Article(id=1156983791462998263, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156983783787421903, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2401851, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1710432000000, receivedDateStr=2024-03-15, revisedDate=1731945600000, revisedDateStr=2024-11-19, acceptedDate=null, acceptedDateStr=null, onlineDate=1753776031604, onlineDateStr=2025-07-29, pubDate=1739808000000, pubDateStr=2025-02-18, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1753776031604, onlineIssueDateStr=2025-07-29, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1753776031604, creator=13701087609, updateTime=1753776031604, updator=13701087609, issue=Issue{id=1156983783787421903, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='5', pageStart='1753', pageEnd='2192', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1753776029774, creator=13701087609, updateTime=1769691857141, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1223739602251436918, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156983783787421903, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1223739602251436919, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1156983783787421903, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=2127, endPage=2134, ext={EN=ArticleExt(id=1156983794000552187, articleId=1156983791462998263, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Prediction of Shared Bicycle Inflow and Outflow Based on Conv3D-GRU Neural Network, columnId=1156262728772735295, journalTitle=Science Technology and Engineering, columnName=Papers·Traffics and Transportations, runingTitle=null, highlight=null, articleAbstract=

Accurately predicting bike-sharing flow is essential for optimizing the supply-demand balance of shared bikes and enhancing urban residents’ travel convenience. To address the issues of low prediction accuracy and insufficient capture of spatiotemporal characteristics in bike-sharing flow prediction, a hybrid convolutional-recurrent neural network (Conv3D-GRU) model was proposed. Using Chicago’s 2022 full-year bike-sharing data, experiments were conducted, and the results were compared with those of the 3D convolutional neural network (3D-CNN) model and the convolutional long short-term memory (ConvLSTM) model. The model performance was evaluated using root mean squared error (RMSE), mean absolute error (MAE), and the coefficient of determination (R2). Experimental results show that compared with the 3D-CNN and ConvLSTM models, Conv3D-GRU is improved by 3.25%, 4.90%, 1.14% and 11.94%, 13.70% and 2.46% on RMSE, MAE and R2, respectively. This demonstrates that the Conv3D-GRU model has lower prediction errors and higher prediction accuracy, making it an effective and reliable approach for forecasting bike-sharing inflow and outflow.

, correspAuthors=Ying-ying LÜ, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, authorCompany=null, fund=null, authors=null, authorsList=Xian-guang JIA, Huan LIU, Chao-qin FENG, Ying-ying LÜ), CN=ArticleExt(id=1156983851886141734, articleId=1156983791462998263, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=基于混合卷积-递归神经网络的共享单车出入流预测, columnId=1156262730664366426, journalTitle=科学技术与工程, columnName=论文·交通运输, runingTitle=null, highlight=null, articleAbstract=

准确预测共享单车流量有助于优化共享单车的供需平衡,提高城市居民的出行便利性。为解决共享单车预测准确性不高以及时空特性捕捉不充分的问题,提出了一种混合卷积-递归神经网络(hybrid convolutional-recurrent neural network)Conv3D-GRU模型,采用芝加哥2022全年共享单车数据进行实验,并与三维卷积神经网络3D-CNN(3D convolutional neural network)模型和卷积长短期记忆网络(Convolutional long short-term memory,ConvLSTM)的预测结果进行比较,使用均方根误差(root mean squared error,RMSE)、平均绝对误差(mean absolute error,MAE)、决定系数R2评估模型性能。实验结果表明,Conv3D-GRU相较于3D-CNN和ConvLSTM模型,在RMSE、MAE以及R2上分别提高了3.25%、4.90%、1.14%和11.94%、13.70%、2.46%,可见Conv3D-GRU模型的预测误差小,预测精度高,能够有效和可靠地适用于共享单车出入流的预测。

, correspAuthors=吕英英, authorNote=null, correspAuthorsNote=
*吕英英(1982—),女,汉族,山西临汾人,硕士,讲师。研究方向:智能交通与大数据。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=M8QqBlhIJuFHdaySDYJKuA==, magXml=bfLTnhLQ2GWwiT+Ldft/3w==, pdfUrl=null, pdf=hicBUoWrHbVY5CYKLCdABg==, pdfFileSize=5074546, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=f4Y+2Qo2ppx4l3cvwBvZ5w==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=AI3gr6OlHf2PF/1KMx8Ocg==, mapNumber=null, authorCompany=null, fund=null, authors=

贾现广(1977—),男,汉族,河南浚县人,硕士,副教授。研究方向:智能交通与大数据。E-mail:

, authorsList=贾现广, 刘欢, 冯超琴, 吕英英)}, authors=[Author(id=1225467180930744335, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=jxg@kust.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1225467181148848166, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, authorId=1225467180930744335, language=EN, stringName=Xian-guang JIA, firstName=Xian-guang, middleName=null, lastName=JIA, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Traffic Engineering, Kunming University of Science and Technology, Kunming 650500, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1225467182545551421, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, authorId=1225467180930744335, language=CN, stringName=贾现广, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 昆明理工大学交通工程学院, 昆明 650500, bio={"content":"

贾现广(1977—),男,汉族,河南浚县人,硕士,副教授。研究方向:智能交通与大数据。E-mail:

"}, bioImg=null, bioContent=

贾现广(1977—),男,汉族,河南浚县人,硕士,副教授。研究方向:智能交通与大数据。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1225467180381291487, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, xref=1, ext=[AuthorCompanyExt(id=1225467180419040226, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180381291487, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Traffic Engineering, Kunming University of Science and Technology, Kunming 650500, China), AuthorCompanyExt(id=1225467180427428836, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180381291487, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 昆明理工大学交通工程学院, 昆明 650500)])]), Author(id=1225467182734295119, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1225467183011119223, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, authorId=1225467182734295119, language=EN, stringName=Huan LIU, firstName=Huan, middleName=null, lastName=LIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Traffic Engineering, Kunming University of Science and Technology, Kunming 650500, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1225467183187280018, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, authorId=1225467182734295119, language=CN, stringName=刘欢, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 昆明理工大学交通工程学院, 昆明 650500, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1225467180381291487, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, xref=1, ext=[AuthorCompanyExt(id=1225467180419040226, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180381291487, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Traffic Engineering, Kunming University of Science and Technology, Kunming 650500, China), AuthorCompanyExt(id=1225467180427428836, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180381291487, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 昆明理工大学交通工程学院, 昆明 650500)])]), Author(id=1225467183380218017, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, 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=1225467183564767408, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, authorId=1225467183380218017, language=EN, stringName=Chao-qin FENG, firstName=Chao-qin, middleName=null, lastName=FENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 School of Traffic Engineering, Kunming University of Science and Technology, Kunming 650500, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1225467183803842756, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, authorId=1225467183380218017, language=CN, stringName=冯超琴, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 昆明理工大学交通工程学院, 昆明 650500, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1225467180381291487, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, xref=1, ext=[AuthorCompanyExt(id=1225467180419040226, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180381291487, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Traffic Engineering, Kunming University of Science and Technology, Kunming 650500, China), AuthorCompanyExt(id=1225467180427428836, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180381291487, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 昆明理工大学交通工程学院, 昆明 650500)])]), Author(id=1225467184055501006, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=20070102@kust.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1225467184302964970, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, authorId=1225467184055501006, language=EN, stringName=Ying-ying LÜ, firstName=Ying-ying, middleName=null, lastName=LÜ, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, *, address=2 School of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1225467184432988409, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, authorId=1225467184055501006, language=CN, stringName=吕英英, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, *, address=2 昆明理工大学信息工程与自动化学院, 昆明 650500, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1225467180721030137, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, xref=2, ext=[AuthorCompanyExt(id=1225467180758778875, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180721030137, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China), AuthorCompanyExt(id=1225467180775556094, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180721030137, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 昆明理工大学信息工程与自动化学院, 昆明 650500)])])], keywords=[Keyword(id=1225467184856613154, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, orderNo=1, keyword=urban transportation), Keyword(id=1225467185066328370, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, orderNo=2, keyword=access flow prediction), Keyword(id=1225467185338958164, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, orderNo=3, keyword=Conv3D-GRU), Keyword(id=1225467185477370213, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, orderNo=4, keyword=bicycle sharing), Keyword(id=1225467187037651327, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, orderNo=5, keyword=spatio-temporal properties), Keyword(id=1225467187293503889, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, orderNo=1, keyword=城市交通), Keyword(id=1225467187486441891, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, orderNo=2, keyword=出入流预测), Keyword(id=1225467187620659633, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, orderNo=3, keyword=Conv3D-GRU), Keyword(id=1225467187792626115, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, orderNo=4, keyword=共享单车), Keyword(id=1225467187972981208, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, orderNo=5, keyword=时空特性)], refs=[Reference(id=1225467192582521596, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2021, volume=159, issue=null, pageStart=112631, pageEnd=null, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=Li C, Xiao W, Zhang D, journalName=Energy Policy, refType=null, unstructuredReference=Li C, Xiao W, Zhang D, et al. Low-carbon transformation of cities: understanding the demand for dockless bike sharing in China[J]. Energy Policy, 2021, 159: 112631., articleTitle=Low-carbon transformation of cities: understanding the demand for dockless bike sharing in China, refAbstract=null), Reference(id=1225467193048089367, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=2, pageStart=300, pageEnd=306, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=黄赶祥, 张薇, 徐迪, journalName=交通运输系统工程与信息, refType=null, unstructuredReference=黄赶祥, 张薇, 徐迪. 共享单车对我国主要城市交通拥堵的影响[J]. 交通运输系统工程与信息, 2023, 23(2): 300-306., articleTitle=共享单车对我国主要城市交通拥堵的影响, refAbstract=null), Reference(id=1225467193421382442, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2023, volume=23, issue=2, pageStart=300, pageEnd=306, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=Huang Ganxiang, Zhang Wei, Xu Di, journalName=Transportation Systems Engineering and Information, refType=null, unstructuredReference=Huang Ganxiang, Zhang Wei, Xu Di. The impact of shared bicycles on traffic congestion in major cities of China[J]. Transportation Systems Engineering and Information, 2023, 23(2): 300-306., articleTitle=The impact of shared bicycles on traffic congestion in major cities of China, refAbstract=null), Reference(id=1225467193593348917, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=2, pageStart=192, pageEnd=195, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=杨柳, journalName=信息技术与信息化, refType=null, unstructuredReference=杨柳. 深度学习在交通运输中应用综述[J]. 信息技术与信息化, 2022(2): 192-195., articleTitle=深度学习在交通运输中应用综述, refAbstract=null), Reference(id=1225467193782092606, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=2, pageStart=192, pageEnd=195, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=Yang Liu, journalName=Information Technology and Informatization, refType=null, unstructuredReference=Yang Liu. A review of deep learning applications in transportation[J]. Information Technology and Informatization, 2022(2): 192-195., articleTitle=A review of deep learning applications in transportation, refAbstract=null), Reference(id=1225467193912116039, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2018, volume=18, issue=32, pageStart=89, pageEnd=94, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=种颖珊, 韩晓明, journalName=科学技术与工程, refType=null, unstructuredReference=种颖珊, 韩晓明. 基于随机森林与时空聚类的共享单车站点需求量预测[J]. 科学技术与工程, 2018, 18(32): 89-94., articleTitle=基于随机森林与时空聚类的共享单车站点需求量预测, refAbstract=null), Reference(id=1225467194096665432, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2018, volume=18, issue=32, pageStart=89, pageEnd=94, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=Zhong Yingshan, Han Xiaoming, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Zhong Yingshan, Han Xiaoming. Prediction of shared bicycle station demand based on random forest and spatiotemporal clustering[J]. Science Technology and Engineering, 2018, 18(32): 89-94., articleTitle=Prediction of shared bicycle station demand based on random forest and spatiotemporal clustering, refAbstract=null), Reference(id=1225467194281214824, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=20, pageStart=8344, pageEnd=8349, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=曹旦旦, 范书瑞, 张艳, journalName=科学技术与工程, refType=null, unstructuredReference=曹旦旦, 范书瑞, 张艳, 等. 基于长短期记忆神经网络模型的共享单车短时需求量预测[J]. 科学技术与工程, 2020, 20(20): 8344-8349., articleTitle=基于长短期记忆神经网络模型的共享单车短时需求量预测, refAbstract=null), Reference(id=1225467194541261680, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2020, volume=20, issue=20, pageStart=8344, pageEnd=8349, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=Cao Dandan, Fan Shurui, Zhang Yan, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Cao Dandan, Fan Shurui, Zhang Yan, et al. Short-term demand forecasting for shared bicycles based on long short-term memory neural network model[J]. Science Technology and Engineering, 2020, 20(20): 8344-8349., articleTitle=Short-term demand forecasting for shared bicycles based on long short-term memory neural network model, refAbstract=null), Reference(id=1225467195950547839, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2020, volume=47, issue=12, pageStart=77, pageEnd=85, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=许淼, 刘宏飞, 初凯, journalName=湖南大学学报: 自然科学版, refType=null, unstructuredReference=许淼, 刘宏飞, 初凯. 基于AM-LSTM模型的共享单车时空需求预测[J]. 湖南大学学报: 自然科学版, 2020, 47(12): 77-85., articleTitle=基于AM-LSTM模型的共享单车时空需求预测, refAbstract=null), Reference(id=1225467196059599758, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2020, volume=47, issue=12, pageStart=77, pageEnd=85, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=Xu Miao, Liu Hongfei, Chu Kai, journalName=Journal of Hunan University: Natural Sciences, refType=null, unstructuredReference=Xu Miao, Liu Hongfei, Chu Kai. Spatiotemporal demand prediction of bike-sharing based on AM-LSTM model[J]. Journal of Hunan University: Natural Sciences, 2020, 47(12): 77-85., articleTitle=Spatiotemporal demand prediction of bike-sharing based on AM-LSTM model, refAbstract=null), Reference(id=1225467196244149150, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2021, volume=9, issue=null, pageStart=124337, pageEnd=124347, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=Collini E, Nesi P, Pantaleo G, journalName=IEEE Access, refType=null, unstructuredReference=Collini E, Nesi P, Pantaleo G. Deep learning for short-term prediction of available bikes on bike-sharing stations[J]. IEEE Access, 2021, 9: 124337-124347., articleTitle=Deep learning for short-term prediction of available bikes on bike-sharing stations, refAbstract=null), Reference(id=1225467196369978280, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=35, pageStart=15247, pageEnd=15254, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=刘恒孜, 贺玉龙, 宋太龙, journalName=科学技术与工程, refType=null, unstructuredReference=刘恒孜, 贺玉龙, 宋太龙, 等. 共享单车需求预测及调度优化[J]. 科学技术与工程, 2021, 21(35): 15247-15254., articleTitle=共享单车需求预测及调度优化, refAbstract=null), Reference(id=1225467196479030195, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2021, volume=21, issue=35, pageStart=15247, pageEnd=15254, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=Liu Hengzi, He Yulong, Song Tailong, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Liu Hengzi, He Yulong, Song Tailong, et al. Shared bicycle demand forecasting and dispatch optimization[J]. Science Technology and Engineering, 2021, 21(35): 15247-15254., articleTitle=Shared bicycle demand forecasting and dispatch optimization, refAbstract=null), Reference(id=1225467196659385285, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=9, pageStart=1871, pageEnd=1876, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=刘耿耿, 朱予涵, 郭灿阳, journalName=小型微型计算机系统, refType=null, unstructuredReference=刘耿耿, 朱予涵, 郭灿阳. 基于双向长短期记忆网络的共享单车流量预测[J]. 小型微型计算机系统, 2021, 42(9): 1871-1876., articleTitle=基于双向长短期记忆网络的共享单车流量预测, refAbstract=null), Reference(id=1225467196982346715, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=9, pageStart=1871, pageEnd=1876, url=null, language=null, rfNumber=[9], rfOrder=15, authorNames=Liu Genggeng, Zhu Yuhan, Guo Canyang, journalName=Microcomputer Systems, refType=null, unstructuredReference=Liu Genggeng, Zhu Yuhan, Guo Canyang. Shared bicycle flow prediction based on bidirectional long short-term memory network[J]. Microcomputer Systems, 2021, 42(9): 1871-1876., articleTitle=Shared bicycle flow prediction based on bidirectional long short-term memory network, refAbstract=null), Reference(id=1225467197171090407, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2018, volume=null, issue=36, pageStart=92, pageEnd=102, url=null, language=null, rfNumber=[10], rfOrder=16, authorNames=Yang H, Xie K, Ozbay K, journalName=Transportation Research Record, refType=null, unstructuredReference=Yang H, Xie K, Ozbay K, et al. Use of deep learning to predict daily usage of bike sharing systems[J]. Transportation Research Record, 2018(36): 92-102., articleTitle=Use of deep learning to predict daily usage of bike sharing systems, refAbstract=null), Reference(id=1225467197397582827, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2021, volume=561, issue=null, pageStart=274, pageEnd=285, url=null, language=null, rfNumber=[11], rfOrder=17, authorNames=Zi W, Xiong W, Chen H, journalName=Information Sciences, refType=null, unstructuredReference=Zi W, Xiong W, Chen H, et al. TAGCN: station-level demand prediction for bike-sharing system via a temporal attention graph convolution network[J]. Information Sciences, 2021, 561: 274-285., articleTitle=TAGCN: station-level demand prediction for bike-sharing system via a temporal attention graph convolution network, refAbstract=null), Reference(id=1225467197590520823, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=686, pageEnd=693, url=null, language=null, rfNumber=[12], rfOrder=18, authorNames=Guo R, Jiang Z, Huang J, journalName=2019 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computing, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI), refType=null, unstructuredReference=Guo R, Jiang Z, Huang J, et al. BikeNet: accurate bike demand prediction using graph neural networks for station rebalancing[C]// 2019 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computing, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI). New York: IEEE, 2019: 686-693., articleTitle=BikeNet: accurate bike demand prediction using graph neural networks for station rebalancing, refAbstract=null), Reference(id=1225467197837983756, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2021, volume=23, issue=8, pageStart=10923, pageEnd=10934, url=null, language=null, rfNumber=[13], rfOrder=19, authorNames=Li X, Xu Y, Chen Q, journalName=IEEE Transactions on Intelligent Transportation Systems, refType=null, unstructuredReference=Li X, Xu Y, Chen Q, et al. Short-term forecast of bicycle usage in bike sharing systems: a spatial-temporal memory network[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(8): 10923-10934., articleTitle=Short-term forecast of bicycle usage in bike sharing systems: a spatial-temporal memory network, refAbstract=null), Reference(id=1225467197988978712, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2024, volume=27, issue=2, pageStart=55, pageEnd=59, url=null, language=null, rfNumber=[14], rfOrder=20, authorNames=王俊, 于爱荣, journalName=软件工程, refType=null, unstructuredReference=王俊, 于爱荣. 基于ConvLSTM的南京地区共享单车需求预测研究[J]. 软件工程, 2024, 27(2): 55-59., articleTitle=基于ConvLSTM的南京地区共享单车需求预测研究, refAbstract=null), Reference(id=1225467198219665455, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2024, volume=27, issue=2, pageStart=55, pageEnd=59, url=null, language=null, rfNumber=[14], rfOrder=21, authorNames=Wang Jun, Yu Airong, journalName=Software Engineering, refType=null, unstructuredReference=Wang Jun, Yu Airong. Study on shared bicycle demand prediction in Nanjing based on ConvLSTM[J]. Software Engineering, 2024, 27(2): 55-59., articleTitle=Study on shared bicycle demand prediction in Nanjing based on ConvLSTM, refAbstract=null), Reference(id=1225467198383243321, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2019, volume=31, issue=null, pageStart=1665, pageEnd=1677, url=null, language=null, rfNumber=[15], rfOrder=22, authorNames=Ai Y, Li Z, Gan M, journalName=Neural Computing and Applications, refType=null, unstructuredReference=Ai Y, Li Z, Gan M, et al. A deep learning approach on short-term spatiotemporal distribution forecasting of dockless bike-sharing system[J]. Neural Computing and Applications, 2019, 31: 1665-1677., articleTitle=A deep learning approach on short-term spatiotemporal distribution forecasting of dockless bike-sharing system, refAbstract=null), Reference(id=1225467198613930065, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2021, volume=8, issue=1, pageStart=1, pageEnd=74, url=null, language=null, rfNumber=[16], rfOrder=23, authorNames=Alzubaidi L, Zhang J, Humaidi A J, journalName=Journal of Big Data, refType=null, unstructuredReference=Alzubaidi L, Zhang J, Humaidi A J, et al. Review of deep learning: concepts, CNN architectures, challenges, applications, future directions[J]. Journal of Big Data, 2021, 8(1): 1-74., articleTitle=Review of deep learning: concepts, CNN architectures, challenges, applications, future directions, refAbstract=null), Reference(id=1225467198773313631, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2014, volume=null, issue=null, pageStart=1409, pageEnd=null, url=null, language=null, rfNumber=[17], rfOrder=24, authorNames=Cho K, Van Merriënboer B, Bahdanau D, journalName=ArXiv Preprint ArXiv, refType=null, unstructuredReference=Cho K, Van Merriënboer B, Bahdanau D, et al. On the properties of neural machine translation: Encoder-decoder approaches[J]. ArXiv Preprint ArXiv, 2014: 1409.1259., articleTitle=On the properties of neural machine translation: Encoder-decoder approaches, refAbstract=null), Reference(id=1225467199020777584, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=25, authorNames=张金雷, 杨立兴, 高自友, journalName=深度学习与交通大数据实战, refType=null, unstructuredReference=张金雷, 杨立兴, 高自友. 深度学习与交通大数据实战[M]. 北京: 清华大学出版社, 2022., articleTitle=null, refAbstract=null), Reference(id=1225467200451035276, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=26, authorNames=Zhang Jinlei, Yang Lixing, Gao Ziyou, journalName=Practical applications of deep learning and traffic big data, refType=null, unstructuredReference=Zhang Jinlei, Yang Lixing, Gao Ziyou. Practical applications of deep learning and traffic big data[M]. Beijing: Tsinghua University Press, 2022., articleTitle=null, refAbstract=null), Reference(id=1225467200815939738, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2022, volume=24, issue=4, pageStart=4621, pageEnd=4630, url=null, language=null, rfNumber=[19], rfOrder=27, authorNames=Lee S H, Ku H C, journalName=IEEE Transactions on Intelligent Transportation Systems, refType=null, unstructuredReference=Lee S H, Ku H C. A dual attention-based recurrent neural network for short-term bike sharing usage demand prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 24(4): 4621-4630., articleTitle=A dual attention-based recurrent neural network for short-term bike sharing usage demand prediction, refAbstract=null), Reference(id=1225467200996294821, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=null, pageStart=857, pageEnd=862, url=null, language=null, rfNumber=[20], rfOrder=28, authorNames=Liang Y, Huang G, Zhao Z, journalName=null, refType=null, unstructuredReference=Liang Y, Huang G, Zhao Z. Bike sharing demand prediction based on knowledge sharing across modes:a graph-based deep learning approach[C]//2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC). Macau: IEEE, 2022: 857-862., articleTitle=Bike sharing demand prediction based on knowledge sharing across modes:a graph-based deep learning approach, refAbstract=null)], funds=[Fund(id=1225467192066622168, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, awardId=71961012, language=CN, fundingSource=国家自然科学基金(71961012), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1225467180381291487, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, xref=1, ext=[AuthorCompanyExt(id=1225467180419040226, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180381291487, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 School of Traffic Engineering, Kunming University of Science and Technology, Kunming 650500, China), AuthorCompanyExt(id=1225467180427428836, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180381291487, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 昆明理工大学交通工程学院, 昆明 650500)]), AuthorCompany(id=1225467180721030137, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, xref=2, ext=[AuthorCompanyExt(id=1225467180758778875, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180721030137, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 School of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China), AuthorCompanyExt(id=1225467180775556094, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, companyId=1225467180721030137, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 昆明理工大学信息工程与自动化学院, 昆明 650500)])], figs=[ArticleFig(id=1225467188392411645, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, label=Fig.1, caption=Conv3D-GRU model architecture diagram, figureFileSmall=yGgw3n5Y/3jsMvgyyMCkDw==, figureFileBig=wP3++dJczVHjRCaYfNi5bA==, tableContent=null), ArticleFig(id=1225467188522435082, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, label=图1, caption=Conv3D-GRU模型结构图, figureFileSmall=yGgw3n5Y/3jsMvgyyMCkDw==, figureFileBig=wP3++dJczVHjRCaYfNi5bA==, tableContent=null), ArticleFig(id=1225467188715373090, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, label=Fig.2, caption=Shared bicycle outflow prediction results, figureFileSmall=vWDGMQodCB3sBHt0DX27tg==, figureFileBig=kvOTStjBQ1BT5EMn6p7h1g==, tableContent=null), ArticleFig(id=1225467188849590833, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, label=图2, caption=共享单车出流预测结果, figureFileSmall=vWDGMQodCB3sBHt0DX27tg==, figureFileBig=kvOTStjBQ1BT5EMn6p7h1g==, tableContent=null), ArticleFig(id=1225467189017363007, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, label=Fig.3, caption=Shared bicycle inflow prediction results, figureFileSmall=3+380iZkxydXI9TZvo3vKQ==, figureFileBig=lS1ivyK2OuXbnz9JKHPMGw==, tableContent=null), ArticleFig(id=1225467189189329489, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, label=图3, caption=共享单车入流预测结果, figureFileSmall=3+380iZkxydXI9TZvo3vKQ==, figureFileBig=lS1ivyK2OuXbnz9JKHPMGw==, tableContent=null), ArticleFig(id=1225467189386461787, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, label=Fig.4, caption=Comparison chart of bike-sharing flow prediction, figureFileSmall=2NZdmredKMDe/fYwWaBobw==, figureFileBig=GvL0kIRMjJxuF4B2lP49zA==, tableContent=null), ArticleFig(id=1225467189617148528, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, label=图4, caption=共享单车流量预测对比图, figureFileSmall=2NZdmredKMDe/fYwWaBobw==, figureFileBig=GvL0kIRMjJxuF4B2lP49zA==, tableContent=null), ArticleFig(id=1225467189822669451, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, label=Table 1, caption=

Data example

, figureFileSmall=null, figureFileBig=null, tableContent=
started_at ended_at start_lat start_lng end_lat end_lng
2022/1/13 11:59 2022/1/13 12:02 42.012 800 -87.665 906 42.012 56 -87.674 4
2022/1/10 08:41 2022/1/10 08:46 42.012 763 -87.665 967 42.012 56 -87.674 4
2022/1/25 04:53 2022/1/25 04:58 41.925 602 -87.653 708 41.925 33 -87.665 8
2022/1/04 00:18 2022/1/04 00:33 41.983 593 -87.669 154 41.961 51 -87.671 4
2022/1/20 01:31 2022/1/20 01:37 41.877 850 -87.624 080 41.884 62 -87.627 8
), ArticleFig(id=1225467189961081501, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, label=表1, caption=

数据字段示例

, figureFileSmall=null, figureFileBig=null, tableContent=
started_at ended_at start_lat start_lng end_lat end_lng
2022/1/13 11:59 2022/1/13 12:02 42.012 800 -87.665 906 42.012 56 -87.674 4
2022/1/10 08:41 2022/1/10 08:46 42.012 763 -87.665 967 42.012 56 -87.674 4
2022/1/25 04:53 2022/1/25 04:58 41.925 602 -87.653 708 41.925 33 -87.665 8
2022/1/04 00:18 2022/1/04 00:33 41.983 593 -87.669 154 41.961 51 -87.671 4
2022/1/20 01:31 2022/1/20 01:37 41.877 850 -87.624 080 41.884 62 -87.627 8
), ArticleFig(id=1225467191315841705, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=EN, label=Table 2, caption=

Comparison of evaluation metrics among different models

, figureFileSmall=null, figureFileBig=null, tableContent=
预测模型 Conv3D-GRU 3D-CNN ConvLSTM
RMSE/10-3 3.989 4.123 4.530
MAE/10-3 0.932 0.980 1.080
R2/% 92.37 91.33 90.15
), ArticleFig(id=1225467191496196791, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1156983791462998263, language=CN, label=表2, caption=

各模型评价指标对比

, figureFileSmall=null, figureFileBig=null, tableContent=
预测模型 Conv3D-GRU 3D-CNN ConvLSTM
RMSE/10-3 3.989 4.123 4.530
MAE/10-3 0.932 0.980 1.080
R2/% 92.37 91.33 90.15
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2401851, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2401851, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2401851, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2401851, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于混合卷积-递归神经网络的共享单车出入流预测
收藏切换
PDF下载
贾现广 1 , 刘欢 1 , 冯超琴 1 , 吕英英 2, *
科学技术与工程 | 论文·交通运输 2025,25(5): 2127-2134
收起
收藏切换
科学技术与工程 | 论文·交通运输 2025, 25(5): 2127-2134
基于混合卷积-递归神经网络的共享单车出入流预测
全屏
贾现广1 , 刘欢1, 冯超琴1, 吕英英2, *
作者信息
  • 1 昆明理工大学交通工程学院, 昆明 650500
  • 2 昆明理工大学信息工程与自动化学院, 昆明 650500
  • 贾现广(1977—),男,汉族,河南浚县人,硕士,副教授。研究方向:智能交通与大数据。E-mail:

通讯作者:

*吕英英(1982—),女,汉族,山西临汾人,硕士,讲师。研究方向:智能交通与大数据。E-mail:
Prediction of Shared Bicycle Inflow and Outflow Based on Conv3D-GRU Neural Network
Xian-guang JIA1 , Huan LIU1, Chao-qin FENG1, Ying-ying LÜ2, *
Affiliations
  • 1 School of Traffic Engineering, Kunming University of Science and Technology, Kunming 650500, China
  • 2 School of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China
出版时间: 2025-02-18 doi: 10.12404/j.issn.1671-1815.2401851
文章导航
收藏切换

准确预测共享单车流量有助于优化共享单车的供需平衡,提高城市居民的出行便利性。为解决共享单车预测准确性不高以及时空特性捕捉不充分的问题,提出了一种混合卷积-递归神经网络(hybrid convolutional-recurrent neural network)Conv3D-GRU模型,采用芝加哥2022全年共享单车数据进行实验,并与三维卷积神经网络3D-CNN(3D convolutional neural network)模型和卷积长短期记忆网络(Convolutional long short-term memory,ConvLSTM)的预测结果进行比较,使用均方根误差(root mean squared error,RMSE)、平均绝对误差(mean absolute error,MAE)、决定系数R2评估模型性能。实验结果表明,Conv3D-GRU相较于3D-CNN和ConvLSTM模型,在RMSE、MAE以及R2上分别提高了3.25%、4.90%、1.14%和11.94%、13.70%、2.46%,可见Conv3D-GRU模型的预测误差小,预测精度高,能够有效和可靠地适用于共享单车出入流的预测。

城市交通  /  出入流预测  /  Conv3D-GRU  /  共享单车  /  时空特性

Accurately predicting bike-sharing flow is essential for optimizing the supply-demand balance of shared bikes and enhancing urban residents’ travel convenience. To address the issues of low prediction accuracy and insufficient capture of spatiotemporal characteristics in bike-sharing flow prediction, a hybrid convolutional-recurrent neural network (Conv3D-GRU) model was proposed. Using Chicago’s 2022 full-year bike-sharing data, experiments were conducted, and the results were compared with those of the 3D convolutional neural network (3D-CNN) model and the convolutional long short-term memory (ConvLSTM) model. The model performance was evaluated using root mean squared error (RMSE), mean absolute error (MAE), and the coefficient of determination (R2). Experimental results show that compared with the 3D-CNN and ConvLSTM models, Conv3D-GRU is improved by 3.25%, 4.90%, 1.14% and 11.94%, 13.70% and 2.46% on RMSE, MAE and R2, respectively. This demonstrates that the Conv3D-GRU model has lower prediction errors and higher prediction accuracy, making it an effective and reliable approach for forecasting bike-sharing inflow and outflow.

urban transportation  /  access flow prediction  /  Conv3D-GRU  /  bicycle sharing  /  spatio-temporal properties
贾现广, 刘欢, 冯超琴, 吕英英. 基于混合卷积-递归神经网络的共享单车出入流预测. 科学技术与工程, 2025 , 25 (5) : 2127 -2134 . DOI: 10.12404/j.issn.1671-1815.2401851
Xian-guang JIA, Huan LIU, Chao-qin FENG, Ying-ying LÜ. Prediction of Shared Bicycle Inflow and Outflow Based on Conv3D-GRU Neural Network[J]. Science Technology and Engineering, 2025 , 25 (5) : 2127 -2134 . DOI: 10.12404/j.issn.1671-1815.2401851
随着城市的快速发展,共享单车作为一种新型绿色交通出行方式为居民的日常出行提供了便捷[1]。在中国,共享单车自2016年兴起,迅速成为城市短途出行的重要方式。近年来,随着技术的进步和大数据分析的应用,共享单车的使用效率和用户体验不断提升。目前,中国共享单车市场呈现出智能化、精细化管理、绿色出行和多元化发展的趋势。有效预测和管理共享单车流量对于减少交通拥堵、减少能源消耗、提升城市环境质量具有至关重要的作用[2]
过去的研究主要侧重于传统方法如自回归移动平均模型(autoregressive integrated moving average,ARIMA)和支持向量机(support vector machine,SVM)[3]。然而,这些传统模型往往假设数据呈线性且独立分布,难以处理共享单车出入流中的非线性和时空依赖关系。近年来,机器学习在时间序列预测领域展现出优异的性能,尤其是在交通流量预测上取得了显著的成果[3],许多研究者开始将机器学习算法应用于共享单车需求预测。
种颖珊等[4]结合随机森林和时空聚类预测站点的共享单车需求。曹旦旦等[5]使用长短时记忆(long short-term memory,LSTM)神经网络来预测每小时的共享单车需求。许淼等[6]提出了一种引入注意力机制的长短时记忆神经网络模型(attention mechanism long short-term memory network,AM-LSTM),其预测精度优于单一的LSTM模型。Collini等[7]对共享单车系统中的可用共享单车进行短期预测,经过对比得出(bidirectional long short-term memory,Bi-LSTM)是最佳模型。刘恒孜等[8]提出了季节性灰色Markov模型来预测共享单车需求量。刘耿耿等[9]提出一种基于双向长短期记忆(BiLSTM)的深度网络模型以预测未来的共享单车流量,结果表明预测效果优于ANN,RNN和LSTM模型。Yang等[10]利用卷积神经网络(convolutional neural networks,CNN)来预测城市级和站点级的每日单车借还量。
尽管上述的研究提出了许多基于机器学习的预测方法,并在一定程度上解决了共享单车需求预测的问题,但这些方法并未完全同时考虑到共享单车的空间特性和时间依赖性。为了解决这个问题,Zi等[11]提出了一种新颖的具有时间注意力的深度图卷积网络模型(time attention graph convolutional network,TAGCN),用于每个站点的共享单车的出入站数量预测。Guo等[12]应用时空图神经网络(space-time graph neural network,ST-GNN)建立了自行车需求预测模型。上述学者都采用了图卷积网络(graph convolutional network,GCN)捕获空间相关性,但是GCN通常是静态的,它在考虑邻近节点的连接关系时不会考虑时间因素,难以捕获节点在不同时间点的状态变化。Li等[13]结合2D-CNN和LSTM提出了一种时空记忆网络(space-time memory network,STMN)来预测共享单车的短期使用情况。王俊等[14]和Ai等[15]都采用卷积长短期记忆网络(ConvLSTM)预测共享单车,并表明ConvLSTM模型比2D-CNN和LSTM模型具有更好的性能,但是ConvLSTM网络只在顶层的高级特征上建立时间联系,而没有充分利用低层特征上的空间和时间相关性。
共享单车的使用具有明显的时空特性,用户的出行需求也随着时间和地点的变化而变化,只考虑静态的空间特性或时间依赖性可能无法全面捕捉到共享单车的实际情况,从而导致预测不准确。因此,针对现有研究在捕捉共享单车的时空特性和综合处理多层次信息方面存在的不足,现提出一种基于Conv3D-GRU神经网络的共享单车出入流预测方法。Conv3D-GRU神经网络模型通过结合3D-CNN和GRU的优势,有效地连接高层次和低层次的空间特征和时间特征,使得模型更具适应性和泛化能力,能够更好地捕捉共享单车需求的时空变化规律,提高模型预测的准确性。
卷积神经网络(CNN)是一种深度学习算法,擅长处理具有网格结构的数据。根据处理数据的维度不同,CNN可分为一维卷积神经网络(1D-CNN)、二维卷积神经网络(2D-CNN)和三维卷积神经网络(3D-CNN)[16],其基本组成都包括卷积层、激活层、池化层和全连接层。与1D-CNN和2D-CNN相比,3D-CNN通过滑动窗口在数据的三维索引上移动,计算特征映射,从而识别出数据随时间和空间变化的规律性。3D-CNN计算公式为
f*g(x,y,z)=∑∑∑f(i,j,k)g(x-i,y-j,z-k)
式(1)中:f为输入图像;g为卷积核;(x,y,z)分别为卷积核在3个维度上的空间坐标;(i,j,k)为3D卷积核内的局部坐标;*表示卷积运算。
GRU是循环神经网络(recurrent neural network,RNN)的一种变体,能够解决传统RNN面临的梯度消失和长期依赖问题[17]。GRU通过其特有的门控机制,如重置门和更新门,使得模型在处理时间序列数据时能够更有效地保留长期依赖信息。
结合当前时刻输入信息xtht-1,GRU可以得到当前隐藏节点的输出yt和传递给下一个节点的隐藏状态ht,其内部计算公式如下。
zt=σ(Wz[ht-1,xt])
rt=σ(Wr[ht-1,xt])
${\stackrel{~}{h}}_{t}$=tanh(W[rt*ht-1,xt])
ht=(1-zt)*ht-1+zt*${\stackrel{~}{h}}_{t}$
式中:xt为当前时刻输入信息;ht-1ht分别为上一时刻和传递到下一时刻的隐藏状态; ${\stackrel{~}{h}}_{t}$为候选隐藏状态;rtzt分别为重置门和更新门;σ和tanh分别为Sigmoid函数和tanh函数;WzWrW分别为各阶段的权重参数。
全连接层(fully connected layer,FC)是神经网络中的一种基本层,也是最常用的层之一。在全连接层中,输入向量通过线性变换和激活函数的非线性映射,产生输出向量,其计算公式为
y=f(Wht+b)
式(6)中:ht为GRU层在时间步t的输出向量;W为全连接层的权重矩阵;b为全连接层的偏置向量;f为激活函数。
为有效捕捉共享单车数据中的空间位置特征及其随时间的变化规律,提出了一种混合卷积-递归神经网络(Conv3D-GRU)模型,所构建的Conv3D-GRU模型结构由5层3D-CNN、1层GRU和1层全连接层组成,该模型通过联合3D-CNN和GRU可学习到共享单车数据的空间特征及时间依赖关系,模型架构如图1所示。
(1)3D-CNN层。共享单车数据在城市不同区域展现出明显的使用差异,为了有效地处理这些空间信息,选择了3D-CNN。具体地,该模型将所处理的张量数据输入到2层3D-CNN用于识别和提取共享单车数量在不同时间和地点上的变化趋势、规律及特征。通过在日期、小时、经度和纬度索引上进行滑动的卷积操作,计算出一系列特征映射,捕获数据随时间和地点变化的规律性,从而学习到共享单车流量随着时间的推移在城市中不同地点的分布情况。最后引入了3层3D-CNN,进一步提取更高级的时空特征,层级化的特征提取过程能更深入地捕捉共享单车流量的时空复杂性,使得模型能够更准确地预测城市中的共享单车流量。
(2)GRU层。共享单车的使用模式不仅与空间因素相关,还与时间紧密相关。例如,通勤高峰期和节假日的单车使用模式可能有显著差异。为此,模型引入了GRU层捕捉时间序列数据中的动态变化和长期依赖关系。具体地,在空间特征通过3D-CNN层提取后,引入GRU识别出共享单车数据中不同时间段内的流量特征,使模型能够全面捕捉共享单车使用的时间依赖性。
(3)全连接层(FC)。全连接层主要用于将经过GRU层处理后的时间序列特征转换为适合进一步处理的高维特征向量,从而增强模型对共享单车流量变化细节的捕捉能力。
芝加哥是美国的主要大城市之一,具有代表性的城市结构和繁忙的交通系统,其共享单车系统广泛覆盖城市的不同区域,反映了从城市核心到郊区的多样化单车使用模式,其数据记录相对完整和准确。本文采用的数据集是2022年美国芝加哥全年共享单车行程数据,其中每条行程数据包含以下字段:订单ID、车辆类型、行程开始和结束的时间戳、起始和结束站点的名称和标识符、起始和结束站点的经纬度坐标,以及用户类型(是否为注册会员)。由于本文只研究共享单车的出入流量,故对数据进行筛选,去除无关的字段,只保留订单数据的起始时间(‘started_at’),结束时间(‘ended_at’),起始经度(‘start_lng’)和纬度(‘start_lat’),以及结束经度(‘end_lng’)和纬度(‘end_lat’)。数据字段示例如表1所示。
为了将原始的共享单车数据转化为模型能够理解和处理的形式,同时剔除无关信息、挖掘空间和时间的关联性,并降低计算负担,对原始共享单车订单数据进行以下处理。
(1)网格划分。基于经纬度,将城市分成32×32个网格,网格的范围设置为经度-87.83~-87.50,纬度为41.65~42.06,每个网格为一个子区域且大小相等,用Q={q1,1,…,qi,j,…,qm,n}来表示这些子区域的集合,其中,qi,j为第i行,第j列的子区域。使用地理位置与网格的起点的差值除以网格间距得到每个地理位置在网格系统中的索引。网格索引计算公式为
lng_ind =floor[(lng - lng_start) / lng_interval]
lat_ind = floor[(lat - lat_start) / lat_interval]
式中:lat和lng分别为共享单车的经度和纬度;lng_start和lat_start分别为网格的起始经度和纬度;lat_interval和lng_interval分别为经度和纬度的网格间距;floor表示向下取整。
(2)网格出入流张量计算。提取数据集中的每个共享单车行程的起始和结束时间,将原始格式(如“2022-01-01 09:30”)分解为日期和时间两部分,提取出日期和小时。根据日期、小时和网格索引,统计每个网格在每个小时的共享单车入流和出流。共享单车的出入流[18]是指在每个小时内,每个网格区域共享单车的离开和进入数量。设P为共享单车轨迹的集合,给定一个子区域qi,j,其相应的入流与出流公式定义为
$\boldsymbol{x}_{\mathrm{in}, i, j}^{t}=\sum_{E_{r} \in P}|l>1| h_{l-1} \notin q_{i, j} \wedge h_{l} \in q_{i, j} \mid$
$\boldsymbol{x}_{\mathrm{out}, i, j}^{t}=\sum_{E_{r} \in P}|l>1| h_{l} \in q_{i, j} \wedge h_{l+1} \notin q_{i, j} \mid$
式中:Er:h1h2→…→hErP中在时间间隔t的一段子轨迹;hlqi,j表示hl位于区域qi,j中;|·|代表集合的势。将在时间间隔t的入流和出流定义为出入流张量(Tensor)χt∈Rm×n×2ti,j,0=${x}_{in,i,j}^{t}$,${\chi }_{i,j,1}^{t}$=${x}_{out,i,j}^{t}$
最后,将以上统计数据存储为一个五维数组,其中的维度分别是日期、小时、网格的经度索引、网格的纬度索引和流动性(0代表入流,1代表出流),构建一个描述城市共享单车流动性的空间时间数据集,为进一步的数据分析和建模提供了基础。
采用Python编程语言进行编程,实验环境配置如下:处理器为Intel(R) Core(TM) i5-8250U CPU @1.60GHz(最大可加速至1.80 GHz),内存容量为16 GB,操作系统为Windows 10,模型的构建基于PyTorch深度学习框架。模型使用均方误差损失(MSE)作为训练的损失函数,并采用Adam优化器来更新模型的权重。学习率设置为0.001,批次大小为32,epoch设置为100,设置提前终止策略(EarlyStopping)防止所构建的模型出现过拟合。将共享单车出入流数据进行归一化处理,按照8∶1∶1划分训练集、验证集和测试集,使用3个时间步长预测未来1个时间步长的共享单车出入流。基于Conv3D-GRU神经网络的共享单车出入流预测模型的各层参数设置如下。
(1)模型首先采用2个三维卷积层进行空间特征提取。第1个卷积层从2个通道的输入映射到16个通道的输出,其中卷积核的大小为3,步长为1,且进行了1单位的零填充。第2个卷积层从16个通道的输入映射到32个通道的输出,参数设置与第1层相同。每个卷积层后采用ReLU激活函数,以引入非线性并增强模型的表达能力。
(2)空间特征提取后,采用1个GRU层处理时间序列数据以提取时间层次的特征。该层的输入尺寸为32×32×32,隐藏层单元数设置为64。
(3)模型通过1个全连接层将GRU的输出从64维映射到64×32×32维,调整数据的维度,使其适应后续的卷积层。
(4)最后通过3个三维卷积层将时序特征映射回原始的空间分辨率。其中,前2个卷积层将64个通道的输入映射到64个通道的输出和32个通道的输出,其中卷积核的大小为3,步长为1,且进行了1单位的零填充。最后1个卷积层则将32个通道的输入映射到2个通道的输出,卷积核的大小设为(1,3,3),步长为1,且进行了(0,1,1)单位的零填充。
为了验证所构建的Conv3D-GRU模型的有效性,将Conv3D-GRU模型和3D-CNN模型以及ConvLSTM模型进行平行对照实验。某小时共享单车出入流的真实值和各模型的预测值结果热力图如图2图3所示,Conv3D-GRU模型的预测效果比3D-CNN和ConvLSTM模型更好。
图2可知,3个预测模型的出流预测结果均与真实值存在一定程度的差异,但是从整体上相比而言,Conv3D-GRU模型的预测结果与真实值最为接近,其次是3D-CNN模型,ConvLSTM模型的预测效果最差。进一步分析每个网格的预测结果,Conv3D-GRU模型对每个网格的出流预测比3D-CNN和ConvLSTM模型更准确。
图3可知,整体上,Conv3D-GRU模型预测的入流结果比3D-CNN和ConvLSTM模型更接近真实值。局部上,3D-CNN模型对每个网格的入流预测比ConvLSTM模型好,但是3D-CNN模型对部分网格的预测与真实值有较大差异,相比之下,Conv3D-GRU模型对每个网格的入流预测精度更高。
为了更全面地评估模型在时间序列预测上的性能,绘制某网格一段时间内的预测流量与实际流量的折线图,如图4所示。可以观察到,3个模型都捕捉到了流量的总体趋势,但在预测准确性上存在差异。3DCNN-GRU模型在大多数时间点上比其他模型更接近实际流量数据。这表明,在处理共享单车流量预测问题时,结合了卷积神经网络和循环神经网络的3DCNN-GRU模型能提供更高的预测准确性。
为了更加精确地描述模型的各种性能,使用均方根误差RMSE、平均绝对误差MAE及决定系数R2这3种指标来评价3种模型的性能,RMSE、MAE[19]R2[20]定义如下。
RMSE=$\sqrt{\frac{1}{n}\sum _{i}^{n}({y}_{i}-{\stackrel{\wedge }{y}}_{i})}$
MAE=$\frac{1}{n}\sum _{i=1}^{n}\left|{\stackrel{\wedge }{y}}_{{}_{i}}-{y}_{i}\right|$
R2=1-SSE/SST
式中:${\stackrel{\wedge }{y}}_{i}$为共享单车流量的预测值;yi为共享单车流量的真实值;SSE为误差平方和;SST为总平方和。
结果显示,Conv3D-GRU模型在预测城市共享单车流量方面表现出了良好的性能。具体的结果如表2所示。
RMSE越小表示模型预测值与实际值误差小,模型预测精度高。Conv3D-GRU模型相对于3D-CNN模型提升了约3.25%,相对于ConvLSTM模型提升了约11.94%。RMSE的值越小,模型的预测性能越好。说明Conv3D-GRU模型的预测值与真实观测值之间差异更小,模型的预测精度更高。
MAE衡量模型对每个数据点的误差的平均值。Conv3D-GRU模型相对于3D-CNN模型提升了约4.90%,相对于ConvLSTM模型提升了约13.70%。这说明Conv3D-GRU模型的平均预测误差更小,即Conv3D-GRU模型对每个数据点的预测更为准确。
R2度量模型解释因变量方差的比例,较高的R2表示模型能够更好地解释因变量(目标变量)的方差,即更好地捕捉数据的变化趋势。Conv3D-GRU模型相对于3D-CNN模型提升了约1.14%,相对于ConvLSTM模型提升了约2.46%。
这表明Conv3D-GRU模型更好地适应了数据的变化趋势,使得其对数据的解释能力更强。
Conv3D-GRU模型在RMSE、MAE、R2指标上均表现出比3D-CNN和ConvLSTM模型更高的性能。与3D-CNN和ConvLSTM模型相比,Conv3D-GRU模型的预测精度更高,预测值与真实值的误差更小,拟合更准确,对数据变化趋势的解释能力更强,说明Conv3D-GRU模型在预测共享单车出入流上更具有优势。
考虑到共享单车流量存在的时空特性,提出了一种基于混合卷积-递归神经网络(Conv3D-GRU)的模型,融合了三维卷积神经网络(3D-CNN)和门控循环单元(GRU)的特性。通过引入三维卷积神经网络,模型从多个时间步长和空间位置中捕获特征,从而更全面地捕捉不同时间和空间维度的变化。GRU在处理序列数据时,能够更好地捕捉和理解数据中的时序依赖性,进一步提升了模型的预测性能。结果表明,Conv3D-GRU模型相比于3D-CNN模型,在RMSE、MAE、R2三项指标上分别提升了3.25%、4.90%、1.14%;Conv3D-GRU模型相比于ConvLSTM模型,在RMSE、MAE、R2三项指标上分别提升了11.94%、13.70%、2.46%。Conv3D-GRU模型的预测结果更接近真实情况,对每个数据点的预测更为精确,能更好捕捉数据的变化趋势,模型拟合程度更高,可解释性更强,说明所提出的Conv3D-GRU模型在共享单车流量预测方面和基准模型相比误差低、精度更高,具有应用于共享单车流量预测的潜力,但仍需进一步的深入研究和优化,以充分发挥其优越性能并应对局限性。
  • 国家自然科学基金(71961012)
参考文献 引证文献
排序方式:
[1]
Li C, Xiao W, Zhang D, et al. Low-carbon transformation of cities: understanding the demand for dockless bike sharing in China[J]. Energy Policy, 2021, 159: 112631.
[2]
黄赶祥, 张薇, 徐迪. 共享单车对我国主要城市交通拥堵的影响[J]. 交通运输系统工程与信息, 2023, 23(2): 300-306.
Huang Ganxiang, Zhang Wei, Xu Di. The impact of shared bicycles on traffic congestion in major cities of China[J]. Transportation Systems Engineering and Information, 2023, 23(2): 300-306.
[3]
杨柳. 深度学习在交通运输中应用综述[J]. 信息技术与信息化, 2022(2): 192-195.
Yang Liu. A review of deep learning applications in transportation[J]. Information Technology and Informatization, 2022(2): 192-195.
[4]
种颖珊, 韩晓明. 基于随机森林与时空聚类的共享单车站点需求量预测[J]. 科学技术与工程, 2018, 18(32): 89-94.
Zhong Yingshan, Han Xiaoming. Prediction of shared bicycle station demand based on random forest and spatiotemporal clustering[J]. Science Technology and Engineering, 2018, 18(32): 89-94.
[5]
曹旦旦, 范书瑞, 张艳, 等. 基于长短期记忆神经网络模型的共享单车短时需求量预测[J]. 科学技术与工程, 2020, 20(20): 8344-8349.
Cao Dandan, Fan Shurui, Zhang Yan, et al. Short-term demand forecasting for shared bicycles based on long short-term memory neural network model[J]. Science Technology and Engineering, 2020, 20(20): 8344-8349.
[6]
许淼, 刘宏飞, 初凯. 基于AM-LSTM模型的共享单车时空需求预测[J]. 湖南大学学报: 自然科学版, 2020, 47(12): 77-85.
Xu Miao, Liu Hongfei, Chu Kai. Spatiotemporal demand prediction of bike-sharing based on AM-LSTM model[J]. Journal of Hunan University: Natural Sciences, 2020, 47(12): 77-85.
[7]
Collini E, Nesi P, Pantaleo G. Deep learning for short-term prediction of available bikes on bike-sharing stations[J]. IEEE Access, 2021, 9: 124337-124347.
[8]
刘恒孜, 贺玉龙, 宋太龙, 等. 共享单车需求预测及调度优化[J]. 科学技术与工程, 2021, 21(35): 15247-15254.
Liu Hengzi, He Yulong, Song Tailong, et al. Shared bicycle demand forecasting and dispatch optimization[J]. Science Technology and Engineering, 2021, 21(35): 15247-15254.
[9]
刘耿耿, 朱予涵, 郭灿阳. 基于双向长短期记忆网络的共享单车流量预测[J]. 小型微型计算机系统, 2021, 42(9): 1871-1876.
Liu Genggeng, Zhu Yuhan, Guo Canyang. Shared bicycle flow prediction based on bidirectional long short-term memory network[J]. Microcomputer Systems, 2021, 42(9): 1871-1876.
[10]
Yang H, Xie K, Ozbay K, et al. Use of deep learning to predict daily usage of bike sharing systems[J]. Transportation Research Record, 2018(36): 92-102.
[11]
Zi W, Xiong W, Chen H, et al. TAGCN: station-level demand prediction for bike-sharing system via a temporal attention graph convolution network[J]. Information Sciences, 2021, 561: 274-285.
[12]
Guo R, Jiang Z, Huang J, et al. BikeNet: accurate bike demand prediction using graph neural networks for station rebalancing[C]// 2019 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computing, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI). New York: IEEE, 2019: 686-693.
[13]
Li X, Xu Y, Chen Q, et al. Short-term forecast of bicycle usage in bike sharing systems: a spatial-temporal memory network[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(8): 10923-10934.
[14]
王俊, 于爱荣. 基于ConvLSTM的南京地区共享单车需求预测研究[J]. 软件工程, 2024, 27(2): 55-59.
Wang Jun, Yu Airong. Study on shared bicycle demand prediction in Nanjing based on ConvLSTM[J]. Software Engineering, 2024, 27(2): 55-59.
[15]
Ai Y, Li Z, Gan M, et al. A deep learning approach on short-term spatiotemporal distribution forecasting of dockless bike-sharing system[J]. Neural Computing and Applications, 2019, 31: 1665-1677.
[16]
Alzubaidi L, Zhang J, Humaidi A J, et al. Review of deep learning: concepts, CNN architectures, challenges, applications, future directions[J]. Journal of Big Data, 2021, 8(1): 1-74.
[17]
Cho K, Van Merriënboer B, Bahdanau D, et al. On the properties of neural machine translation: Encoder-decoder approaches[J]. ArXiv Preprint ArXiv, 2014: 1409.1259.
[18]
张金雷, 杨立兴, 高自友. 深度学习与交通大数据实战[M]. 北京: 清华大学出版社, 2022.
Zhang Jinlei, Yang Lixing, Gao Ziyou. Practical applications of deep learning and traffic big data[M]. Beijing: Tsinghua University Press, 2022.
[19]
Lee S H, Ku H C. A dual attention-based recurrent neural network for short-term bike sharing usage demand prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 24(4): 4621-4630.
[20]
Liang Y, Huang G, Zhao Z. Bike sharing demand prediction based on knowledge sharing across modes:a graph-based deep learning approach[C]//2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC). Macau: IEEE, 2022: 857-862.
2025年第25卷第5期
PDF下载
365
148
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2401851
  • 接收时间:2024-03-15
  • 首发时间:2025-07-29
  • 出版时间:2025-02-18
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-03-15
  • 修回日期:2024-11-19
基金
国家自然科学基金(71961012)
作者信息
    1 昆明理工大学交通工程学院, 昆明 650500
    2 昆明理工大学信息工程与自动化学院, 昆明 650500

通讯作者:

*吕英英(1982—),女,汉族,山西临汾人,硕士,讲师。研究方向:智能交通与大数据。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2401851
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