Article(id=1207271182483669821, tenantId=1146029695717560320, journalId=1205116964453384197, issueId=1207271180105499439, articleNumber=null, orderNo=null, doi=10.20040/j.cnki.1000-7709.2025.20242243, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1730044800000, receivedDateStr=2024-10-28, revisedDate=1734278400000, revisedDateStr=2024-12-16, acceptedDate=null, acceptedDateStr=null, onlineDate=1765765479918, onlineDateStr=2025-12-15, pubDate=1758729600000, pubDateStr=2025-09-25, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1765765479918, onlineIssueDateStr=2025-12-15, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1765765479918, creator=13701087609, updateTime=1765765479918, updator=13701087609, issue=Issue{id=1207271180105499439, tenantId=1146029695717560320, journalId=1205116964453384197, year='2025', volume='43', issue='9', pageStart='1', pageEnd='220', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=null, createTime=1765765479351, creator=13701087609, updateTime=1765765681303, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1207272027254247478, tenantId=1146029695717560320, journalId=1205116964453384197, issueId=1207271180105499439, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1207272027254247479, tenantId=1146029695717560320, journalId=1205116964453384197, issueId=1207271180105499439, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=101, endPage=105, ext={EN=ArticleExt(id=1207271184002007900, articleId=1207271182483669821, tenantId=1146029695717560320, journalId=1205116964453384197, language=EN, title=Prediction Model of Shield Tunneling Speed Based on VMD-DBO-Stacking Ensemble Learning, columnId=null, journalTitle=Water Resources and Power, columnName=null, runingTitle=null, highlight=null, articleAbstract=

Addressing the issues of single model algorithm, low accuracy, and poor generalization in existing shield tunneling speed prediction methods, this study proposes a shield tunneling speed prediction approach to improve prediction accuracy based on Variational Mode Decomposition (VMD), Dung Beetle Optimizer (DBO), and Stacking ensemble learning. Firstly, to obtain more effective data, VMD is applied to decompose and reconstruct the original data to obtain denoised construction parameter data for subsequent model prediction. Secondly, based on the ensemble learning strategy, Support Vector Regression (SVR), Random Forest (RF), and Extreme Gradient Boosting (XGBoost) models are selected as base learners, while Gaussian Process Regression (GPR) is chosen as the meta-learner to construct a Stacking ensemble learning prediction model with higher prediction accuracy and stronger generalization ability. Thirdly, to further enhance prediction accuracy, DBO is employed to optimize the hyperparameters of the ensemble learning model. Finally, this prediction method is applied to the shield tunneling construction of a water diversion tunnel project in Henan Province and compared with other prediction methods. Compared to other single models (SVR, RF, XGBoost), the results indicate that the proposed method achieves higher prediction accuracy, with average accuracy improvements of 7.76%, 6.70%, and 4.97%, respectively, providing a new approach for shield tunneling speed prediction.

, 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, authorCompany=null, fund=null, authors=null, authorsList=Zi-ang DENG, Yu-xian ZHANG, Ji-xun ZHANG), CN=ArticleExt(id=1207271188804485206, articleId=1207271182483669821, tenantId=1146029695717560320, journalId=1205116964453384197, language=CN, title=基于VMD-DBO-Stacking集成学习的盾构掘进速度预测模型, columnId=1207271182139736888, journalTitle=水电能源科学, columnName=水工结构、水工材料与水利工程施工, runingTitle=null, highlight=null, articleAbstract=

针对现有盾构掘进速度预测方法存在的模型算法单一、精度不高和泛化性较差等问题,为了提高盾构掘进速度预测精度,建立一种基于变分模态分解(VMD)、蜣螂优化算法(DBO)和Stacking(VMD-DBO-Stacking)集成学习的盾构掘进速度预测模型。首先,为了得到更有效的数据,采用VMD对原始数据进行分解重构得到去噪后的施工参数数据用于后续模型预测;其次,基于集成学习策略,选取支持向量回归(SVR)模型、随机森林(RF)模型、极端梯度提升(XGBoost)模型作为基学习器,高斯过程回归(GPR)模型作为元学习器,从而构建预测精度更高、泛化性更强的Stacking集成学习预测模型;然后,为了进一步提高预测精度,采用DBO对集成学习模型进行超参数优化;最后,将此预测模型用于河南某引水隧洞工程盾构施工中并与其他预测模型进行对比。结果表明,与其他单一模型(SVR、RF、XGBoost)相比,所建模型具有更高的预测精度,平均精度分别提升7.76%、6.70%、4.97%,为盾构掘进速度预测提供一种新思路。

, correspAuthors=null, authorNote=null, correspAuthorsNote=
张继勋(1974-),男,博士、副教授、硕导,研究方向为水工地下工程信息化,E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=OmC3O2s53AinGX3NmHoIFA==, magXml=EzAcm85AeQgevJv8ca0nLQ==, pdfUrl=null, pdf=/t0j6GodLiD290F/oNviGw==, pdfFileSize=1503229, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=3T4HYYetPxLrxm8O0aumSw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=SQ2VyCpdYWKibJbIyS4g+Q==, mapNumber=null, authorCompany=null, fund=null, authors=

邓子昂(2001-),男,硕士研究生,研究方向为水工地下工程信息化,E-mail:

, authorsList=邓子昂, 张玉贤, 张继勋)}, authors=[Author(id=1207271189274247297, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=dengziang1917@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1207271189429436561, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, authorId=1207271189274247297, language=EN, stringName=Zi-ang DENG, firstName=Zi-ang, middleName=null, lastName=DENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1207271189563654301, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, authorId=1207271189274247297, language=CN, stringName=邓子昂, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=河海大学水利水电学院,江苏 南京 210098, bio={"content":"

邓子昂(2001-),男,硕士研究生,研究方向为水工地下工程信息化,E-mail:

"}, bioImg=null, bioContent=

邓子昂(2001-),男,硕士研究生,研究方向为水工地下工程信息化,E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1207271189131640950, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, xref=null, ext=[AuthorCompanyExt(id=1207271189140029559, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, companyId=1207271189131640950, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China), AuthorCompanyExt(id=1207271189169389689, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, companyId=1207271189131640950, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=河海大学水利水电学院,江苏 南京 210098)])]), Author(id=1207271189664317606, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, 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=1207271189836284081, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, authorId=1207271189664317606, language=EN, stringName=Yu-xian ZHANG, firstName=Yu-xian, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1207271189957918905, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, authorId=1207271189664317606, language=CN, stringName=张玉贤, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=河海大学水利水电学院,江苏 南京 210098, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1207271189131640950, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, xref=null, ext=[AuthorCompanyExt(id=1207271189140029559, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, companyId=1207271189131640950, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China), AuthorCompanyExt(id=1207271189169389689, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, companyId=1207271189131640950, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=河海大学水利水电学院,江苏 南京 210098)])]), Author(id=1207271190071165125, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=zhangjixun@hhu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1207271190188605648, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, authorId=1207271190071165125, language=EN, stringName=Ji-xun ZHANG, firstName=Ji-xun, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1207271190293463259, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, authorId=1207271190071165125, language=CN, stringName=张继勋, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=河海大学水利水电学院,江苏 南京 210098, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1207271189131640950, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, xref=null, ext=[AuthorCompanyExt(id=1207271189140029559, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, companyId=1207271189131640950, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China), AuthorCompanyExt(id=1207271189169389689, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, companyId=1207271189131640950, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=河海大学水利水电学院,江苏 南京 210098)])])], keywords=[Keyword(id=1207271190582870254, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=EN, orderNo=1, keyword=shield tunneling machine), Keyword(id=1207271190876471548, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=EN, orderNo=2, keyword=tunneling speed), Keyword(id=1207271191061020934, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=EN, orderNo=3, keyword=variational mode decomposition), Keyword(id=1207271192361255199, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=EN, orderNo=4, keyword=Dung Beetle Optimizer), Keyword(id=1207271192449335593, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=EN, orderNo=5, keyword=Stacking ensemble learning), Keyword(id=1207271192570970417, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=CN, orderNo=1, keyword=盾构), Keyword(id=1207271192700993856, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=CN, orderNo=2, keyword=掘进速度), Keyword(id=1207271192789074250, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=CN, orderNo=3, keyword=变分模态分解), Keyword(id=1207271192889737556, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=CN, orderNo=4, keyword=蜣螂优化算法), Keyword(id=1207271193019760995, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=CN, orderNo=5, keyword=Stacking集成学习)], refs=[Reference(id=1207271194793951716, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2023, volume=45, issue=4, pageStart=121, pageEnd=125, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=李强, 李子阳, 许林军, journalName=人民黄河, refType=null, unstructuredReference=李强, 李子阳, 许林军, 等. 引水隧洞盾构法施工风险源辨识与分析[J]. 人民黄河, 2023, 45(4): 121-125., articleTitle=引水隧洞盾构法施工风险源辨识与分析, refAbstract=null), Reference(id=1207271194898809323, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2024, volume=45, issue=1, pageStart=107, pageEnd=113, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=裴浩东, 叶社保, 杨平, journalName=郑州大学学报(工学版), refType=null, unstructuredReference=裴浩东, 叶社保, 杨平, 等. 软土地层盾构掘进参数分析及掘进速度预测[J]. 郑州大学学报(工学版), 2024, 45(1): 107-113., articleTitle=软土地层盾构掘进参数分析及掘进速度预测, refAbstract=null), Reference(id=1207271195028832758, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2021, volume=49, issue=4, pageStart=373, pageEnd=379, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=张哲铭, 李晓瑜, 姬建, journalName=河海大学学报(自然科学版), refType=null, unstructuredReference=张哲铭, 李晓瑜, 姬建. 基于LS-SVM的TBM掘进参数预测模型[J]. 河海大学学报(自然科学版), 2021, 49(4): 373-379., articleTitle=基于LS-SVM的TBM掘进参数预测模型, refAbstract=null), Reference(id=1207271195175633406, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2021, volume=29, issue=5, pageStart=1186, pageEnd=1200, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=仉文岗, 唐理斌, 陈福勇, journalName=应用基础与工程科学学报, refType=null, unstructuredReference=仉文岗, 唐理斌, 陈福勇, 等. 基于4种超参数优化算法及随机森林模型预测TBM掘进速度[J]. 应用基础与工程科学学报, 2021, 29(5): 1186-1200., articleTitle=基于4种超参数优化算法及随机森林模型预测TBM掘进速度, refAbstract=null), Reference(id=1207271195263713801, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2020, volume=39, issue=5, pageStart=139, pageEnd=145, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=赵光祖, 王亚旭, 李尧, journalName=河南理工大学学报(自然科学版), refType=null, unstructuredReference=赵光祖, 王亚旭, 李尧, 等. 基于优化BP神经网络的TBM性能预测[J]. 河南理工大学学报(自然科学版), 2020, 39(5): 139-145., articleTitle=基于优化BP神经网络的TBM性能预测, refAbstract=null), Reference(id=1207271195364377105, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2024, volume=57, issue=2, pageStart=174, pageEnd=185, url=null, language=null, rfNumber=[6], rfOrder=5, authorNames=祝玉珊, 王晓玲, 崔博, journalName=天津大学学报(自然科学与工程技术版), refType=null, unstructuredReference=祝玉珊, 王晓玲, 崔博, 等. 坝基灌浆量预测ISSA-Stacking集成学习代理模型研究[J]. 天津大学学报(自然科学与工程技术版), 2024, 57(2): 174-185., articleTitle=坝基灌浆量预测ISSA-Stacking集成学习代理模型研究, refAbstract=null), Reference(id=1207271195527954968, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2023, volume=44, issue=1, pageStart=55, pageEnd=73, url=null, language=null, rfNumber=[7], rfOrder=6, authorNames=江星星, 宋秋昱, 杜贵府, journalName=仪器仪表学报, refType=null, unstructuredReference=江星星, 宋秋昱, 杜贵府, 等. 变分模式分解方法研究与应用综述[J]. 仪器仪表学报, 2023, 44(1): 55-73., articleTitle=变分模式分解方法研究与应用综述, refAbstract=null), Reference(id=1207271195653784101, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2024, volume=60, issue=24, pageStart=97, pageEnd=109, url=null, language=null, rfNumber=[8], rfOrder=7, authorNames=娄革伟, 郑永煌, 陈均, journalName=计算机工程与应用, refType=null, unstructuredReference=娄革伟, 郑永煌, 陈均, 等. 混合多策略改进的蜣螂优化算法[J]. 计算机工程与应用, 2024, 60(24): 97-109., articleTitle=混合多策略改进的蜣螂优化算法, refAbstract=null), Reference(id=1207271196878520881, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2023, volume=42, issue=5, pageStart=634, pageEnd=640, url=null, language=null, rfNumber=[9], rfOrder=8, authorNames=彭继慎, 郝茗, 宋立业, journalName=辽宁工程技术大学学报(自然科学版), refType=null, unstructuredReference=彭继慎, 郝茗, 宋立业, 等. 基于TSSA-SVR算法的TBM掘进速度预测[J]. 辽宁工程技术大学学报(自然科学版), 2023, 42(5): 634-640., articleTitle=基于TSSA-SVR算法的TBM掘进速度预测, refAbstract=null), Reference(id=1207271196970795577, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2022, volume=59, issue=6, pageStart=14, pageEnd=23, url=null, language=null, rfNumber=[10], rfOrder=9, authorNames=杜庆峰, 张双俐, 张晨曦, journalName=现代隧道技术, refType=null, unstructuredReference=杜庆峰, 张双俐, 张晨曦, 等. 基于均值滤波去噪和XGBoost算法的泥水平衡盾构掘进速度预测方法[J]. 现代隧道技术, 2022, 59(6): 14-23., articleTitle=基于均值滤波去噪和XGBoost算法的泥水平衡盾构掘进速度预测方法, refAbstract=null), Reference(id=1207271197058875965, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2024, volume=42, issue=3, pageStart=133, pageEnd=136, 53, url=null, language=null, rfNumber=[11], rfOrder=10, authorNames=杨辉斌, 郑德仁, 王贺龙, journalName=水电能源科学, refType=null, unstructuredReference=杨辉斌, 郑德仁, 王贺龙, 等. 基于GA-SVR的管网异常漏损检测[J]. 水电能源科学, 2024, 42(3): 133-136, 53., articleTitle=基于GA-SVR的管网异常漏损检测, refAbstract=null), Reference(id=1207271197180510788, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2024, volume=42, issue=11, pageStart=60, pageEnd=63, url=null, language=null, rfNumber=[12], rfOrder=11, authorNames=柏晓鹏, 南瑞川, 杜甫, journalName=水电能源科学, refType=null, unstructuredReference=柏晓鹏, 南瑞川, 杜甫, 等. 基于随机森林的地下水溶质运移替代模型研究[J]. 水电能源科学, 2024, 42(11): 60-63., articleTitle=基于随机森林的地下水溶质运移替代模型研究, refAbstract=null), Reference(id=1207271197344088656, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2023, volume=41, issue=11, pageStart=77, pageEnd=81, url=null, language=null, rfNumber=[13], rfOrder=12, authorNames=李晨阳, 郑东健, journalName=水电能源科学, refType=null, unstructuredReference=李晨阳, 郑东健. 基于多层次数据处理的NGO-XGBoost大坝变形预测模型及其应用[J]. 水电能源科学, 2023, 41(11): 77-81., articleTitle=基于多层次数据处理的NGO-XGBoost大坝变形预测模型及其应用, refAbstract=null), Reference(id=1207271197495083610, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2023, volume=41, issue=10, pageStart=67, pageEnd=71, url=null, language=null, rfNumber=[14], rfOrder=13, authorNames=杨爽, 薛晔, journalName=水电能源科学, refType=null, unstructuredReference=杨爽, 薛晔. 基于GGRA-GPR模型的洪涝灾害直接经济损失预评估[J]. 水电能源科学, 2023, 41(10): 67-71., articleTitle=基于GGRA-GPR模型的洪涝灾害直接经济损失预评估, refAbstract=null), Reference(id=1207271197608329827, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, doi=null, pmid=null, pmcid=null, year=2024, volume=44, issue=5, pageStart=1012, pageEnd=1028, url=null, language=null, rfNumber=[15], rfOrder=14, authorNames=张建明, 侍克斌, 贾运甫, journalName=隧道建设(中英文), refType=null, unstructuredReference=张建明, 侍克斌, 贾运甫, 等. 基于VMD与加权RF的TBM掘进速度预测SHAP解释模型[J]. 隧道建设(中英文), 2024, 44(5): 1012-1028., articleTitle=基于VMD与加权RF的TBM掘进速度预测SHAP解释模型, refAbstract=null)], funds=[Fund(id=1207271194450018777, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, awardId=202102AF080001, language=CN, fundingSource=云南省重大科技专项计划项目(202102AF080001), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1207271189131640950, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, xref=null, ext=[AuthorCompanyExt(id=1207271189140029559, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, companyId=1207271189131640950, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China), AuthorCompanyExt(id=1207271189169389689, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, companyId=1207271189131640950, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=河海大学水利水电学院,江苏 南京 210098)])], figs=[ArticleFig(id=1207271193162367345, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=EN, label=Fig. 1, caption=Prediction process of the VMD-DBO-Stacking integrated forecasting model, figureFileSmall=IMMWB9e5qy9zDf0OzGWQLw==, figureFileBig=3T4HYYetPxLrxm8O0aumSw==, tableContent=null), ArticleFig(id=1207271193288196478, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=CN, label=图1, caption=VMD-DBO-Stacking集成预测模型预测流程, figureFileSmall=IMMWB9e5qy9zDf0OzGWQLw==, figureFileBig=3T4HYYetPxLrxm8O0aumSw==, tableContent=null), ArticleFig(id=1207271193627935131, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=EN, label=Fig. 2, caption=Prediction result of shield tunneling speed using an ensemble model, figureFileSmall=ND1PTMl4/ORGHHrgwbRAsQ==, figureFileBig=uWctdVJ5e3JZjYsolemctA==, tableContent=null), ArticleFig(id=1207271193745375653, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=CN, label=图2, caption=集成模型盾构掘进速度预测结果, figureFileSmall=ND1PTMl4/ORGHHrgwbRAsQ==, figureFileBig=uWctdVJ5e3JZjYsolemctA==, tableContent=null), ArticleFig(id=1207271193925730738, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=EN, label=Fig. 3, caption=Prediction result of various models, figureFileSmall=wC1YzE9nKSPVXvJWsBAFHg==, figureFileBig=efPad9caq3LZxyBFc24nTQ==, tableContent=null), ArticleFig(id=1207271194026394042, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=CN, label=图3, caption=各模型预测结果, figureFileSmall=wC1YzE9nKSPVXvJWsBAFHg==, figureFileBig=efPad9caq3LZxyBFc24nTQ==, tableContent=null), ArticleFig(id=1207271194131251651, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=EN, label=Tab. 1, caption=

Results of evaluation indicators for different models

, figureFileSmall=null, figureFileBig=null, tableContent=
预测模型RRMSE/(mm·min-1MMAE/(mm·min-1R2
VMD-DBO-Stacking1.2211.0900.972
VMD-DBO-SVR2.2221.6650.902
VMD-DBO-RF2.1041.6260.911
VMD-DBO-XGBoost1.9191.5530.926
), ArticleFig(id=1207271194311606733, tenantId=1146029695717560320, journalId=1205116964453384197, articleId=1207271182483669821, language=CN, label=表1, caption=

各模型评价指标结果

, figureFileSmall=null, figureFileBig=null, tableContent=
预测模型RRMSE/(mm·min-1MMAE/(mm·min-1R2
VMD-DBO-Stacking1.2211.0900.972
VMD-DBO-SVR2.2221.6650.902
VMD-DBO-RF2.1041.6260.911
VMD-DBO-XGBoost1.9191.5530.926
)], attaches=null, journal=Journal(id=1205114436760272898, delFlag=0, nameCn=水电能源科学, nameEn=Water Resources and Power, nameHistory1=null, nameHistory2=null, issn=1000-7709, eissn=, cn=42-1231/TK, coden=null, periodic=月刊, language=CN, oaType=1, 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=, officePhone=, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=MdM62FQAbNFxgeQbfsV63A==, journalPrice=null, startedYear=null, abbrevIsoEn=Water Resources and Power, journalRemark=null, publicationField=null, createdTime=1765251271671, updatedTime=1765529918575, createdBy=18614031015, updatedBy=13041195026, firstLetterCn=W, firstLetterEn=W, subjectCode=Engineering, subjectName=Engineering, subjectCodeEn=Engineering, subjectNameEn=null, picCn=MdM62FQAbNFxgeQbfsV63A==, picEn=bVvVzb1Vw5wPWDTRDsA6dw==, jcr=null, cjcr=null, exts=[JournalExt(id=1206283167728730402, language=CN, name=水电能源科学, 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=1765529918844, updatedTime=1765529918844, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=https://sdny.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://sdny.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://sdny.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1206283167800033571, language=EN, name=Water Resources and Power, 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=1765529918861, updatedTime=1765529918861, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=https://sdny.cbpt.cnki.net/index.aspx?t=1, submissionEditorUrl=https://sdny.cbpt.cnki.net/index.aspx?t=3, submissionReviewUrl=https://sdny.cbpt.cnki.net/index.aspx?t=2, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1205116964453384197, websiteList=[Website(id=1205118568041001173, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1205116964453384197, 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/sdnykx/CN, language=CN, createTime=1765252256641, createBy=18614031015, updateTime=1765252461034, updateBy=18614031015, name=水电能源科学-中文, tplId=1146099689490845704, title=水电能源科学, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1205152539890131310, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568041001173, code=articleTextType, value=kx, createTime=1765260356161, updateTime=1765260356161, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152539864965483, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568041001173, code=banner, value=null, createTime=1765260356155, updateTime=1765260356155, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152539919491441, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568041001173, code=grayFlag, value=0, createTime=1765260356168, updateTime=1765260356168, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152539856576874, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568041001173, code=logo, value=https://castjournals.cast.org.cn/joweb/sdnykx/CN/file/pic?fileId=bX+sWBfK/Ed3Ho8UDVuTcA==, createTime=1765260356153, updateTime=1765260356153, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152539936268659, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568041001173, code=minRunFlag, value=0, createTime=1765260356172, updateTime=1765260356172, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152539881742701, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568041001173, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/sdnykx/CN/file/pic, createTime=1765260356159, updateTime=1765260356159, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152539927880050, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568041001173, code=silenceFlag, value=0, createTime=1765260356170, updateTime=1765260356170, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152539873354092, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568041001173, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1765260356157, updateTime=1765260356157, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152539902714223, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568041001173, code=themeColor, value=null, createTime=1765260356164, updateTime=1765260356164, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152539911102832, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568041001173, code=themeStyle, value=null, createTime=1765260356166, updateTime=1765260356166, creator=18614031015, updator=18614031015)]), Website(id=1205118568120692965, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1205116964453384197, 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/sdnykx/EN, language=EN, createTime=1765252256660, createBy=18614031015, updateTime=1765252477325, updateBy=18614031015, name=水电能源科学-英文, tplId=1146101810881728533, title=Water Resources and Power, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1205152566700122488, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568120692965, code=articleTextType, value=kx, createTime=1765260362553, updateTime=1765260362553, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152566679150965, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568120692965, code=banner, value=null, createTime=1765260362548, updateTime=1765260362548, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152566721094011, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568120692965, code=grayFlag, value=0, createTime=1765260362558, updateTime=1765260362558, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152566670762356, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568120692965, code=logo, value=https://castjournals.cast.org.cn/joweb/sdnykx/EN/file/pic?fileId=bX+sWBfK/Ed3Ho8UDVuTcA==, createTime=1765260362546, updateTime=1765260362546, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152566742065533, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568120692965, code=minRunFlag, value=0, createTime=1765260362563, updateTime=1765260362563, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152566695928183, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568120692965, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/sdnykx/EN/file/pic, createTime=1765260362552, updateTime=1765260362552, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152566733676924, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568120692965, code=silenceFlag, value=0, createTime=1765260362561, updateTime=1765260362561, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152566687539574, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568120692965, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1765260362550, updateTime=1765260362550, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152566708511097, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568120692965, code=themeColor, value=null, createTime=1765260362555, updateTime=1765260362555, creator=18614031015, updator=18614031015), WebsiteProps(id=1205152566716899706, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1205118568120692965, code=themeStyle, value=null, createTime=1765260362557, updateTime=1765260362557, creator=18614031015, updator=18614031015)])], journalTitle=水电能源科学, weixinUrl=null, journalUrl=null, iacademicId=null, status=1, seqNo=null, journalTitleEn=Water Resources and Power, journalPhotoCn=MdM62FQAbNFxgeQbfsV63A==, journalPhotoEn=bVvVzb1Vw5wPWDTRDsA6dw==, journalFirstLetter=W, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/sdnykx/CN/10.20040/j.cnki.1000-7709.2025.20242243, detailUrlEn=https://castjournals.cast.org.cn/joweb/sdnykx/EN/10.20040/j.cnki.1000-7709.2025.20242243, pdfUrlCn=https://castjournals.cast.org.cn/joweb/sdnykx/CN/PDF/10.20040/j.cnki.1000-7709.2025.20242243, pdfUrlEn=https://castjournals.cast.org.cn/joweb/sdnykx/EN/PDF/10.20040/j.cnki.1000-7709.2025.20242243, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于VMD-DBO-Stacking集成学习的盾构掘进速度预测模型
收藏切换
PDF下载
邓子昂 , 张玉贤 , 张继勋
水电能源科学 | 水工结构、水工材料与水利工程施工 2025,43(9): 101-105
收起
收藏切换
水电能源科学 | 水工结构、水工材料与水利工程施工 2025, 43(9): 101-105
基于VMD-DBO-Stacking集成学习的盾构掘进速度预测模型
全屏
邓子昂 , 张玉贤, 张继勋
作者信息
  • 河海大学水利水电学院,江苏 南京 210098
  • 邓子昂(2001-),男,硕士研究生,研究方向为水工地下工程信息化,E-mail:

通讯作者:

张继勋(1974-),男,博士、副教授、硕导,研究方向为水工地下工程信息化,E-mail:
Prediction Model of Shield Tunneling Speed Based on VMD-DBO-Stacking Ensemble Learning
Zi-ang DENG , Yu-xian ZHANG, Ji-xun ZHANG
Affiliations
  • College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China
出版时间: 2025-09-25 doi: 10.20040/j.cnki.1000-7709.2025.20242243
文章导航
收藏切换

针对现有盾构掘进速度预测方法存在的模型算法单一、精度不高和泛化性较差等问题,为了提高盾构掘进速度预测精度,建立一种基于变分模态分解(VMD)、蜣螂优化算法(DBO)和Stacking(VMD-DBO-Stacking)集成学习的盾构掘进速度预测模型。首先,为了得到更有效的数据,采用VMD对原始数据进行分解重构得到去噪后的施工参数数据用于后续模型预测;其次,基于集成学习策略,选取支持向量回归(SVR)模型、随机森林(RF)模型、极端梯度提升(XGBoost)模型作为基学习器,高斯过程回归(GPR)模型作为元学习器,从而构建预测精度更高、泛化性更强的Stacking集成学习预测模型;然后,为了进一步提高预测精度,采用DBO对集成学习模型进行超参数优化;最后,将此预测模型用于河南某引水隧洞工程盾构施工中并与其他预测模型进行对比。结果表明,与其他单一模型(SVR、RF、XGBoost)相比,所建模型具有更高的预测精度,平均精度分别提升7.76%、6.70%、4.97%,为盾构掘进速度预测提供一种新思路。

盾构  /  掘进速度  /  变分模态分解  /  蜣螂优化算法  /  Stacking集成学习

Addressing the issues of single model algorithm, low accuracy, and poor generalization in existing shield tunneling speed prediction methods, this study proposes a shield tunneling speed prediction approach to improve prediction accuracy based on Variational Mode Decomposition (VMD), Dung Beetle Optimizer (DBO), and Stacking ensemble learning. Firstly, to obtain more effective data, VMD is applied to decompose and reconstruct the original data to obtain denoised construction parameter data for subsequent model prediction. Secondly, based on the ensemble learning strategy, Support Vector Regression (SVR), Random Forest (RF), and Extreme Gradient Boosting (XGBoost) models are selected as base learners, while Gaussian Process Regression (GPR) is chosen as the meta-learner to construct a Stacking ensemble learning prediction model with higher prediction accuracy and stronger generalization ability. Thirdly, to further enhance prediction accuracy, DBO is employed to optimize the hyperparameters of the ensemble learning model. Finally, this prediction method is applied to the shield tunneling construction of a water diversion tunnel project in Henan Province and compared with other prediction methods. Compared to other single models (SVR, RF, XGBoost), the results indicate that the proposed method achieves higher prediction accuracy, with average accuracy improvements of 7.76%, 6.70%, and 4.97%, respectively, providing a new approach for shield tunneling speed prediction.

shield tunneling machine  /  tunneling speed  /  variational mode decomposition  /  Dung Beetle Optimizer  /  Stacking ensemble learning
邓子昂, 张玉贤, 张继勋. 基于VMD-DBO-Stacking集成学习的盾构掘进速度预测模型. 水电能源科学, 2025 , 43 (9) : 101 -105 . DOI: 10.20040/j.cnki.1000-7709.2025.20242243
Zi-ang DENG, Yu-xian ZHANG, Ji-xun ZHANG. Prediction Model of Shield Tunneling Speed Based on VMD-DBO-Stacking Ensemble Learning[J]. Water Resources and Power, 2025 , 43 (9) : 101 -105 . DOI: 10.20040/j.cnki.1000-7709.2025.20242243
近年来,盾构法施工因其具有施工快速安全、对地层扰动小等优点成为大部分隧洞工程的施工方法[1]。在实际工程中,盾构掘进效率是盾构机性能的一个关键指标,而掘进速度是影响掘进效率的主要因素[2],掘进速度控制不当不仅会影响施工时间,还会导致隧洞塌方和地表沉降过大等问题,因此建立能够精准预测盾构掘进速度的预测模型对保证工程进度、安全并降低工程成本至关重要。随着大数据与人工智能的飞速发展,机器学习因其强大的数据处理和分析能力,逐渐被用于盾构预测领域。张哲铭等[3]采用支持向量机(SVR)结合最小二乘法(LS)对TBM的掘进速度进行了预测,验证了LS-SVR机器学习预测TBM参数的可行性;仉文岗等[4]采用随机森林(RF)预测模型结合4种超参数优化算法,实现了对盾构掘进速度的预测;赵光祖等[5]采用模拟退火算法(SA)和遗传算法(GA)优化BP神经网络从而建立TBM性能预测模型,并对掘进速度进行预测。然而,单一的预测模型容易低估输入参数不确定性,从而导致预测结果鲁棒性不足[6]。鉴此,本文基于集成学习策略,建立一种基于变分模态分解(VMD)、蜣螂优化算法(DBO)和Stacking的(VMD-DBO-Stacking)集成预测模型,即首先采用VMD对盾构施工数据进行重构去噪,然后采用SVR、RF、XGBoost、GPR建立Stacking集成模型,并采用DBO优化其超参数,从而实现对盾构掘进速度的预测。实例应用表明所建的预测模型具有更高的精度和泛化性能,为盾构掘进速度预测提供一种新方法。
变分模态分解是一种完全非递归的自适应信号分解方法[7]。主要可分为变分问题的构造和求解两部分。
数据经VMD分解可得到K个IMF分量,将每个IMF定义为调幅调频函数,其表示为:
式中,ukt)为调幅调频函数;Akt)为模态的瞬时幅值;ϕkt)为相位函数。
对各个调幅调频函数,经过希尔伯特变换得到解析信号,然后加入指数项将对应的频谱调制到相应的基带上,最后估计模态的信号带宽并引入限制条件,构建变分问题如下:
式中,uk为分解后第k个频率分量;ωk为第k个中心频率;K为模态个数;∂t为偏导数算子;δt)为关于时间t的狄拉克分布;j为虚数,-j2=1;t为时间;*为卷积算子;ft)为原始信号。
将二次惩罚因子和拉格朗日乘数与式(2)结合,得到的增广拉格朗日其表达式为:
式中,L(·)为拉格朗日函数表达式;α为二次惩罚因子;λt)为拉格朗日乘数。
通过交替方向乘子法求解式(3)的最优解,模态分量的更新公式为:
式中,n为迭代次数;ω为频率参数;为维纳滤波剩余量;fω)、uiω)、λω)分别为ft)、uit)、λt)的傅里叶变换;τ为拉格朗日乘子项步长。
蜣螂优化算法(DBO)是一种基于蜣螂生物行为过程提出的群体智能优化算法,比起粒子群算法、麻雀搜索算法、灰狼优化算法等传统优化算法具有收敛速度快、收敛精度高和稳定性能好等优势[8]。蜣螂优化算法主要分为滚球、繁衍、觅食和偷窃4部分,其运行流程包括初始化蜣螂种群矩阵、评估各蜣螂的适应度、更新蜣螂位置并确保在搜索范围内以及迭代更新最优解,直至达到全局最优。具体公式可参考文献[8]。
Stacking集成模型通过多个基学习器和一个元学习器优化整合预测,充分利用数据集和各学习器优势,从而提升预测精度。为了保证基学习器选择的合理性和准确性,在已有研究的基础上,选取在掘进速度预测领域已证实有效的单一预测模型[4,9-10]作为基学习器,即SVR、RF、XGBoost作为3个基学习器,采用GPR作为元学习器,关于这些机器学习模型的原理和介绍可参考文献[11-14],将3个基学习器和元学习器其组合成Stacking集成模型,其具体步骤如下。
步骤1 将盾构掘进数据按照8:2的比例划分为训练集T和测试集S。对训练集T进行五折交叉验证:将训练集分成5等份,每一折选取4份为训练数据,剩下的1份为验证数据。以任意一折为例,使用训练数据对基学习器进行训练,分别得到其在验证数据上的预测值Yt=y1ty2ty3ty4ty5t)和在测试集S上的预测值Ut=u1tu2tu3tu4tu5t)。其中YtUt分别表示第t个基学习器。
步骤2 将验证数据的预测值进行重新组合形成元学习器的训练样本,在测试集上的预测值进行平均取值形成元学习器的测试样本
步骤3 另外两个基学习器重复步骤1、2,将最后得到的元学习器的训练样本和测试样本进行平均取值得到最终元学习器的训练集和测试集。
步骤4 使用步骤3中得到的数据集对元学习器进行训练测试,并得到最终的预测结果。
VMD-DBO-Stacking盾构掘进速度预测模型的基本流程见图1,具体步骤如下。
步骤1 首先对原始盾构施工数据进行预处理,然后采用VMD算法进行去噪处理,获取去噪后的盾构施工数据,最后对施工数据进行归一化处理。
步骤2 将处理后的施工数据按比例划分为训练集和测试集并输入Stacking集成模型中,并采用DBO算法对集成模型中基学习器进行超参数优化,确定最优超参数组合。
步骤3 根据最优超参数组合建立VMD-DBO-Stacking集成预测模型,并利用模型对盾构掘进速度进行预测,得到最终预测结果。
步骤4 采用均方根误差(RRMSE)、平方绝对误差(MMAE)、样本回归值(R2)作为掘进速度预测的评价指标对预测性能进行评价。
河南某引水隧洞盾构工程隧洞全长2 545 m,内径6.8 m,宽1.5 m,主要穿越粉质粘土、重粉质壤土、中砂层、细砂层。该工程采用泥水平衡盾构,工程前期主要穿越砂土地层,地层基本不发生变化,因此忽略地层参数的影响。随着盾构施工的掘进,盾构施工的主要参数被实时上传至控制系统,该工程的原始数据每60 s采集1次,选取该工程2021年2月25日~4月27日的数据,共65 536组,盾构掘进参数21个(掘进速度、刀盘转速、刀盘扭矩、总推进力、润滑油脂压力、刀盘变频器输出频率、推进泵出口压力、推进油缸压力、推进油缸行程、铰接油缸压力、铰接油缸行程、注浆压力、外部水源压力、浆液流量、进泥口压力、出泥口压力、排浆泵电流、排浆泵速度、开挖仓压力、工作仓压力、连接桥拖拉压力)。
由于盾构机启闭及掘进过程中会出现不稳定的数据波动,因此在分析数据前需清洗异常值,从而筛选出有效数据。首先要剔除原始数据集中的大量空数据,即删除盾构机处于非工作状态的数据。采用构造二值状态判别函数的方法进行判定,主要公式为:
式中,M为状态判别函数;RTFV分别为刀盘转速、刀盘扭矩、总推进力、掘进速度。
在实际工程中,若与刀盘转速、刀盘扭矩、总推进力、掘进速度有关的任意参数为0,则可认为盾构机处于非工作状态,即式(5)中M=0的数据对应盾构机非工作状态,应被剔除。
然后采用VMD对数据进行去噪处理。根据文献[15]中方法进行VMD参数选择,首先设定惩罚因子为2 000,模态分解个数的取值范围为[2,10],接着用不同模态分解个数值对原始数据集进行VMD分解。对每个模态分解个数值,将分解得到的子序列重构并计算样本熵。通过比较样本熵,选择样本熵最小值对应的模态分解个数值作为最优参数。最终确定VMD的模态分解个数和惩罚因子分别设为9、2 000。样本熵公式为:
式中,m为维数;r为相似容限;N为数据长度;Bm+1r)、Bmr)分别为在m+1、m维下相似容限为r时的相似数据点的数量。
VMD去噪具体步骤如下。
步骤1 根据模态分解个数将输入数据分解为9个IMFs,并根据式(6)得到每个IMF的样本熵。
步骤2 根据两个相邻的变分模态的样本熵差,可确定有效的IMFs,此时IMFs突变点表示为:
式中,SE(·)为样本熵值;IIMFiIIMFi+1分别为分解得到的第ii+1个本征模态函数分量。
步骤3 重建有效的IMFs。此时去噪后的数据可表示为:
步骤4 为了减小数据之间不同量纲和尺度差异的影响,对输入、输出的数据按照最大最小化准则进行归一化处理。其公式为:
式中,x′为样本归一化后的数据值;x为原始数据值;xmaxxmin分别为某一参数的最大值、最小值。
经过数据处理后得到掘进数据14 056组,为计算方便,仅选取前3 500组数据进行计算演示,将80%数据作为训练集,20%数据作为测试集,选取掘进速度作为输出参数,其余参数为输入参数。采用五折交叉验证训练Stacking集成模型,同时采用DBO优化其超参数。其中DBO最大迭代次数为100,种群数为30,对基学习器SVR惩罚因子和核函数系数、RF决策树数量和最大深度以及XGBoost学习率、决策树数量和最大深度进行寻优;运用训练好的DBO-Stacking模型对盾构掘进速度进行预测,其预测结果见图2
图2可看出,本文所建立的VMD-DBO-Stacking集成预测模型的盾构掘进速度预测值与真实值具有较高的一致性,其掘进速度的最大预测残差为1.08 mm/min,相对误差为1.55%,由此证明了本文所建立的集成预测模型具有较好的预测效果。
为了验证本文集成预测模型的优越性,对比分析了本文模型与各个基学习器(VMD-DBO-SVR、VMD-DBO-RF、VMD-DBO-XGBoost)构建的单一预测模型,各模型预测结果见图3,预测评价指标结果见表1
图3可看出,4种模型均能够较好地预测出盾构掘进速度,但与其他单一预测模型相比集成预测模型预测结果更加贴合真实值,其相对误差更小。可见本文建立的基于VMD-DBO-Stacking集成预测模型具有更高的预测精度。由表1可知,单一模型的预测精度均在0.9以上,说明SVR、RF、XGBoost作为基学习器的可行性与合理性,而集成预测模型的RRMSEMMAER2分别为1.221 mm/min、1.09 mm/min、0.972,相比于SVR、RF、XGBoost预测模型,RRMSE降低了45.0%、42.0%、36.4%,MMAE降低了34.5%、33.0%、29.8%,R2提高了7.76%、6.70%、4.97%,说明集成学习模型预测性能要高于单一预测模型。
a. 通过构造二值状态判别函数和VMD去噪处理有效提升了原始盾构施工数据的质量,为后续模型预测提供了更准确的数据集。
b. 采用集成学习策略,以SVR、RF、XGBoost作为基学习器,GPR作为元学习器构建Stacking集成学习预测模型,并采用DBO分别优化基学习器的超参数,进一步提升了模型的预测精度。这为盾构掘进速度预测模型的改进提供了一种新的方法。
c. 与单一模型(SVR、RF、XGBoost)相比,本文所建立的VMD-DBO-Stacking集成预测模型在预测精度上分别提升了7.76%、6.70%、4.97%,展现出更高的预测精度和泛化能力。
  • 云南省重大科技专项计划项目(202102AF080001)
参考文献 引证文献
排序方式:
[1]
李强, 李子阳, 许林军, 等. 引水隧洞盾构法施工风险源辨识与分析[J]. 人民黄河, 2023, 45(4): 121-125.
[2]
裴浩东, 叶社保, 杨平, 等. 软土地层盾构掘进参数分析及掘进速度预测[J]. 郑州大学学报(工学版), 2024, 45(1): 107-113.
[3]
张哲铭, 李晓瑜, 姬建. 基于LS-SVM的TBM掘进参数预测模型[J]. 河海大学学报(自然科学版), 2021, 49(4): 373-379.
[4]
仉文岗, 唐理斌, 陈福勇, 等. 基于4种超参数优化算法及随机森林模型预测TBM掘进速度[J]. 应用基础与工程科学学报, 2021, 29(5): 1186-1200.
[5]
赵光祖, 王亚旭, 李尧, 等. 基于优化BP神经网络的TBM性能预测[J]. 河南理工大学学报(自然科学版), 2020, 39(5): 139-145.
[6]
祝玉珊, 王晓玲, 崔博, 等. 坝基灌浆量预测ISSA-Stacking集成学习代理模型研究[J]. 天津大学学报(自然科学与工程技术版), 2024, 57(2): 174-185.
[7]
江星星, 宋秋昱, 杜贵府, 等. 变分模式分解方法研究与应用综述[J]. 仪器仪表学报, 2023, 44(1): 55-73.
[8]
娄革伟, 郑永煌, 陈均, 等. 混合多策略改进的蜣螂优化算法[J]. 计算机工程与应用, 2024, 60(24): 97-109.
[9]
彭继慎, 郝茗, 宋立业, 等. 基于TSSA-SVR算法的TBM掘进速度预测[J]. 辽宁工程技术大学学报(自然科学版), 2023, 42(5): 634-640.
[10]
杜庆峰, 张双俐, 张晨曦, 等. 基于均值滤波去噪和XGBoost算法的泥水平衡盾构掘进速度预测方法[J]. 现代隧道技术, 2022, 59(6): 14-23.
[11]
杨辉斌, 郑德仁, 王贺龙, 等. 基于GA-SVR的管网异常漏损检测[J]. 水电能源科学, 2024, 42(3): 133-136, 53.
[12]
柏晓鹏, 南瑞川, 杜甫, 等. 基于随机森林的地下水溶质运移替代模型研究[J]. 水电能源科学, 2024, 42(11): 60-63.
[13]
李晨阳, 郑东健. 基于多层次数据处理的NGO-XGBoost大坝变形预测模型及其应用[J]. 水电能源科学, 2023, 41(11): 77-81.
[14]
杨爽, 薛晔. 基于GGRA-GPR模型的洪涝灾害直接经济损失预评估[J]. 水电能源科学, 2023, 41(10): 67-71.
[15]
张建明, 侍克斌, 贾运甫, 等. 基于VMD与加权RF的TBM掘进速度预测SHAP解释模型[J]. 隧道建设(中英文), 2024, 44(5): 1012-1028.
2025年第43卷第9期
PDF下载
231
95
引用本文
BibTeX
文章信息
doi: 10.20040/j.cnki.1000-7709.2025.20242243
  • 接收时间:2024-10-28
  • 首发时间:2025-12-15
  • 出版时间:2025-09-25
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-10-28
  • 修回日期:2024-12-16
基金
云南省重大科技专项计划项目(202102AF080001)
作者信息
    河海大学水利水电学院,江苏 南京 210098

通讯作者:

张继勋(1974-),男,博士、副教授、硕导,研究方向为水工地下工程信息化,E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/sdnykx/CN/10.20040/j.cnki.1000-7709.2025.20242243
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