Article(id=1148106715691737606, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1148106709542892487, articleNumber=1003-3033(2025)04-0137-08, orderNo=null, doi=10.16265/j.cnki.issn1003-3033.2025.04.1398, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1731600000000, receivedDateStr=2024-11-15, revisedDate=1739462400000, revisedDateStr=2025-02-14, acceptedDate=null, acceptedDateStr=null, onlineDate=1751659571812, onlineDateStr=2025-07-05, pubDate=1745769600000, pubDateStr=2025-04-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1751659571812, onlineIssueDateStr=2025-07-05, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1751659571812, creator=13701087609, updateTime=1751659571812, updator=13701087609, issue=Issue{id=1148106709542892487, tenantId=1146029695717560320, journalId=1146031787341344770, year='2025', volume='35', issue='4', pageStart='1', pageEnd='264', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=1, specialIssue=0, createTime=1751659570346, creator=13701087609, updateTime=1757560692417, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1172857809499730113, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1148106709542892487, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1172857809499730114, tenantId=1146029695717560320, journalId=1146031787341344770, issueId=1148106709542892487, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=137, endPage=144, ext={EN=ArticleExt(id=1149758072526127123, articleId=1148106715691737606, tenantId=1146029695717560320, journalId=1146031787341344770, language=EN, title=Hazard prediction model of tunnel water inrush based on stacking ensemble learning, columnId=1149733269173878863, journalTitle=China Safety Science Journal, columnName=Safety engineering technology, runingTitle=null, highlight=null, articleAbstract=

In order to solve the problems that machine learning exists in the hazard intelligent prediction field of tunnel water inrush,such as relatively simple models and imperfect prediction accuracy,a prediction model based on the stacking ensemble learning was proposed. Firstly,the tunnel water inrush disaster dataset was established by collecting 232 groups of water inrush disaster data from 95 tunnels,and the data was preprocessed. Then,3 base learners and 2 meta learners were selected to train 8 sets of stacking ensemble models in different combinations,and 6 sets of optimal ensemble models were selected. Finally,the optimal stacking ensemble model was selected by comparing and analyzing the prediction results of 6 groups of parameters optimized and stacking ensemble model with the grid search parameters and the 5-fold cross-validation hyperparameter optimization model. The results show that SVM(Support Vector Machine )+NB (Naive Bayes) + LR (Linear Regression) ensemble model is obtained after the optimal single model SVM is improved with the stacking ensemble learning method. Its accuracy,recall,and F1 score are 0.94,0.91,and 0.92,respectively. The overall prediction effect is better than that of other compative models,and it can accurately predict the hazard level of tunnel water inrush.

, correspAuthors=Nian ZHANG, 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=Jiale LU, Nian ZHANG, Mengmeng NIU, Fei WAN), CN=ArticleExt(id=1148106722197102798, articleId=1148106715691737606, tenantId=1146029695717560320, journalId=1146031787341344770, language=CN, title=基于Stacking集成学习的隧道突水危险预测模型, columnId=1149733269727526997, journalTitle=中国安全科学学报, columnName=安全工程技术, runingTitle=null, highlight=null, articleAbstract=

为解决机器学习在隧道突水危险智能预测领域存在的模型较单一和预测精度不够理想等问题,提出一种基于Stacking集成学习方法的预测模型。首先,通过搜集95条隧道共计232组隧道突水灾害数据建立隧道突水灾害数据集,并进行数据预处理;然后,选取3种基学习器和2种元学习器以不同组合方式训练出8组Stacking集成模型,并筛选出6组较优的集成模型;最后,使用网格搜索调参并结合5折交叉验证超参数调优模型,对比分析6组参数调优后的Stacking集成模型的预测结果,选择出最优Stacking集成模型。结果表明:采用Stacking集成学习方法改进最优单模型支持向量机(SVM)后得到SVM+朴素贝叶斯(NB )+线性回归(LR)集成模型,其精确率、召回率和F1分数分别达到0.94、0.91和0.92,整体预测效果优于其他对比模型,可准确预测隧道突水危险等级。

, correspAuthors=张念 副教授, authorNote=null, correspAuthorsNote=
**张 念(1984—),男,湖北襄阳人,博士,副教授,主要从事隧道及地下工程安全与算法优化方面的研究。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=slkLZHoUy/Bc6Ki72pkjCg==, magXml=Bh9pozCons1r1qqvnkT3jA==, pdfUrl=null, pdf=olDoorlNQhx6fj87zcTN1g==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=i4qdORtHjCHrPb6ENC1HYA==, mapNumber=null, authorCompany=null, fund=null, authors=

卢佳乐 (1998—),男,山西运城人,硕士研究生,主要研究方向为隧道工程安全与算法优化。E-mail:

, authorsList=卢佳乐, 张念 副教授, 牛萌萌, 万飞 研究员)}, authors=[Author(id=1165198346412106717, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=lujiale0477@link.tyut.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1165198346466632672, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, authorId=1165198346412106717, language=EN, stringName=Jiale LU, firstName=Jiale, middleName=null, lastName=LU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 College of Civil Engineering,Taiyuan University of Technology,Taiyuan Shanxi 030024,China
2 Research Center of Tunneling and Underground Engineering of Ministry of Education,Beijing Jiaotong University,Beijing 100044,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1165198346533741538, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, authorId=1165198346412106717, language=CN, stringName=卢佳乐, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 太原理工大学 土木工程学院,山西 太原 030024
2 北京交通大学 隧道及地下工程教育部工程研究中心,北京 100044, bio={"img":"5wasjIG1R3wG6QvV7UIRtg==","content":"

卢佳乐 (1998—),男,山西运城人,硕士研究生,主要研究方向为隧道工程安全与算法优化。E-mail:

"}, bioImg=5wasjIG1R3wG6QvV7UIRtg==, bioContent=

卢佳乐 (1998—),男,山西运城人,硕士研究生,主要研究方向为隧道工程安全与算法优化。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1165198346194002896, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, xref=1, ext=[AuthorCompanyExt(id=1165198346202391505, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346194002896, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Civil Engineering,Taiyuan University of Technology,Taiyuan Shanxi 030024,China), AuthorCompanyExt(id=1165198346206585810, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346194002896, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 太原理工大学 土木工程学院,山西 太原 030024)]), AuthorCompany(id=1165198346269500372, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, xref=2, ext=[AuthorCompanyExt(id=1165198346277888981, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346269500372, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Research Center of Tunneling and Underground Engineering of Ministry of Education,Beijing Jiaotong University,Beijing 100044,China), AuthorCompanyExt(id=1165198346282083286, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346269500372, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京交通大学 隧道及地下工程教育部工程研究中心,北京 100044)])]), Author(id=1165198346600850406, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=zhangnian@tyut.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1165198346676347882, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, authorId=1165198346600850406, language=EN, stringName=Nian ZHANG, firstName=Nian, middleName=null, lastName=ZHANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, **, address=1 College of Civil Engineering,Taiyuan University of Technology,Taiyuan Shanxi 030024,China
2 Research Center of Tunneling and Underground Engineering of Ministry of Education,Beijing Jiaotong University,Beijing 100044,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1165198346760233965, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, authorId=1165198346600850406, language=CN, stringName=张念 副教授, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, **, address=1 太原理工大学 土木工程学院,山西 太原 030024
2 北京交通大学 隧道及地下工程教育部工程研究中心,北京 100044, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1165198346194002896, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, xref=1, ext=[AuthorCompanyExt(id=1165198346202391505, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346194002896, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Civil Engineering,Taiyuan University of Technology,Taiyuan Shanxi 030024,China), AuthorCompanyExt(id=1165198346206585810, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346194002896, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 太原理工大学 土木工程学院,山西 太原 030024)]), AuthorCompany(id=1165198346269500372, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, xref=2, ext=[AuthorCompanyExt(id=1165198346277888981, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346269500372, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Research Center of Tunneling and Underground Engineering of Ministry of Education,Beijing Jiaotong University,Beijing 100044,China), AuthorCompanyExt(id=1165198346282083286, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346269500372, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京交通大学 隧道及地下工程教育部工程研究中心,北京 100044)])]), Author(id=1165198346831537135, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, 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=1165198346902840306, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, authorId=1165198346831537135, language=EN, stringName=Mengmeng NIU, firstName=Mengmeng, middleName=null, lastName=NIU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 College of Civil Engineering,Taiyuan University of Technology,Taiyuan Shanxi 030024,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1165198346969949171, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, authorId=1165198346831537135, language=CN, stringName=牛萌萌, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1 太原理工大学 土木工程学院,山西 太原 030024, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1165198346194002896, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, xref=1, ext=[AuthorCompanyExt(id=1165198346202391505, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346194002896, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Civil Engineering,Taiyuan University of Technology,Taiyuan Shanxi 030024,China), AuthorCompanyExt(id=1165198346206585810, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346194002896, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 太原理工大学 土木工程学院,山西 太原 030024)])]), Author(id=1165198347053835252, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1165198347129332727, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, authorId=1165198347053835252, language=EN, stringName=Fei WAN, firstName=Fei, middleName=null, lastName=WAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 Research Institute of Highway Ministry of Transport,Beijing 100088,China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1165198347217413112, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, authorId=1165198347053835252, language=CN, stringName=万飞 研究员, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 交通运输部公路科学研究所,北京 100088, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1165198346344997848, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, xref=3, ext=[AuthorCompanyExt(id=1165198346349192153, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346344997848, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Research Institute of Highway Ministry of Transport,Beijing 100088,China), AuthorCompanyExt(id=1165198346353386458, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346344997848, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 交通运输部公路科学研究所,北京 100088)])])], keywords=[Keyword(id=1165198347385185277, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, orderNo=1, keyword=Stacking ensemble learning), Keyword(id=1165198347452294144, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, orderNo=2, keyword=tunnel water inrush), Keyword(id=1165198347511013377, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, orderNo=3, keyword=prediction model), Keyword(id=1165198347586510851, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, orderNo=4, keyword=hazard level), Keyword(id=1165198347641036804, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, orderNo=5, keyword=machine learning), Keyword(id=1165198347703951366, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, orderNo=1, keyword=Stacking集成学习), Keyword(id=1165198347775254538, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, orderNo=2, keyword=隧道突水), Keyword(id=1165198347838169098, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, orderNo=3, keyword=预测模型), Keyword(id=1165198347909472268, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, orderNo=4, keyword=危险等级), Keyword(id=1165198347972386828, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, orderNo=5, keyword=机器学习)], refs=[Reference(id=1165198350581243986, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2006, volume=16, issue=1, pageStart=114, pageEnd=118, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=白明洲, 许兆义, 王连俊, journalName=中国安全科学学报, refType=null, unstructuredReference=白明洲, 许兆义, 王连俊, 等. 复杂岩溶地区隧道施工突水地质灾害研究[J]. 中国安全科学学报, 2006, 16(1):114-118., articleTitle=复杂岩溶地区隧道施工突水地质灾害研究, refAbstract=null), Reference(id=1165198350665130068, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2006, volume=16, issue=1, pageStart=114, pageEnd=118, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=BAI Mingzhou, XU Zhaoyi, WANG Lianjun, journalName=China Safety Science Journal, refType=null, unstructuredReference=BAI Mingzhou, XU Zhaoyi, WANG Lianjun, et al. Study on water outburst geological disaster of tunnel construction at complicated karst zone[J]. China Safety Science Journal, 2006, 16(1): 114-118., articleTitle=Study on water outburst geological disaster of tunnel construction at complicated karst zone, refAbstract=null), Reference(id=1165198350728044630, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2021, volume=114, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=XUE Yiguo, KONG Fanmeng, LI Shucai, journalName=Tunnelling and Underground Space Technology, refType=null, unstructuredReference=XUE Yiguo, KONG Fanmeng, LI Shucai, et al. Water and mud inrush hazard in underground engineering: genesis, evolution and prevention[J]. Tunnelling and Underground Space Technology, 2021, 114: DOI: 10.1016/J.TUST.2021.103987., articleTitle=Water and mud inrush hazard in underground engineering: genesis, evolution and prevention, refAbstract=null), Reference(id=1165198350786764888, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=3, authorNames=王璐, journalName=基于遗传算法和支持向量机的西南岩溶越岭隧道涌水量预测, refType=null, unstructuredReference=王璐. 基于遗传算法和支持向量机的西南岩溶越岭隧道涌水量预测[D]. 成都: 成都理工大学, 2019., articleTitle=null, refAbstract=null), Reference(id=1165198350853873754, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2019, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=WANG Lu, journalName=Prediction of water inflow in southwest karst crossing-mountain tunnel based on genetic algorithm and support vector machine, refType=null, unstructuredReference=WANG Lu. Prediction of water inflow in southwest karst crossing-mountain tunnel based on genetic algorithm and support vector machine[D]. Chengdu: Chengdu University of Technology, 2019., articleTitle=null, refAbstract=null), Reference(id=1165198350925176924, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2022, volume=32, issue=增1, pageStart=171, pageEnd=177, url=null, language=null, rfNumber=[4], rfOrder=5, authorNames=马天行, 林允, 周晓斌, journalName=中国安全科学学报, refType=null, unstructuredReference=马天行, 林允, 周晓斌, 等. 煤层底板突水危险性预测的熵权-正态云模型[J]. 中国安全科学学报, 2022, 32(增1):171-177., articleTitle=煤层底板突水危险性预测的熵权-正态云模型, refAbstract=null), Reference(id=1165198350983897182, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2022, volume=32, issue=S1, pageStart=171, pageEnd=177, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=MA Tianxing, LIN Yun, ZHOU Xiaobin, journalName=China Safety Science Journal, refType=null, unstructuredReference=MA Tianxing, LIN Yun, ZHOU Xiaobin, et al. Entropy weight-normal cloud model for water inrush risk prediction of coal seam floor[J]. China Safety Science Journal, 2022, 32(S1): 171-177., articleTitle=Entropy weight-normal cloud model for water inrush risk prediction of coal seam floor, refAbstract=null), Reference(id=1165198351055200352, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=7, authorNames=柏成浩, journalName=基于机器学习的岩溶隧道突水突泥灾害风险智能预测方法研究, refType=null, unstructuredReference=柏成浩. 基于机器学习的岩溶隧道突水突泥灾害风险智能预测方法研究[D]. 济南: 山东大学, 2021., articleTitle=null, refAbstract=null), Reference(id=1165198351122309218, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=BO Chenghao, journalName=Research on intelligent prediction method of hazard risk of water and mud inrush in karst tunnel based on machine learning, refType=null, unstructuredReference=BO Chenghao. Research on intelligent prediction method of hazard risk of water and mud inrush in karst tunnel based on machine learning[D]. Jinan: Shandong University, 2021., articleTitle=null, refAbstract=null), Reference(id=1165198351181029476, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2016, volume=null, issue=null, pageStart=171, pageEnd=null, url=null, language=null, rfNumber=[6], rfOrder=9, authorNames=周志华, journalName=机器学习, refType=null, unstructuredReference=周志华. 机器学习[M]. 北京: 清华大学出版社, 2016:171-190., articleTitle=null, refAbstract=null), Reference(id=1165198351248138342, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2023, volume=406, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=10, authorNames=GUO Wangda, ZHANG Jinxi, MURTAZA M, journalName=Construction and Building Materials, refType=null, unstructuredReference=GUO Wangda, ZHANG Jinxi, MURTAZA M, et al. An ensemble learning with sequential model-based optimization approach for pavement roughness estimation using smartphone sensor data[J]. Construction and Building Materials, 2023, 406: DOI:10.1016/J.CONBUILDMAT.2023.133293., articleTitle=An ensemble learning with sequential model-based optimization approach for pavement roughness estimation using smartphone sensor data, refAbstract=null), Reference(id=1165198351306858600, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=1992, volume=5, issue=2, pageStart=241, pageEnd=259, url=null, language=null, rfNumber=[8], rfOrder=11, authorNames=WOLPERT D H, journalName=Neural Networks, refType=null, unstructuredReference=WOLPERT D H. Stacked generalization[J]. Neural Networks, 1992, 5(2): 241-259., articleTitle=Stacked generalization, refAbstract=null), Reference(id=1165198351357190250, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2024, volume=14, issue=2, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[9], rfOrder=12, authorNames=ZHANG Nian, NIU Mengmeng, WAN Fei, journalName=Applied Sciences, refType=null, unstructuredReference=ZHANG Nian, NIU Mengmeng, WAN Fei, et al. Hazard prediction of water inrush in water-rich tunnels based on random forest algorithm[J]. Applied Sciences, 2024, 14(2):DOI:10.3390/APP14020867., articleTitle=Hazard prediction of water inrush in water-rich tunnels based on random forest algorithm, refAbstract=null), Reference(id=1165198351436882028, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2022, volume=14, issue=4, pageStart=1292, pageEnd=1303, url=null, language=null, rfNumber=[10], rfOrder=13, authorNames=YAN Tao, SHEN Shuilong, ZHOU Annan, journalName=Journal of Rock Mechanics and Geotechnical Engineering: English Edition, refType=null, unstructuredReference=YAN Tao, SHEN Shuilong, ZHOU Annan, et al. Prediction of geological characteristics from shield operational parameters by integrating grid search and K-fold cross validation into stacking classification algorithm[J]. Journal of Rock Mechanics and Geotechnical Engineering: English Edition, 2022, 14(4): 1292-1303., articleTitle=Prediction of geological characteristics from shield operational parameters by integrating grid search and K-fold cross validation into stacking classification algorithm, refAbstract=null), Reference(id=1165198351487213678, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2024, volume=71, issue=1, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[11], rfOrder=14, authorNames=XU Ying, journalName=Journal of Engineering and Applied Science, refType=null, unstructuredReference=XU Ying. Research on cooling load estimation through optimal hybrid models based on naive bayes[J]. Journal of Engineering and Applied Science, 2024, 71(1):DOI:10.1186/S44147-024-00396-9., articleTitle=Research on cooling load estimation through optimal hybrid models based on naive bayes, refAbstract=null), Reference(id=1165198351554322544, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2023, volume=16, issue=4, pageStart=390, pageEnd=398, url=null, language=null, rfNumber=[12], rfOrder=15, authorNames=LI Yanlong, YIN Qiaogang, ZHANG Ye, journalName=Water Science and Engineering, refType=null, unstructuredReference=LI Yanlong, YIN Qiaogang, ZHANG Ye, et al. Deformation prediction model of concrete face rockfill dams based on an improved random forest model[J]. Water Science and Engineering, 2023, 16(4): 390-398., articleTitle=Deformation prediction model of concrete face rockfill dams based on an improved random forest model, refAbstract=null), Reference(id=1165198351625625714, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2021, volume=38, issue=1, pageStart=51, pageEnd=56, url=null, language=null, rfNumber=[13], rfOrder=16, authorNames=NIE Xiaobo, LI Haibin, journalName=Journal of Donghua University: English Edition, refType=null, unstructuredReference=NIE Xiaobo, LI Haibin. Structural reliability analysis based on support vector machine and dual neural network direct integration method[J]. Journal of Donghua University: English Edition, 2021, 38(1): 51-56., articleTitle=Structural reliability analysis based on support vector machine and dual neural network direct integration method, refAbstract=null), Reference(id=1165198351688540276, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, doi=null, pmid=null, pmcid=null, year=2022, volume=16, issue=10, pageStart=1249, pageEnd=1266, url=null, language=null, rfNumber=[14], rfOrder=17, authorNames=SOHEILA K, CHENG Bin, JOSE T, journalName=Frontiers of Structural and Civil Engineering, refType=null, unstructuredReference=SOHEILA K, CHENG Bin, JOSE T. Structural performance assessment of GFRP elastic gridshells by machine learning interpretability methods[J]. Frontiers of Structural and Civil Engineering, 2022, 16(10): 1249-1266., articleTitle=Structural performance assessment of GFRP elastic gridshells by machine learning interpretability methods, refAbstract=null)], funds=[Fund(id=1165198350300225611, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, awardId=YDZJSX20231A021, language=CN, fundingSource=中央引导地方科技发展资金资助(YDZJSX20231A021), fundOrder=null, country=null), Fund(id=1165198350358945869, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, awardId=TUC2024-03, language=CN, fundingSource=隧道及地下工程教育部工程研究中心(北京交通大学)开放研究基金资助(TUC2024-03), fundOrder=null, country=null), Fund(id=1165198350413471823, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, awardId=2024-9006, language=CN, fundingSource=中央级公益性科研院所基本科研业务费项目(2024-9006), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1165198346194002896, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, xref=1, ext=[AuthorCompanyExt(id=1165198346202391505, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346194002896, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Civil Engineering,Taiyuan University of Technology,Taiyuan Shanxi 030024,China), AuthorCompanyExt(id=1165198346206585810, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346194002896, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 太原理工大学 土木工程学院,山西 太原 030024)]), AuthorCompany(id=1165198346269500372, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, xref=2, ext=[AuthorCompanyExt(id=1165198346277888981, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346269500372, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Research Center of Tunneling and Underground Engineering of Ministry of Education,Beijing Jiaotong University,Beijing 100044,China), AuthorCompanyExt(id=1165198346282083286, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346269500372, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 北京交通大学 隧道及地下工程教育部工程研究中心,北京 100044)]), AuthorCompany(id=1165198346344997848, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, xref=3, ext=[AuthorCompanyExt(id=1165198346349192153, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346344997848, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Research Institute of Highway Ministry of Transport,Beijing 100088,China), AuthorCompanyExt(id=1165198346353386458, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, companyId=1165198346344997848, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 交通运输部公路科学研究所,北京 100088)])], figs=[ArticleFig(id=1165198348144353295, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Fig.1, caption=Flow chart of stacking ensemble learning, figureFileSmall=pOASzyNz/bI/iXjB+uObNw==, figureFileBig=KAMRarEbycmkokawdmA4yw==, tableContent=null), ArticleFig(id=1165198348215656465, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=图1, caption=Stacking集成学习流程, figureFileSmall=pOASzyNz/bI/iXjB+uObNw==, figureFileBig=KAMRarEbycmkokawdmA4yw==, tableContent=null), ArticleFig(id=1165198348299542547, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Fig.2, caption=Distribution of samples under each hazard level, figureFileSmall=E0mph8Ir6m5BOfBXO6w6ow==, figureFileBig=Pl4zb7P9KsbnbnAwi5s41w==, tableContent=null), ArticleFig(id=1165198348349874197, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=图2, caption=各危险等级下的样本分布, figureFileSmall=E0mph8Ir6m5BOfBXO6w6ow==, figureFileBig=Pl4zb7P9KsbnbnAwi5s41w==, tableContent=null), ArticleFig(id=1165198348416983063, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Fig.3, caption=Water inrush hazard prediction process, figureFileSmall=r59QwhcEoOM7jgbm3W7N5A==, figureFileBig=36TD5rDV5XSiY69JnD93rQ==, tableContent=null), ArticleFig(id=1165198348479897625, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=图3, caption=突水危险预测流程, figureFileSmall=r59QwhcEoOM7jgbm3W7N5A==, figureFileBig=36TD5rDV5XSiY69JnD93rQ==, tableContent=null), ArticleFig(id=1165198348563783707, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Fig.4, caption=Evaluation metrics of each single model before hyper-parameters optimization, figureFileSmall=O4BMMbhADRjupYoBTBnJEg==, figureFileBig=sMoiR//4lT+tBfR4jOp7Wg==, tableContent=null), ArticleFig(id=1165198348630892573, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=图4, caption=超参数调优前单模型各评价指标, figureFileSmall=O4BMMbhADRjupYoBTBnJEg==, figureFileBig=sMoiR//4lT+tBfR4jOp7Wg==, tableContent=null), ArticleFig(id=1165198348714778656, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Fig.5, caption=Comparison of evaluation metrics of 8 groups of Stacking ensemble models, figureFileSmall=7FZhHgs+xOcwt/KPHWMM5w==, figureFileBig=uXtW473e9GEiw8GnRmDH3A==, tableContent=null), ArticleFig(id=1165198348777693217, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=图5, caption=8组Stacking集成模型评价指标对比, figureFileSmall=7FZhHgs+xOcwt/KPHWMM5w==, figureFileBig=uXtW473e9GEiw8GnRmDH3A==, tableContent=null), ArticleFig(id=1165198348861579300, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Fig.6, caption=Comparison of evaluation metrics of 6 groups of Stacking ensemble models before and after optimization, figureFileSmall=MDgO4AfxHD2OMbtBS12Giw==, figureFileBig=YEpkLzB58E3TwjERXhPFkw==, tableContent=null), ArticleFig(id=1165198348941271078, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=图6, caption=优化前后6组Stacking集成模型评价指标对比, figureFileSmall=MDgO4AfxHD2OMbtBS12Giw==, figureFileBig=YEpkLzB58E3TwjERXhPFkw==, tableContent=null), ArticleFig(id=1165198349050322985, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Fig.7, caption=Comparison of evaluation metrics for SVM and SNL, figureFileSmall=gQ+THbezxRHlId1LnjZJJw==, figureFileBig=o5oCAWeUg+lzdB+E/mJq0A==, tableContent=null), ArticleFig(id=1165198349134209068, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=图7, caption=SVM与SNL评价指标对比, figureFileSmall=gQ+THbezxRHlId1LnjZJJw==, figureFileBig=o5oCAWeUg+lzdB+E/mJq0A==, tableContent=null), ArticleFig(id=1165198349213900846, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Fig.8, caption=Confusion matrix comparison of SVM and SNL, figureFileSmall=0RSgVudaHxGTq4mAPc4B2w==, figureFileBig=breye+U2v0P19ROu04mHrA==, tableContent=null), ArticleFig(id=1165198349297786928, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=图8, caption=SVM与SNL的混淆矩阵对比, figureFileSmall=0RSgVudaHxGTq4mAPc4B2w==, figureFileBig=breye+U2v0P19ROu04mHrA==, tableContent=null), ArticleFig(id=1165198349364895795, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Fig.9, caption=ROC curve comparison of SVM and SNL, figureFileSmall=IHRxPzac2EhYAi9Ez8MeOQ==, figureFileBig=F0c6cQs+pWqZKSJUdzCMYQ==, tableContent=null), ArticleFig(id=1165198349427810358, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=图9, caption=SVM与SNL的ROC曲线对比, figureFileSmall=IHRxPzac2EhYAi9Ez8MeOQ==, figureFileBig=F0c6cQs+pWqZKSJUdzCMYQ==, tableContent=null), ArticleFig(id=1165198349494919225, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Table 1, caption=

Data set of tunnel water inrush

, figureFileSmall=null, figureFileBig=null, tableContent=
序号 预处理 地层岩性 不良地质 岩层倾角/
(°)
负地形
面积比
围岩级别 水动力
分带
危险等级
断面1 原始数据集 白云岩、页
岩、泥岩
F11断层主
断裂带
58~60 低中山区 V级 暗河管道水 IV级
数值化 3 4 59 50 4 4 4
标准化 0.67 1.00 0.66 0.53 1.00 1.00 4
断面2 原始数据集 灰岩 褶皱翼部 V级 III级
数值化 4 3 4 3
标准化 1.00 0.67 0.46 0.56 1.00 0.67 3
断面232 原始数据集 石英砂岩 背斜核部 60° II级 交替带 IV级
数值化 1 4 60 1 3 4
标准化 1.00 1.00 0.67 0.52 0.00 0.67 4
), ArticleFig(id=1165198349578805307, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=表1, caption=

隧道突水灾害数据集

, figureFileSmall=null, figureFileBig=null, tableContent=
序号 预处理 地层岩性 不良地质 岩层倾角/
(°)
负地形
面积比
围岩级别 水动力
分带
危险等级
断面1 原始数据集 白云岩、页
岩、泥岩
F11断层主
断裂带
58~60 低中山区 V级 暗河管道水 IV级
数值化 3 4 59 50 4 4 4
标准化 0.67 1.00 0.66 0.53 1.00 1.00 4
断面2 原始数据集 灰岩 褶皱翼部 V级 III级
数值化 4 3 4 3
标准化 1.00 0.67 0.46 0.56 1.00 0.67 3
断面232 原始数据集 石英砂岩 背斜核部 60° II级 交替带 IV级
数值化 1 4 60 1 3 4
标准化 1.00 1.00 0.67 0.52 0.00 0.67 4
), ArticleFig(id=1165198349675274300, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Table 2, caption=

Optimal hyper-parameters for three models

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 超参数选取范围 最佳超参数
RF max_depth:[1-10]
n_estimators:[1-10]
max_depth:[8]
n_estimators:[10]
SVM C:[0.1,1,10,100]
gamma:[0.001,0.01,0.1,1]
kernel:[linear,RBF,poly]
C:[1]
gamma:[1]
kernel:[RBF]
AB learning_rate:[0.01,0.1,1]
n_estimators:[50,100,150]
learning_rate:[0.1]
n_estimators:[50]
), ArticleFig(id=1165198349725605949, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=表2, caption=

3种模型的最佳超参数取值

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 超参数选取范围 最佳超参数
RF max_depth:[1-10]
n_estimators:[1-10]
max_depth:[8]
n_estimators:[10]
SVM C:[0.1,1,10,100]
gamma:[0.001,0.01,0.1,1]
kernel:[linear,RBF,poly]
C:[1]
gamma:[1]
kernel:[RBF]
AB learning_rate:[0.01,0.1,1]
n_estimators:[50,100,150]
learning_rate:[0.1]
n_estimators:[50]
), ArticleFig(id=1165198349780131902, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Table 3, caption=

Comparison of each evaluation metric of single model before and after optimization

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 RF SVM AB
优化前 P 0.81 0.91 0.73
R 0.81 0.89 0.70
F1 0.80 0.90 0.70
优化后 P 0.84↑ 0.83↓ 0.80↑
R 0.79↓ 0.81↓ 0.79↑
F1 0.80△ 0.81↓ 0.79↑
), ArticleFig(id=1165198349843046464, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=表3, caption=

优化前后单模型各评价指标对比

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 RF SVM AB
优化前 P 0.81 0.91 0.73
R 0.81 0.89 0.70
F1 0.80 0.90 0.70
优化后 P 0.84↑ 0.83↓ 0.80↑
R 0.79↓ 0.81↓ 0.79↑
F1 0.80△ 0.81↓ 0.79↑
), ArticleFig(id=1165198349914349634, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Table 4, caption=

Optimal hyper-parameters for two meta learners

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 超参数选取范围 最佳超参数
LR C:[ 0.1,1,10,100]
Penalty:[L1,L2]
Solver:[liblinear,saga]
C:[1]
Penalty:[L1]
Solver:[saga]
Ridge alpha:[0.01,0.1,
1.0,10.0]
alpha:[0.1]
), ArticleFig(id=1165198349989847108, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=表4, caption=

2种元学习器的最佳超参数取值

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 超参数选取范围 最佳超参数
LR C:[ 0.1,1,10,100]
Penalty:[L1,L2]
Solver:[liblinear,saga]
C:[1]
Penalty:[L1]
Solver:[saga]
Ridge alpha:[0.01,0.1,
1.0,10.0]
alpha:[0.1]
), ArticleFig(id=1165198350065344582, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=EN, label=Table 5, caption=

8 groups of Stacking ensemble models

, figureFileSmall=null, figureFileBig=null, tableContent=
组合 模型名称 基学习器 元学习器
1 RNL RF+NB LR
2 RNR Ridge
3 RSL RF+SVM LR
4 RSR Ridge
5 SNL SVM+NB LR
6 SNR Ridge
7 RNSL RF+NB+SVM LR
8 RNSR Ridge
), ArticleFig(id=1165198350132453448, tenantId=1146029695717560320, journalId=1146031787341344770, articleId=1148106715691737606, language=CN, label=表5, caption=

8组Stacking集成模型

, figureFileSmall=null, figureFileBig=null, tableContent=
组合 模型名称 基学习器 元学习器
1 RNL RF+NB LR
2 RNR Ridge
3 RSL RF+SVM LR
4 RSR Ridge
5 SNL SVM+NB LR
6 SNR Ridge
7 RNSL RF+NB+SVM LR
8 RNSR Ridge
)], attaches=null, journal=Journal(id=1123942128916217864, delFlag=0, nameCn=中国安全科学学报, nameEn=China Safety Science Journal, nameHistory1=null, nameHistory2=null, issn=1003-3033, eissn=, cn=11-2865/X, coden=null, periodic=0, language=CN, oaType=0, 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=fkqsFM6VKlHC4gCtS5XqTw==, journalPrice=null, startedYear=null, abbrevIsoEn=Chin Saf Sci J, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754269350027, createdBy=null, updatedBy=13701087609, firstLetterCn=C, firstLetterEn=C, subjectCode=Engineering, subjectName=工程, subjectCodeEn=Engineering, subjectNameEn=null, picCn=fkqsFM6VKlHC4gCtS5XqTw==, picEn=SHn9HgqSxtJrOcAxqD++4Q==, jcr=null, cjcr=null, exts=[JournalExt(id=1159052918994595848, language=CN, name=中国安全科学学报, 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.cssjj.com.cn/, createdTime=1754269350050, updatedTime=1754269350050, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.cssjj.com.cn/CN/column/item15.shtml, submissionAuthorUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, option={"copyright":""}), JournalExt(id=1159052919040733193, language=EN, name=China Safety Science Journal, 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.cssjj.com.cn/EN/1003-3033/home.shtml, createdTime=1754269350061, updatedTime=1754269350061, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=https://synbioj.cip.com.cn/EN/column/column3.shtml, submissionAuthorUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionReviewUrl=https://zgaqkxxbauthor.manuscriptcloud.com/login, submissionCeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, submissionAeEditorUrl=https://zgaqkxxbeditor.manuscriptcloud.com/login, option={"copyright":""})], databaseList=null, tenantJournalId=1146031787341344770, websiteList=[Website(id=1148243202345263519, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031787341344770, 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/zgaqkxxb/CN, language=CN, createTime=1751692112766, createBy=18614031015, updateTime=1753502583634, updateBy=18614031015, name=《中国安全科学学报》中文站点, tplId=1146099689490845704, title=中国安全科学学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148618794941046792, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=articleTextType, value=kx, createTime=1751781661020, updateTime=1751781661020, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794911686661, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=banner, value=null, createTime=1751781661012, updateTime=1751781661012, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794894909444, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=tui0IVO9FMwB61HHtX5scg==, createTime=1751781661008, updateTime=1751781661008, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794932658183, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751781661017, updateTime=1751781661017, creator=18614031015, updator=18614031015), WebsiteProps(id=1148618794924269574, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202345263519, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751781661015, updateTime=1751781661015, creator=18614031015, updator=18614031015)]), Website(id=1155836763751993353, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146031787341344770, 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/zgaqkxxb/EN, language=EN, createTime=1753502558893, createBy=18614031015, updateTime=1753524450387, updateBy=18614031015, name=《中国安全科学学报》英文站点, tplId=1146101810881728533, title=China Safety Science Journal, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155895925743669425, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=articleTextType, value=kx, createTime=1753516664205, updateTime=1753516664205, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925722697902, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=banner, value=null, createTime=1753516664200, updateTime=1753516664200, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925714309293, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=tui0IVO9FMwB61HHtX5scg==, createTime=1753516664198, updateTime=1753516664198, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925735280816, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753516664203, updateTime=1753516664203, creator=18614031015, updator=18614031015), WebsiteProps(id=1155895925731086511, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155836763751993353, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753516664202, updateTime=1753516664202, creator=18614031015, updator=18614031015)])], journalTitle=中国安全科学学报, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=China Safety Science Journal, journalPhotoCn=fkqsFM6VKlHC4gCtS5XqTw==, journalPhotoEn=SHn9HgqSxtJrOcAxqD++4Q==, journalFirstLetter=C, journalRecommend=null, journalNew=null, journalCollection=1, 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/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2025.04.1398, detailUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/10.16265/j.cnki.issn1003-3033.2025.04.1398, pdfUrlCn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/PDF/10.16265/j.cnki.issn1003-3033.2025.04.1398, pdfUrlEn=https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/PDF/10.16265/j.cnki.issn1003-3033.2025.04.1398, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于Stacking集成学习的隧道突水危险预测模型
收藏切换
PDF下载
卢佳乐 1, 2 , 张念 副教授 1, 2, ** , 牛萌萌 1 , 万飞 研究员 3
中国安全科学学报 | 安全工程技术 2025,35(4): 137-144
收起
收藏切换
中国安全科学学报 | 安全工程技术 2025, 35(4): 137-144
基于Stacking集成学习的隧道突水危险预测模型
全屏
卢佳乐1, 2 , 张念 副教授1, 2, ** , 牛萌萌1, 万飞 研究员3
作者信息
  • 1 太原理工大学 土木工程学院,山西 太原 030024
  • 2 北京交通大学 隧道及地下工程教育部工程研究中心,北京 100044
  • 3 交通运输部公路科学研究所,北京 100088
  • 卢佳乐 (1998—),男,山西运城人,硕士研究生,主要研究方向为隧道工程安全与算法优化。E-mail:

通讯作者:

**张 念(1984—),男,湖北襄阳人,博士,副教授,主要从事隧道及地下工程安全与算法优化方面的研究。E-mail:
Hazard prediction model of tunnel water inrush based on stacking ensemble learning
Jiale LU1, 2 , Nian ZHANG1, 2, ** , Mengmeng NIU1, Fei WAN3
Affiliations
  • 1 College of Civil Engineering,Taiyuan University of Technology,Taiyuan Shanxi 030024,China
  • 2 Research Center of Tunneling and Underground Engineering of Ministry of Education,Beijing Jiaotong University,Beijing 100044,China
  • 3 Research Institute of Highway Ministry of Transport,Beijing 100088,China
出版时间: 2025-04-28 doi: 10.16265/j.cnki.issn1003-3033.2025.04.1398
文章导航
收藏切换

为解决机器学习在隧道突水危险智能预测领域存在的模型较单一和预测精度不够理想等问题,提出一种基于Stacking集成学习方法的预测模型。首先,通过搜集95条隧道共计232组隧道突水灾害数据建立隧道突水灾害数据集,并进行数据预处理;然后,选取3种基学习器和2种元学习器以不同组合方式训练出8组Stacking集成模型,并筛选出6组较优的集成模型;最后,使用网格搜索调参并结合5折交叉验证超参数调优模型,对比分析6组参数调优后的Stacking集成模型的预测结果,选择出最优Stacking集成模型。结果表明:采用Stacking集成学习方法改进最优单模型支持向量机(SVM)后得到SVM+朴素贝叶斯(NB )+线性回归(LR)集成模型,其精确率、召回率和F1分数分别达到0.94、0.91和0.92,整体预测效果优于其他对比模型,可准确预测隧道突水危险等级。

Stacking集成学习  /  隧道突水  /  预测模型  /  危险等级  /  机器学习

In order to solve the problems that machine learning exists in the hazard intelligent prediction field of tunnel water inrush,such as relatively simple models and imperfect prediction accuracy,a prediction model based on the stacking ensemble learning was proposed. Firstly,the tunnel water inrush disaster dataset was established by collecting 232 groups of water inrush disaster data from 95 tunnels,and the data was preprocessed. Then,3 base learners and 2 meta learners were selected to train 8 sets of stacking ensemble models in different combinations,and 6 sets of optimal ensemble models were selected. Finally,the optimal stacking ensemble model was selected by comparing and analyzing the prediction results of 6 groups of parameters optimized and stacking ensemble model with the grid search parameters and the 5-fold cross-validation hyperparameter optimization model. The results show that SVM(Support Vector Machine )+NB (Naive Bayes) + LR (Linear Regression) ensemble model is obtained after the optimal single model SVM is improved with the stacking ensemble learning method. Its accuracy,recall,and F1 score are 0.94,0.91,and 0.92,respectively. The overall prediction effect is better than that of other compative models,and it can accurately predict the hazard level of tunnel water inrush.

Stacking ensemble learning  /  tunnel water inrush  /  prediction model  /  hazard level  /  machine learning
卢佳乐, 张念 副教授, 牛萌萌, 万飞 研究员. 基于Stacking集成学习的隧道突水危险预测模型. 中国安全科学学报, 2025 , 35 (4) : 137 -144 . DOI: 10.16265/j.cnki.issn1003-3033.2025.04.1398
Jiale LU, Nian ZHANG, Mengmeng NIU, Fei WAN. Hazard prediction model of tunnel water inrush based on stacking ensemble learning[J]. China Safety Science Journal, 2025 , 35 (4) : 137 -144 . DOI: 10.16265/j.cnki.issn1003-3033.2025.04.1398
隧道突水作为隧道施工阶段可能遭遇最严重的地质灾害之一,会造成严重的经济损失和人员伤亡。如果能在隧道设计阶段或施工早期阶段预测出突水灾害的危险性,对于后续采取相应的灾害防治措施具有重要的指导意义[1-2]
传统的隧道突水灾害预测方法,如层次分析法、综合评价法、专家评判法等,存在主观性强、适用性差和预测准确度低等问题,而机器学习法可避免上述问题。目前,诸多学者通过结合机器学习法智能预测隧道突水灾害,如王璐[3]构建遗传算法-支持向量机(Support Vector Machine,SVM)模型,开展隧道涌水量预测,取得了一定成效;马天行等[4]利用熵权法结合正态云模型预测煤层底板突水危险性,预测准确性有所提高;柏成浩[5]通过建立决策树、SVM和随机森林(Random Forest,RF)3种机器学习模型,并基于Python语言开发出风险智能预测平台,成功预测出特尔莫隧道的突水风险等级。现有文献在预测隧道突水灾害时,大部分只使用较为单一的传统机器学习模型,或简单模型的线性融合,导致模型整体预测精度不够理想。集成学习[6-7]作为一种能在各种机器学习任务上提高准确率的强有力技术,组合多个学习器,共同完成学习任务,能达到比单个学习器更好的适用性能。
鉴于此,笔者拟提出基于堆栈法(Stacking)集成学习方法的预测模型,通过选取合适的机器学习模型,运用Stacking集成学习算法,集成不同类型的学习器来预测隧道突水危险等级,以期提升隧道施工智能化管控水平。
Stacking算法最早由WOLPERT[8]于1992年提出,其使用不同类别的学习器进行第一轮学习,将得到的预测结果输入到第二轮学习器中再进行学习,得到最终预测结果。Stacking集成学习流程如图1所示。
对于隧道突水危险预测而言,Stacking算法的步骤如下:
1) 输入隧道突水灾害数据集 D = X i Y i i = 1,2 n,其中, X i为第 i个样本的特征向量, Y i为第 i个样本对应的标签值。将数据集 D平均分成 K个子集 D 1 D 2 D k。一般情况下,默认 K = 5
2) 选择一种基学习器,将数据集 D K - 1份作为训练集,剩下的1份作为预测集。利用 D 2 D 3 D 4 D 5中样本建立模型,分别预测预测集 D 1和测试集 T 1,得到预测结果1和测试结果1,将预测结果1作为第二层元学习器的新训练集 T i * ( i = 1,2 3 ),以此类推得到5组新训练集 D i * ( i = 1,2 5 )。将得到的5组测试结果 T i ( i = 1,2 5 )取平均值作为第二层元学习器的新测试集 T i * ( i = 1,2 3 )
3) 选择其他基学习器,重复步骤2),分别得到其预测结果,将新训练集 D i * ( i = 1,2 5 )和新测试集 T i * ( i = 1,2 3 )输入到第二层元学习器中进行训练,得到的预测结果作为最终预测结果。
4) Stacking集成学习模型训练完毕,最终预测结果即为隧道突水危险预测结果,实现综合学习器的学习能力。
通过中国知网和科学网,以及Google学术、各省应急管理厅公布的新闻事故报告、百度文库和百度学术等,统计与收集隧道突水灾害相关文献以及案例,得到原始数据。主要记录已发生突水灾害的隧道案例,数据集文件类型为.xlsx,共计95条隧道,232组断面数据。隧道突水灾害变量信息包括隧道概况、水文地质条件和施工扰动因素3个方面,涵盖隧道突水灾害发生的孕灾环境和致灾因子,对于后续选取特征变量以及模型的计算和优化有重要的指导意义。数据集中各危险等级下的样本分布情况如图2所示。
图2可知:数据集中各危险等级(I、II、III、IV)下的样本个数依次为17、17、52和146,总计232个。其中,危险等级IV下收集到的断面数据最多,说明大多数隧道突水危险程度极高,极易造成严重损失。危险等级I和II下的样本量与危险等级III和IV下的样本量差异较大,为保证后续模型训练以及优化的准确性,对数据集的预处理方法和模型选取及优化方法合理选用。
预处理数据集,主要包括数值化处理、离群值检测与替换、缺失数据补充以及数据集的标准化等,具体操作过程及内容如下:
1) 统计分析选取特征变量。隧道突水的危险性主要由隧道所处的水文地质条件决定。通过统计数据集中各致灾因子的出现频率和总结学者们研究成果[57],得出地质条件中地层岩性、不良地质、岩层倾角和围岩级别影响较大;水文条件中,负地形面积比和水动力分带影响较大。因此,决定选取6个致灾因子作为隧道突水危险预测模型所输入的特征变量。
2) 采用文献[9]中的隧道突水灾害相关分级标准数值化处理数据集。数据集中的定性变量与定量变量,其数值化过程按照突水危险致灾因子划分标准所对应的数值取值即可。其中,定性变量包括地层岩性、不良地质、水动力分带和围岩级别;定量变量包括岩层倾角和负地形面积比。
3) 采用箱线图法检测数据集的离群值。对于同一变量在不同危险等级下的离群值替换过程,定性变量取其当前危险等级下的数据众数作为替换值;定量变量取当前危险等级下数据平均值作为替换值。
4) 对于数据集中缺失数据的补充方法,采用同步骤3)所述操作。
5) 数据集的标准化采用最大最小归一化公式[5]进行处理。
预处理后,得到标准化数据集,见表1,由表1可知:对于断面2,缺失变量为岩层倾角、负地形面积比和水动力分带,依据数据预处理方法,分别取其当前危险等级下的平均值、众数和众数进行补充,经标准化处理后分别为0.46、0.56和0.67;对于断面232,缺失变量为负地形面积比,依据上述数据预处理方法,取其当前危险等级下的众数进行补充,经标准化处理后分别为0.52和0.52。对于未列出的其他断面,采取上述方法完成原始数据集的预处理工作,由此得到完整的标准化后的隧道突水灾害数据集。
隧道突水危险预测在机器学习中属于分类问题,拟用的评价指标[10]有精确率P、召回率RF1分数。
选用4种不同类型的机器学习模型,分别为朴素贝叶斯[11](Naive Bayes,NB)、RF[12]、SVM [13]和自适应算法[14](Adaptive Boosting,AB)。隧道突水危险预测流程如图3所示。
为提高Stacking集成模型的分类预测效果,在进行Stacking模型组合预测隧道突水危险等级前,先训练选取的单模型,进行超参数调优,并对比参数调优后的单模型的分类预测效果。
将数据集按8∶2比例划分为训练集和测试集,得到185份训练集和47份测试集。输入数据,训练NB、RF、SVM和AB这4种模型,得到隧道突水危险预测结果如图4所示。
图4可知:超参数调优前各单模型的PRF1值差异较大。SVM各值较其他3种单模型最高;AB模型各值最低,在未进行超参数调优的情况下表现较差;NB模型和RF模型预测性能相近,但NB模型的预测精确性稍好。综上所述,在未进行超参数调优的情况下,SVM在此次预测的所有单模型中表现最好。
4种模型中,NB模型不需要进行模型训练,也无需调整超参数,其余3种模型均采用网格搜索结合K折交叉验证(K=5)进行超参数调优。RF、SVM和AB这3种模型的最佳超参数取值见表2。其中,对于RF模型,max_depth为决策树的最大深度,控制着树的复杂程度;n_estimators代表决策树的数量。对于SVM模型,C为惩罚系数,表示模型对误差的容忍程度。gamma代表核函数的宽度,决定数据点之间影响程度的衰减速度;kernel表示核函数,通常用于处理非线性数据拟合;RBF代表径向基函数。对于AB模型,learning_rate为学习率,表示模型更新过程所采用的梯度下降或提升的方向和幅度。
输入3种模型的最佳超参数并重新训练模型,得到各单模型优化后(除NB外)的PRF1值。表3为超参数调优前后的单模型各评价指标对比。
表3可知:经超参数调优,RF模型精确性提高,全面性下降,F1值不变,可知RF模型经超参数调优后性能无明显变化;SVM模型总体性能下降,推测原因为数据量较少,模型存在过拟合风险;AB模型性能有所提高,但其各值仍较其他模型低。综上所述,此次隧道突水危险预测单模型优化前后模型性能表现最好的为超参数优化前的SVM模型。
由上述分析可知:超参数优化前的SVM模型在4种单模型中表现最好,AB模型调优前后的各评价指标值均较低,其精确性和全面性较差,可能不适用于此次预测,故排除。因此,选取NB、RF和SVM模型作为此次预测的Stacking集成模型的第一层基学习器。
为降低模型的过拟合风险,第二层的元学习器选择较为简单的模型。线性回归(Linear Regression,LR)可配合L1、L2正则化进一步防止模型过拟合,选为第一种元学习器;另选择泛化能力较好,可配合L2正则化的岭回归作为第2种元学习器加以对比。同样,为得到学习能力较好的元学习器,对LR和岭回归(Ridge)采取超参数调优方法,2种元学习器的最佳超参数取值见表4。其中,对于LR模型,Penalty为正则化项,用于控制模型的复杂度;Solver代表所用的优化策略。对于Ridge模型,alpha为正则化参数。
设计试验观测不同组合方式下的Stacking集成模型的分类预测效果。以不同组合方式对比3种基学习器和2种元学习器,共计8组集成模型组合。选取的8组Stacking集成模型组合及命名见表5
将数据集按8∶2比例划分为训练集和测试集。输入数据,训练8组Stacking集成模型,得到隧道突水危险预测结果,如图5所示。
图5可知:不同组合方式下的Stacking集成模型的预测效果差异显著。其中,SNL模型各值较其他集成模型得分最高,表明SNL模型预测更精确、误判漏判更少;RSR模型各值较其他集成模型最低;RNL和RNR模型的各值较NB和RF单模型变化不大;RNSL模型各值较最优单模型SVM的低,预测效果不够理想;RNSR模型各值较单模型和其他集成模型最低。综上所述,在未进行超参数优化的条件下,SNL模型为此次预测的所有集成模型中的较优模型。且Stacking集成模型中,第一层的基学习器模型组合并非复杂程度越高,预测效果就越好,基学习器和元学习器的正确选择与组合才是提升模型性能的关键。
RSR和RNSR模型在超参数调优前的模型性能较其他6组集成模型差距大,决定舍弃。仅对优化前集成模型的PRF1值均在0.80以上的6组集成模型输入最佳超参数,并进行模型训练,得到超参数优化前后6组Stacking集成模型的评价指标对比,如图6所示。
图6可知:经超参数调优后,SNL模型各值无明显提高,但其值依旧较其他6组集成模型最高;RNL和RSL模型整体提升较小;RNR模型经优化后略有下降。此外,SNR和RNSL模型优化前后各值未发生改变,表明模型已接近最好表现。综上所述,在超参数优化前后的所有集成模型中,SNL模型为此次隧道突水危险预测模型中的最优Stacking集成模型。
由对比分析可知:此次预测中最优单模型为SVM,最优Stacking集成模型为SNL。为得到最优隧道突水危险预测模型,还需要更深入地对比2个模型,主要包括评价指标、混淆矩阵、被试工作特征曲线(Receiver Operating Characteristic Curve,ROC)。
图7为SVM和SNL模型的评价指标对比,由图7可知:SVM作为单模型中最优模型,但其各值与SNL集成模型相比,SNL模型的值均更高,表明在此次隧道突水危险预测的所有样本中,SNL集成模型对于样本的预测精确性更高。
图8为2个模型的混淆矩阵对比,由图8可知:在测试集样本总数为47份的条件下,SVM与SNL模型的预测正确个数分别为42和43,预测准确率分别为89.36%和91.49%。SNL集成模型在危险等级I、II和III的样本下表现最好,全部预测正确;但在危险等级为IV下与SVM模型相比,多将1份真实危险等级为IV的样本预测为危险等级II,防治措施可能不足。导致这类预测错误的主要原因是数据集样本量欠缺,且样本分布不均衡。期待在数据集更完善、模型更优化的情况下改善此类预测错误。
图9为2个模型的ROC曲线对比,由图9可知:2个模型在此次预测中性能均较好,但SNL集成模型的性能更好。其中,SNL集成模型在危险等级I、III和IV下的预测效果明显优于SVM模型,且在危险等级I下的曲线下面积(Area Under the Curve,AUC)值达到1.000,即第一临界点,完美分类,样本无误判无漏判;而SVM模型仅在危险等级II下的AUC值高出SNL组合模型0.006,差距不大。此外,SVM模型过拟合风险较SNL集成模型更高,第二临界点数有4个,分别为危险等级II下1个,危险等级IV下3个;而SNL集成模型仅在危险等级II下存在1个第二临界点,模型过拟合风险远小于SVM模型。
综上所述,SNL集成模型综合性能优于SVM模型,模型优化更合理,预测更准确。可知:以SVM和NB作为基学习器,LR作为元学习器的Stacking集成模型SNL是此次研究中的最优预测模型,同时,说明采用Stacking集成学习方法改进最优单模型SVM并进行预测是切实可行的。
1) 确定了4个机器学习模型的6个输入变量,分别为地层岩性、不良地质、岩层倾角、负地形面积比、围岩级别和水动力分带;预处理这些数据,并训练4组单模型和8组Stacking集成模型。
2) 筛选得到预测最优单模型SVM和最优Stacking集成模型SNL,并进行对比分析,SNL集成模型是此次隧道突水危险预测的最优模型。
3) 基于Stacking集成学习的SNL模型在预测隧道突水危险等级时,综合性能较其他模型最好,模型更加合理,对于高危险等级的隧道突水灾害适用性更强。采用Stacking集成学习方法对最优单模型SVM的改进是成功的,可实现准确预测隧道突水危险等级。
4) 鉴于模型训练样本较少,且样本分布不够均衡,导致模型在低等级样本下的预测情况较差。后续工作如果能收集到更多更全面的隧道突水信息,并结合Python语言开发出隧道突水智能预测系统,则可进一步提升隧道施工智能化管控水平。
  • 中央引导地方科技发展资金资助(YDZJSX20231A021)
  • 隧道及地下工程教育部工程研究中心(北京交通大学)开放研究基金资助(TUC2024-03)
  • 中央级公益性科研院所基本科研业务费项目(2024-9006)
参考文献 引证文献
排序方式:
[1]
白明洲, 许兆义, 王连俊, 等. 复杂岩溶地区隧道施工突水地质灾害研究[J]. 中国安全科学学报, 2006, 16(1):114-118.
BAI Mingzhou, XU Zhaoyi, WANG Lianjun, et al. Study on water outburst geological disaster of tunnel construction at complicated karst zone[J]. China Safety Science Journal, 2006, 16(1): 114-118.
[2]
XUE Yiguo, KONG Fanmeng, LI Shucai, et al. Water and mud inrush hazard in underground engineering: genesis, evolution and prevention[J]. Tunnelling and Underground Space Technology, 2021, 114: DOI: 10.1016/J.TUST.2021.103987.
[3]
王璐. 基于遗传算法和支持向量机的西南岩溶越岭隧道涌水量预测[D]. 成都: 成都理工大学, 2019.
WANG Lu. Prediction of water inflow in southwest karst crossing-mountain tunnel based on genetic algorithm and support vector machine[D]. Chengdu: Chengdu University of Technology, 2019.
[4]
马天行, 林允, 周晓斌, 等. 煤层底板突水危险性预测的熵权-正态云模型[J]. 中国安全科学学报, 2022, 32(增1):171-177.
MA Tianxing, LIN Yun, ZHOU Xiaobin, et al. Entropy weight-normal cloud model for water inrush risk prediction of coal seam floor[J]. China Safety Science Journal, 2022, 32(S1): 171-177.
[5]
柏成浩. 基于机器学习的岩溶隧道突水突泥灾害风险智能预测方法研究[D]. 济南: 山东大学, 2021.
BO Chenghao. Research on intelligent prediction method of hazard risk of water and mud inrush in karst tunnel based on machine learning[D]. Jinan: Shandong University, 2021.
[6]
周志华. 机器学习[M]. 北京: 清华大学出版社, 2016:171-190.
[7]
GUO Wangda, ZHANG Jinxi, MURTAZA M, et al. An ensemble learning with sequential model-based optimization approach for pavement roughness estimation using smartphone sensor data[J]. Construction and Building Materials, 2023, 406: DOI:10.1016/J.CONBUILDMAT.2023.133293.
[8]
WOLPERT D H. Stacked generalization[J]. Neural Networks, 1992, 5(2): 241-259.
[9]
ZHANG Nian, NIU Mengmeng, WAN Fei, et al. Hazard prediction of water inrush in water-rich tunnels based on random forest algorithm[J]. Applied Sciences, 2024, 14(2):DOI:10.3390/APP14020867.
[10]
YAN Tao, SHEN Shuilong, ZHOU Annan, et al. Prediction of geological characteristics from shield operational parameters by integrating grid search and K-fold cross validation into stacking classification algorithm[J]. Journal of Rock Mechanics and Geotechnical Engineering: English Edition, 2022, 14(4): 1292-1303.
[11]
XU Ying. Research on cooling load estimation through optimal hybrid models based on naive bayes[J]. Journal of Engineering and Applied Science, 2024, 71(1):DOI:10.1186/S44147-024-00396-9.
[12]
LI Yanlong, YIN Qiaogang, ZHANG Ye, et al. Deformation prediction model of concrete face rockfill dams based on an improved random forest model[J]. Water Science and Engineering, 2023, 16(4): 390-398.
[13]
NIE Xiaobo, LI Haibin. Structural reliability analysis based on support vector machine and dual neural network direct integration method[J]. Journal of Donghua University: English Edition, 2021, 38(1): 51-56.
[14]
SOHEILA K, CHENG Bin, JOSE T. Structural performance assessment of GFRP elastic gridshells by machine learning interpretability methods[J]. Frontiers of Structural and Civil Engineering, 2022, 16(10): 1249-1266.
2025年第35卷第4期
PDF下载
475
199
引用本文
BibTeX
文章信息
doi: 10.16265/j.cnki.issn1003-3033.2025.04.1398
  • 接收时间:2024-11-15
  • 首发时间:2025-07-05
  • 出版时间:2025-04-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-11-15
  • 修回日期:2025-02-14
基金
中央引导地方科技发展资金资助(YDZJSX20231A021)
隧道及地下工程教育部工程研究中心(北京交通大学)开放研究基金资助(TUC2024-03)
中央级公益性科研院所基本科研业务费项目(2024-9006)
作者信息
    1 太原理工大学 土木工程学院,山西 太原 030024
    2 北京交通大学 隧道及地下工程教育部工程研究中心,北京 100044
    3 交通运输部公路科学研究所,北京 100088

通讯作者:

**张 念(1984—),男,湖北襄阳人,博士,副教授,主要从事隧道及地下工程安全与算法优化方面的研究。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/zgaqkxxb/CN/10.16265/j.cnki.issn1003-3033.2025.04.1398
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