取积涝分布空间特征并将其展平后,利用KANLayer提取各个时空位置间的全局关联关系,再利用点卷积计算完成时空特征融合。其中,在CNN常用的非线性激活函数A(z)基础上,KANLayer进一步引入了参数化B样条函数Bi(z)组合(图3(d)):
KAN(z)=ωbA(z)+ωsiciBi(z)
式中:z为KANLayer的输入向量;ωbωsci为可学习的权重系数;A(z)默认取为SiLU激活函数:
Az=z1+e-z
由于KAN层采用可学习的基函数组合提取全局非线性特征,其非线性建模能力预期优于仅考虑局部空间特征的CNN模块,故UNet-KAN模块依次连续运用CNN与KAN模块组合,逐级强化对积涝发展时空特征的学习,并由此输出为未来6步(即预见期为1 h)积涝分布图序列。针对预见期1 h以外的积涝过程预报,UNet-KAN模块采用滚动预报策略,即利用首轮预报的末期结果作为新输入,依次递推生成未来2、3 h等时段预测。
在利用UNet-KAN模块得到逐10 min、100 m分辨率的淹没分布后,UNet-KAN-SR模型进一步利用SR模块与多源下垫面信息(图3(a)和(c)),将其降尺度至30 m分辨率。与UNet-KAN模块类似,SR模块以UNet为基础结构,主要由编码器、解码器和跳跃连接组成。其中,编码器利用5层残差卷积单元(ResUnit)及相应的升尺度池化计算模块,自局部至整体提取下垫面环境与积涝分布间的空间关联特征。相似地,解码器利用5层ResUnit及相应的降尺度逆卷积操作,逐级融合来自编码器与跳跃连接的多尺度信息,最终通过3×3卷积、1×1卷积及非线性激活函数,输出相应时间步的30 m分辨率淹没图。为科学引导SR模块利用粗分辨率积涝信息推理得到细分辨率积涝分布,本研究参考影响城市产汇流过程的区域下垫面特征,引入高程、建筑物面积比例、土地利用类型等基底信息,并利用栅格重采样计算,将其与低分辨率积涝分布共同插值至与高分辨率积涝分布一致的栅格坐标系,由此融合得到SR模块的输入张量,用于支撑后续超分辨率计算。
UNet-KAN-SR模型训练分为UNet-KAN模块和SR模块。两个模块均采用均方误差作为损失函数,采用Adam优化器更新参数,批处理大小为2,最大迭代次数设置为100次,并采用基于验证集损失的自适应学习率调度器,使得优化器能在模型性能指标停滞时自动降低学习率,加速参数细化寻优。数据集划分比例采用训练集:验证集:测试集为24∶3∶5。经统计,按图3结构和前述批处理大小配置,UNet-KAN模块参数量约为1081万,SR模块参数量约为169万,两者训练时分别需消耗15 GB大小的GPU显存容量。相比于已有研究常用的时空深度学习模型ConvLSTM9,在相似模型参数量(479万)与同等训练配置的条件下,ConvLSTM模型训练所需的GPU显存容量增长为22 GB,由此可见本研究提出的UNet-KAN-SR模型可有效节约模型训练所需的GPU显存容量。
本研究采用均方根误差(RMSE),平均绝对误差(MAE)、命中率(POD)、误警率(FAR)、关键成功指数(CSI)、峰值偏差百分比(PBAIS)等指标评估UNet-KAN-SR模型用于积涝时空模拟的性能。相关指标的具体计算方法如下:
RMSE=1ni=1n(yi-yi^)2
MAE=1ni=1n|yi-yi^|
POD=TPTP+FN
FAR=FPTP+FP
CSI=TPTP+FP+FN
PBAIS=|PF-PS|PF
式中:yi为参考值;yi^为预测值;n为逐时间步、逐像元统计的样本数;TPFPFN分别为以15 cm为水深阈值29分类统计得到的命中、漏报、空报像元数;PF为预报峰值;PS为模拟峰值。具体而言,当预测值与参考值在同一像元中均大于15 cm时,判定为命中像元;当预测值小于15 cm、参考值大于15 cm时,判定为漏报像元;当预测值大于15 cm、参考值小于15 cm时,判定为空报像元。
从随机生成的32场24 h降雨序列中,选取测试集5场24 h降雨事件,按10 min为起报时间间隔,采用UNet-KAN-SR模型滚动预报各个起报时刻后3 h的积涝分布情况,并与作为参考值的数值模拟结果进行对比,由此评估UNet-KAN-SR模型积涝预报精度。经统计,UNet-KAN-SR模型平均RMSE为9 cm、MAE为2 cm、POD为0.86、FAR为0.31、CSI为0.60,由此说明UNet-KAN-SR模型能命中区域内绝大部分的高积水深点位,综合性能相对较为良好。为评估模型计算效率,本研究采用计算配置为:GPU为Tesla V100S-PCIE-32GB,CPU为Intel(R) Xeon(R) Gold 6226R CPU @ 2.90 GHz。经统计,对于3 h降雨积涝过程计算,UNet-KAN-SR模型平均用时约为3 min,能实现快速城市积涝风险预警,从而为相关应急处置措施的实施预留充足时间。
图4展示了在5场测试集中选取的1场测试降雨情景(总降雨量为209 mm,小时降雨量峰值为73 mm/h)下UNet-KAN-SR模型以T=5 h为起报时刻、预报得到的区域积涝演变过程。经统计,UNet-KAN-SR模型平均RMSE为9 cm、平均POD为0.85且变化幅度不大,说明针对0~3 h积涝预报任务,该模型能较稳定地保持相对可靠的综合性能,预测得到绝大多数高水深区域淹没分布情况,故具有较好的高风险区域预警能力。
考虑到历史易涝点位属于城市积涝风险预警的重点对象,本研究进一步检验了前述测试降雨情景下UNet-KAN-SR模型在相关点位处的积涝预报情况(图5),其中降雨为随机生成的降雨序列中各易涝点的降雨过程。经统计,UNet-KAN-SR模型在各历史易涝点处的RMSE平均值约为25 cm、PBAIS平均值约为19%,说明UNet-KAN-SR模型在高风险点位处仍能维持相对可靠的积涝预报精度,可较好地预见历史易涝点位处峰值积水深情况,由此为积涝风险预警提供有价值的信息。由于历史易涝点积水深度显著高于区域平均水平,导致模型在极端水位变化下的预测误差相比于模型平均误差相对放大。
为检验KAN模块学习积涝时空演变过程非线性特征的优越性,本研究进一步以经典CNN模块为基准,比较在UNet-KAN模块内替换不同层数KAN(图3(b))对预报性能的影响。基于测试集降雨事件,表1总结了模型进行1 h积涝预报的性能统计结果:
表1可知,适当引入KAN层有助于提升模型的预报精度。特别地,当仅替换1层CNN为KAN时,UNet-KAN模型在多个指标方面均取得显著提升,如RMSE从1.45降至1.30 cm、MAE从0.22降至0.17 cm,由此说明KAN层能有效增强模型对积涝时空演变过程的非线性特征建模能力。然而,随着KAN替换层数进一步增加,模型性能的提升效果逐渐减弱。因此,即使KAN层能有效增强智能模型的非线性建模能力,但仍应结合具体实验分析,确定适当的KAN替换层数,实现智能模型非线性拟合与泛化能力间的最优平衡。
地表高程、土地利用类型、建筑物等下垫面条件对城市积涝涉及的产汇流过程具有不可忽视的影响,属于城市积涝数值模拟的重要输入数据30。为分析下垫面特征选取对UNet-KAN-SR模型积涝模拟性能的影响,本研究采取对比实验方法,选用高程Z、建筑物面积比例B和土地利用类型L的不同组合作为SR模块的辅助输入变量并训练相应模块,统计不同下垫面特征组合下UNet-KAN-SR模型的训练时间与均方根误差分布如表2所示。
表2可知,在不引入任何下垫面信息的初始条件下,UNet-KAN-SR模型的RMSE为35 cm,说明若仅以低分辨率积涝分布为输入,智能模型仍难以直接预测相应的高分辨率积涝分布情况。然而,通过引入地表高程、建筑物面积比例和土地利用类型下垫面信息,UNet-KAN-SR模型的RMSE下降至10 cm以内,说明高分辨率下垫面信息能显著提升智能模型的积涝预测能力。图6进一步比较了不同下垫面信息组合下模型RMSE的分布情况。从图中可以看出,尽管在不同信息组合的RMSE上下四分位数范围存在一定波动,但大部分组合RMSE中位数稳定在10 cm附近,说明UNet-KAN-SR模型能凭借其强大的非线性拟合能力,自动学习归纳得到任一下垫面信息与相应分辨率积水深分布间的内在关联规律,从而有效提升积涝预测性能。从实验结果中可以看出,单独引入任一类型的下垫面信息均能获得较为理想的预测效果。这一现象的原因可能在于,模型通过引入KAN模块,显著增强了对非线性特征的建模能力。KAN层通过采用更为复杂的非线性激活函数,能够有效捕捉输入特征之间的全局关联,从而使得即使是较为简化的输入特征,亦能够在模型中发挥重要作用。考虑到引入过量输入特征会显著增加模型训练时间(表2),故通过前述特征优选,能有效缩短智能模型训练时间,提升建模效率。同时,由于高分辨率DEM制备成本相对高昂且地理覆盖范围有限31、土地利用32及建筑物面积占比23等下垫面特征已形成基于高分辨率光学影像的可开放获取产品,故在构建相关智能模型时,可优先考虑选用高分辨率土地利用、建筑物面积占比等信息作为输入,预期能显著降低智能模型构建成本。
针对城市积涝高效预报,本文提出了一种基于两阶段时空解耦学习的UNet-KAN-SR模型,即先利用UNet-KAN模块学习100 m分辨率积涝演变过程,后利用SR模块将其细化映射至30 m分辨率积涝分布,并成功应用于北京城市副中心,取得如下主要结论:
(1)聚焦城市积涝智能预报面临的计算显存瓶颈,UNet-KAN-SR模型解耦训练UNet-KAN模块与SR模块,使得模型无需联合学习30 m分辨率积涝过程的时间演变与空间分布特征,有效降低了计算资源需求。基于测试集降雨事件的评估结果表明,UNet-KAN-SR模型的平均RMSE为9 cm、POD为0.86,仅需3 min即可完成3 h积涝事件计算,故能保障快速可靠的城市积涝预报。
(2)通过对比不同KAN替换层数下UNet-KAN-SR模型性能,本文发现KAN层可有效显著提升智能模型对积涝时空过程的非线性特征建模能力,可降低10%的RMSE。但过度堆叠KAN层易诱导智能模型产生过拟合问题,使得模型性能出现不同程度的下降,故仍应结合具体问题分析,寻求智能模型非线性拟合与泛化能力间的最优平衡。
(3)通过对比不同下垫面信息组合下UNet-KAN-SR模型性能,本文发现高分辨率下垫面信息能显著提升智能模型对于积涝分布的预测能力,但相关预测性能受下垫面信息类型影响较小,因此可通过输入特征优选,有效提升智能模型训练效率,避免因高分辨率地形信息依赖而造成的高昂成本,实现高效智能建模。
需要承认的是,尽管UNet-KAN-SR模型已可对30 m分辨率积涝过程实现高效智能预报,可初步满足城市积涝风险动态研判需求33,但考虑到城市下垫面环境所蕴含的复杂微地形结构,未来研究应聚焦于5 m分辨率乃至更精细尺度的积涝模拟预报,进一步检验UNet-KAN-SR模型及其时空解耦学习策略在提升智能模型效能方面的优越性。此外,为完善不同网络结构在相同计算预算下的比较,后续工作还将尝试引入其他深度学习模块(如Transformer 等模块)替代传统卷积模块,系统评估其在洪涝模拟精度、计算效率与显存消耗方面的综合表现,从而为不同类型模型在城市洪涝预报任务中的适用性提供更全面的认识。

参考文献 引证文献
排序方式:
[1]
苑希民,兰卓青,王丽娜,等. 极端天气城市特大暴雨洪涝灾害特征再分析及应对策略[J]. 水利学报,2024,55(11):1298-1308.
YUAN X M,LAN Z Q,WANG L N,et al. Re-analysis of the characteristics and response strategies of extreme weather urban heavy rainstorm flood disasters[J]. Journal of Hydraulic Engineering,2024,55(11):1298-1308. (in Chinese)
[2]
水文科技热点[J]. 水文,2024,44(6):118.
Hydrological science and technology hotspots[J]. Hydrology,2024,44(6):118. (in Chinese)
[3]
侯精明,张兆安,马利平,等. 基于GPU加速技术的非结构流域雨洪数值模型[J]. 水科学进展,2021,32(4):567-576.
HOU J M,ZHANG Z A,MA L P,et al. A GPU-accelerated unstructured basin rainfall-runoff numerical model[J]. Advances in Water Science,2021,32(4):567-576. (in Chinese)
[4]
张大伟,冯新政,林文青,等. 国内外不同GPU用于洪水模拟时加速性能对比分析[J]. 中国水利水电科学研究院学报(中英文),2024,22(5):530-538.
ZHANG D W,FENG X Z,LIN W Q,et al. Comparative analysis of acceleration performance of different GPUs in flood simulation[J]. Journal of the Chinese Institute of Water Resources and Hydropower Research,2024,22(5):530-538. (in Chinese)
[5]
AURELI F,PROST F,VACONDIO R,et al. A GPU-accelerated shallow-water scheme for surface runoff simulations[J]. Water,2020,12(3):637.
[6]
BENTIVOGLIO R,ISUFI E,JONKMAN S N,et al. Deep learning methods for flood mapping:a review of existing applications and future research directions[J]. Hydrology and Earth System Sciences Discussions,2022,26(16):4345-4378
[7]
林凯荣,欧阳佳娜,马旭民,等. 基于深度学习的城市内涝积水水深预测模型[J]. 水资源保护,2025,41(1):56-63.
LIN K R,OUYANG J N,MA X M,et al. Deep learning-based model for predicting urban waterlogging depth[J]. Water Resources Protection,2025,41(1):56-63. (in Chinese)
[8]
李玉龙,梅超,刘家宏,等. 基于数值模型与机器学习的城市内涝智能模拟[J/OL]. 水利水电技术(中英文),1-15[2025-12-17].
LI Y L,MEI C,LIU J H,et al. Intelligent simulation of urban waterlogging based on numerical models and machine learning[J/OL]. Hydraulic and Hydropower Engineering,1-15[2025-12-17]. (in Chinese)
[9]
BIAN W C,FANG J Y,WANG P,et al. Deep learning surrogate models for spatiotemporal prediction of coastal flooding inundations in Tianjin,China[J]. Journal of Hydrology:Regional Studies,2025,60:102593.
[10]
康永德,康爱泽,侯精明,等. 基于GPU的洪涝模型计算效率提升方法研究[J]. 中国水利,2025(12):64-72.
KANG Y D,KANG A Z,HOU J M,et al. Research on methods to improve computational efficiency of flood models based on GPU[J]. China Water Resources,2025(12):64-72. (in Chinese)
[11]
HE J,ZHANG L M,XIAO T,et al. Deep learning enables super-resolution hydrodynamic flooding process modeling under spatiotemporally varying rainstorms[J]. Water Research,2023,239:120057.
[12]
WANG R K,LIAN J J,YUAN X M,et al. Rapid simulation of floods by considering the spatial and temporal characteristics of inundation[J]. International Journal of Disaster Risk Science,2025,16(3):1-15.
[13]
GUO Z F,LEITAO J P,SIMÕES N E,et al. Data‐driven flood emulation:Speeding up urban flood predictions by deep convolutional neural networks[J]. Journal of Flood Risk Management,2021,14(1):e12684.
[14]
HERATH H M V V,MARSHALL L,SAHA A,et al. Subgrid informed neural networks for high-resolution flood mapping[J]. Journal of Hydrology,2025,660:133329.
[15]
LIU Z M,WANG Y X,VAIDYA S,et al. Kan:Kolmogorov-arnold networks[J]. arXiv,2024:2404.19756.
[16]
REN D F,HU Q,ZHANG T D. EKLT:Kolmogorov-Arnold attention-driven LSTM with Transformer model for river water level prediction[J]. Journal of Hydrology,2025,649:132430.
[17]
WANG C,ZHANG X H,LIU L W. FloodKAN:Integrating kolmogorov-arnold networks for efficient flood extent extraction[J]. Remote Sensing,2025,17(4):564.
[18]
沈佳坤,曹婷. 北京城市副中心创新发展路径研究[J]. 时代经贸,2025,22(3):132-135.
SHEN J K,CAO T. Research on the innovative development path of the sub-center of Beijing[J]. Times of Economy & Trade,2025,22(3):132-135. (in Chinese)
[19]
刘青云,王慧彦. 北京市通州区洪涝灾害风险评估及区划研究[J]. 水利规划与设计,2022(10):76-82.
LIU Q Y,WANG H Y. Flood disaster risk assessment and zoning study in Tongzhou District,Beijing[J]. Water Resources Planning and Design,2022(10):76-82. (in Chinese)
[20]
周正正,刘曙光,WRIGHT D B. 基于随机暴雨移置方法的城市设计暴雨分析[J]. 水科学进展,2020,31(4):583-591.
ZHOU Z Z,LIU S G,WRIGHT D B. Urban design storm analysis based on the stochastic storm transposition method[J]. Advances in Water Science,2020,31(4):583-591. (in Chinese)
[21]
李瑞栋,倪广恒,陆科成,等. 北京“23·7”特大暴雨影响下城市交通应急响应时间动态分析:以北京市通州区为例[J]. 应用基础与工程科学学报,2024,32(2):365-377.
LI R D,NI G H,LU K C,et al. Dynamic analysis of urban traffic emergency response time under the impact of the Beijing "23·7" extreme rainstorm:A case study of Tongzhou District,Beijing[J]. Journal of Basic Science and Engineering,2024,32(2):365-377. (in Chinese)
[22]
KARRA K,KONTGIS C,STATMAN-WEIL Z,et al. Global land use/land cover with Sentinel 2 and deep learning[C]// 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS. 2021
[23]
SHI Q,ZHU J J,LIU Z Y,et al. The last puzzle of global building footprints—Mapping 280 million buildings in East Asia based on VHR images[J]. Journal of Remote Sensing,2024,4:138.
[24]
杨志达. 基于连续性约束交叉相关(COTREC)的降水短临预报方法改进研究[D]. 兰州:兰州大学,2024.
YANG Z D. Research on the Improvement of Precipitation Nowcasting Method Based on Continuity-Constrained Cross-Correlation (COTREC)[D]. Lanzhou:Lanzhou University,2024. (in Chinese)
[25]
BRACHT E,KINDRATENKO V,BRUNNER R J. Sparse spatio-temporal neural network for large-scale forecasting[C]// 2022 IEEE International Conference on Big Data (Big Data). Osaka,2022.
[26]
HAN D,CHOO M,IM J,et al. Precipitation nowcasting using ground radar data and simpler yet better video prediction deep learning[J]. GIScience & Remote Sensing,2023,60(1):2203363.
[27]
RONNEBERGER O,FISCHER P,BROX T. U-net:Convolutional networks for biomedical image segmentation[C]// International Conference on Medical Image Computing and Computer-assisted Intervention. Cham:Springer international publishing,2015
[28]
LI C X,LIU X Y,LI W Y,et al. U-kan makes strong backbone for medical image segmentation and generation[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2025,39(5):4652-4660.
[29]
王丽晶,潘兴瑶,赵飞,等. 北京城市积水内涝风险地图编制与应用[J]. 北京水务,2022(6):24-29,43.
WANG L J,PAN X Y,ZHAO F,et al. Compilation and application of the urban floodwater and waterlogging risk map for Beijing[J]. Beijing Water,2022(6):24-29,43. (in Chinese)
[30]
徐宗学,叶陈雷. 城市暴雨洪涝模拟:原理、模型与展望[J]. 水利学报,2021,52(4):381-392.
XU Z X,YE C L. Urban rainstorm flood simulation:Principles,models,and prospects[J]. Journal of Hydraulic Engineering,2021,52(4):381-392. (in Chinese)
[31]
杨燈,许健,陈德清,等. 激光雷达技术流域地形构建应用现状与展望[J]. 南水北调与水利科技(中英文),2024,22(S1):127-131,135.
YANG D,XU J,CHEN D Q,et al. Application status and prospects of LiDAR technology in basin topography construction[J]. South-to-North Water Transfers and Water Science & Technology,2024,22(S1):127-131,135. (in Chinese)
[32]
LI Z H,HE W,CHENG M F,et al. SinoLC-1:The first 1-meter resolution national-scale land-cover map of China created with the deep learning framework and open-access data[J]. Earth System Science Data Discussions,2023,2023:1-38.
[33]
李国一,刘家宏. 基于TELEMAC-2D模型的深圳洪涝风险评估[J]. 水资源保护,2022,38(5):58-64.
LI G Y,LIU J H. Flood risk assessment of Shenzhen based on TELEMAC-2D model[J]. Water Resources Protection,2022,38(5):58-64. (in Chinese)
2026年第57卷第5期
PDF下载
203
93
引用本文
BibTeX
文章信息
doi: 10.3724/j.slxb.20250474
  • 接收时间:2025-08-16
  • 首发时间:2026-06-25
  • 出版时间:2026-05-20
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-08-16
基金
作者信息
    1.清华大学 水圈科学与水利工程全国重点实验室,北京 100084
    2.香港大学 土木工程系,香港 999077

通讯作者:

李瑞栋(1997—),博士,助理研究员,主要从事城市洪涝研究。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/slxb/CN/10.3724/j.slxb.20250474
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