基于机器学习的钢-UHPC组合结构栓钉抗剪承载力研究

TU398.9; 组合结构良好的界面抗剪性能是充分发挥钢和超高性能混凝土(UHPC)材料性能的关键.为研究钢-UHPC组合结构栓钉剪力连接件抗剪性能,建立了UHPC中包含263个栓钉的推出试验数据库,进行了特征相关性分析,并分析了主要参数影响规律.基于机器学习算法训练并生成了10个栓钉抗剪承载力预测模型,通过8次随机抽样训练验证模型稳定性.采用确定系数(R2)、平均平方对数误差(RMSE)、平均绝对误差(MAE)等指标来评估预测精度,并与3种传统计算方法进行了比较.结果表明,UHPC抗压强度小于150 MPa的试验样本占比高达64.6%,而栓钉直径小于19 mm的试验样本占比达到73%,缺乏更...

Full description

Saved in:
Bibliographic Details
Published in东南大学学报（自然科学版） Vol. 55; no. 1; pp. 146 - 153
Main Authors 戚家南, 杜雨轩, 李立坤, 韩昀芝, 李旻轩, 衣忠强, 邹伟豪
Format Journal Article
LanguageChinese
Published 东南大学长大桥梁安全长寿与健康运维全国重点实验室,南京 211189 2025
东南大学桥梁研究中心,南京 211189%东南大学混凝土及预应力混凝土结构教育部重点实验室,南京 211189%中铁九局集团第一建设有限公司,苏州 215538
东南大学混凝土及预应力混凝土结构教育部重点实验室,南京 211189
Subjects
抗剪承载力
ultra-high performance concrete
机器学习
超高性能混凝土
stud
shear capacity
composite structures
栓钉
machine learning
组合结构
Online AccessGet full text
ISSN1001-0505
DOI10.3969/j.issn.1001-0505.2025.01.017

Cover

Abstract TU398.9; 组合结构良好的界面抗剪性能是充分发挥钢和超高性能混凝土(UHPC)材料性能的关键.为研究钢-UHPC组合结构栓钉剪力连接件抗剪性能,建立了UHPC中包含263个栓钉的推出试验数据库,进行了特征相关性分析,并分析了主要参数影响规律.基于机器学习算法训练并生成了10个栓钉抗剪承载力预测模型,通过8次随机抽样训练验证模型稳定性.采用确定系数(R2)、平均平方对数误差(RMSE)、平均绝对误差(MAE)等指标来评估预测精度,并与3种传统计算方法进行了比较.结果表明,UHPC抗压强度小于150 MPa的试验样本占比高达64.6%,而栓钉直径小于19 mm的试验样本占比达到73%,缺乏更高强度UHPC和大直径栓钉试验数据.传统公式预测精度低于机器学习模型预测精度.随机森林模型预测精度最高,R2 为0.98,MAE与RMSE分别为3.9和6.4;与传统模型最高精度相比,R2 提高了216%,MAE与RMSE分别降低了93%与90%,表明机器学习模型可大幅提升预测精度,为栓钉抗剪承载力计算提供了新的思路.
AbstractList TU398.9; 组合结构良好的界面抗剪性能是充分发挥钢和超高性能混凝土(UHPC)材料性能的关键.为研究钢-UHPC组合结构栓钉剪力连接件抗剪性能,建立了UHPC中包含263个栓钉的推出试验数据库,进行了特征相关性分析,并分析了主要参数影响规律.基于机器学习算法训练并生成了10个栓钉抗剪承载力预测模型,通过8次随机抽样训练验证模型稳定性.采用确定系数(R2)、平均平方对数误差(RMSE)、平均绝对误差(MAE)等指标来评估预测精度,并与3种传统计算方法进行了比较.结果表明,UHPC抗压强度小于150 MPa的试验样本占比高达64.6%,而栓钉直径小于19 mm的试验样本占比达到73%,缺乏更高强度UHPC和大直径栓钉试验数据.传统公式预测精度低于机器学习模型预测精度.随机森林模型预测精度最高,R2 为0.98,MAE与RMSE分别为3.9和6.4;与传统模型最高精度相比,R2 提高了216%,MAE与RMSE分别降低了93%与90%,表明机器学习模型可大幅提升预测精度,为栓钉抗剪承载力计算提供了新的思路.
Abstract_FL Good interfacial shear resistance of composite structure is the key to making full use of the proper-ties of steel and ultra-high performance concrete(UHPC).To study the shear resistance of the stud shear con-nection of steel-UHPC composite structure,the push-out test database of 263 studs in UHPC was established,the characteristic correlation analysis was conducted,and the influence law of main parameters was analyzed.Ten prediction models of the shear capacity of screws were trained and generated based on the machine learn-ing algorithm,and the stability of the models was verified by 8 times random sampling training.The indica-tors such as the determination coefficient(R2),mean square error log(RMSE),and mean absolute error(MAE)were employed to evaluate the prediction precision,and were compared with those of the three tradi-tional computing methods.The results show that the proportion of test samples with compressive strength less than 150 MPa is as high as 64.6%,while the proportion of test samples with screw diameter less than 19 mm is 73%.There is a lack of test data for UHPC with higher strength and large diameter screws.The prediction accuracy of traditional formulas is lower than that of machine learning models.The prediction accuracy of ran-dom forests model is the highest,R2 is 0.98,MAE and RMSE are 3.9 and 6.4,respectively.Compared with the highest accuracy of the traditional model,R2 is increased by 216%,MAE and RMSE are reduced by 93%and 90%,respectively,indicating that the machine learning model can greatly improve the prediction accuracy and provide a new idea for the calculation of shear capacity of stud.
Author 戚家南
韩昀芝
李立坤
衣忠强
杜雨轩
李旻轩
邹伟豪
AuthorAffiliation 东南大学混凝土及预应力混凝土结构教育部重点实验室,南京 211189;东南大学长大桥梁安全长寿与健康运维全国重点实验室,南京 211189;东南大学桥梁研究中心,南京 211189%东南大学混凝土及预应力混凝土结构教育部重点实验室,南京 211189%中铁九局集团第一建设有限公司,苏州 215538
AuthorAffiliation_xml – name: 东南大学混凝土及预应力混凝土结构教育部重点实验室,南京 211189;东南大学长大桥梁安全长寿与健康运维全国重点实验室,南京 211189;东南大学桥梁研究中心,南京 211189%东南大学混凝土及预应力混凝土结构教育部重点实验室,南京 211189%中铁九局集团第一建设有限公司,苏州 215538
Author_FL LI Minxuan
QI Jianan
DU Yuxuan
HAN Yunzhi
ZOU Weihao
LI Likun
YI Zhongqiang
Author_FL_xml – sequence: 1
  fullname: QI Jianan
– sequence: 2
  fullname: DU Yuxuan
– sequence: 3
  fullname: LI Likun
– sequence: 4
  fullname: HAN Yunzhi
– sequence: 5
  fullname: LI Minxuan
– sequence: 6
  fullname: YI Zhongqiang
– sequence: 7
  fullname: ZOU Weihao
Author_xml – sequence: 1
  fullname: 戚家南
– sequence: 2
  fullname: 杜雨轩
– sequence: 3
  fullname: 李立坤
– sequence: 4
  fullname: 韩昀芝
– sequence: 5
  fullname: 李旻轩
– sequence: 6
  fullname: 衣忠强
– sequence: 7
  fullname: 邹伟豪
BookMark eNo9j8tKw0AYhWdRwVr7FoKrxP-fXCZZSrxUKOjCrssknZEWmYJB7ANkYboQaQviBQ0uRAW3ginFl2kS-xZGFOHAgY_DOZwVUlF9JQhZQ9AN13Y3eno3DJWOAKiBBZZOgVo6YClWIdV_vkzqYdj1ASl1gVK7Srayh3SeXuR3aXb9nL09zT-S4iZajB61VuPAK6ZRdnleTMf5fZQn48UozodXWfyax59fs1k2vC2SSfHyvkqWJD8ORf3Pa6S1s33oNbTm_u6et9nUQgTKNAoONwWYHJkDAn2JlqAlMaSPwBiXgYssYI5lmdKUlKNh22VW-EI4jhG4Ro2s__aecSW5Omr3-qcnqlxsd1RnMPB_XgOWn41vUjNmxA
ClassificationCodes TU398.9
ContentType Journal Article
Copyright Copyright © Wanfang Data Co. Ltd. All Rights Reserved.
Copyright_xml – notice: Copyright © Wanfang Data Co. Ltd. All Rights Reserved.
DBID 2B.
4A8
92I
93N
PSX
TCJ
DOI 10.3969/j.issn.1001-0505.2025.01.017
DatabaseName Wanfang Data Journals - Hong Kong
WANFANG Data Centre
Wanfang Data Journals
万方数据期刊 - 香港版
China Online Journals (COJ)
China Online Journals (COJ)
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
DocumentTitle_FL Research on the shear capacity of stud in steel-UHPC composite structure based on machine learning
EndPage 153
ExternalDocumentID dndxxb202501017
GrantInformation_xml – fundername: (国家自然科学基金); (中铁九局科技研发资助项目)
  funderid: (国家自然科学基金); (中铁九局科技研发资助项目)
GroupedDBID 2B.
4A8
92I
93N
ADMLS
ALMA_UNASSIGNED_HOLDINGS
PSX
TCJ
ID FETCH-LOGICAL-s1027-208a4e04a1780e1bf15e2a4e3fb1077afc917c78554f4f2a1366a17ebee883c93
ISSN 1001-0505
IngestDate Thu May 29 04:08:38 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 1
Keywords 抗剪承载力
ultra-high performance concrete
机器学习
超高性能混凝土
stud
shear capacity
composite structures
栓钉
machine learning
组合结构
Language Chinese
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-s1027-208a4e04a1780e1bf15e2a4e3fb1077afc917c78554f4f2a1366a17ebee883c93
PageCount 8
ParticipantIDs wanfang_journals_dndxxb202501017
PublicationCentury 2000
PublicationDate 2025
PublicationDateYYYYMMDD 2025-01-01
PublicationDate_xml – year: 2025
  text: 2025
PublicationDecade 2020
PublicationTitle 东南大学学报（自然科学版）
PublicationTitle_FL Journal of Southeast University(Natural Science Edition)
PublicationYear 2025
Publisher 东南大学长大桥梁安全长寿与健康运维全国重点实验室,南京 211189
东南大学桥梁研究中心,南京 211189%东南大学混凝土及预应力混凝土结构教育部重点实验室,南京 211189%中铁九局集团第一建设有限公司,苏州 215538
东南大学混凝土及预应力混凝土结构教育部重点实验室,南京 211189
Publisher_xml – name: 东南大学长大桥梁安全长寿与健康运维全国重点实验室,南京 211189
– name: 东南大学混凝土及预应力混凝土结构教育部重点实验室,南京 211189
– name: 东南大学桥梁研究中心,南京 211189%东南大学混凝土及预应力混凝土结构教育部重点实验室,南京 211189%中铁九局集团第一建设有限公司,苏州 215538
SSID ssib012290226
ssib002258162
ssib036435511
ssib008679709
ssib023167012
ssib000947520
ssib021009659
ssib057620145
ssib000969306
ssib001128997
ssib006563446
ssib002039847
ssib006703054
ssib051368071
ssib004675274
Score 2.415023
Snippet TU398.9;...
SourceID wanfang
SourceType Aggregation Database
StartPage 146
Title 基于机器学习的钢-UHPC组合结构栓钉抗剪承载力研究
URI https://d.wanfangdata.com.cn/periodical/dndxxb202501017
Volume 55
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnR3LahRBcIgbEC-iqPgmh_RJNk7PTL-OPbuzBDUimEBuYWZ2Vk8bMAmEnHMwOYgkgvhAgwdRwatgQvBnsrvmL6yqGTftIhqFMHSqq6rr0ds11XTXeN6kaGcdPwzSukyDvB4VbVPPuBT1XLWlyUQqFMcLzjN35fRcdGtezI_VbjunllaWs6l87bf3Sv7HqwADv-It2X_w7JApAKAN_oUneBiex_IxSwQzLRZblkT41AlLJDMNgkCXYVZjwzaZlYQDEJ8lihlAjlhimAmYDepz0_caCI5jAgOpz7SuICYkrgl1SWSAECLVBiHaMqOQCv61liCGxS2WaBY38Q-7ACdGhkgeUQNwpPtuTAJqFB_xmxVPGzGrHCWGDRrXCpaAARokrWZakQAKRY0ljQK6codKoWyIXFINt0GIn0a7IG5C1JUULoppknwGlbG60tCaERT0AwwNxozJmIASHaEY9JklyxmNJ01QcDBP092BKe9p06_leHYBroKMPuwCV3EyEDTAVbzSDH0GfATJ71K1qEFjoQICSZBc4HSKFSnbwplRTos4cvgINAh62qAPQEg0dIDzrRwUZk9Ck0M3KggIGbga0QS2jRsBhMXKLWWspNN4whduMC1rLv-yaJSR8edOc_mSxcsK0aPxOzTSUPzGAaaGA0yh1am6bnnPd6RCervbXl3NEAcLJqoT3nigFBc1b9w2Z-7cd7KJSAk3OzL4HVC3eiLuPji7F35o3EMEEPo0l241PGDnZE-QGoXu7oqkaHrUj7UtnWqJHD-14GQ_AUeJjrKLAItUONlICMmDcKoJCh5K7WRzAt5r8JAAne-oLHfSm6zMevNPRqUbi91O2n3gvFzPnvFOV1nxhC2XuLPe2NrDc16z93b3YPdJ__Vu78WH3uf3B193Bi_XD7fe0XI12FvvPX082Nvuv1nv72wfbm30N5_3Nj71N75939_vbb4a7DwbfPxy3ptrJbON6Xr10Zf6EuQ6CtZ3nUaFH6Vcab_gWYeLIgBI2Mm4r1TayQ1XucLTtZ2oE6RgBAm4EIsKrcPchBe8WnexW1z0JlIByPAITCQiXvg6zwoIPpnKsiLLZXTJm6i0XqgW9aWFkbl0-e8oV7xT2C63Za96teVHK8U1SFSWs-vVBPwBMCTjPQ
linkProvider EBSCOhost
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=%E5%9F%BA%E4%BA%8E%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E7%9A%84%E9%92%A2-UHPC%E7%BB%84%E5%90%88%E7%BB%93%E6%9E%84%E6%A0%93%E9%92%89%E6%8A%97%E5%89%AA%E6%89%BF%E8%BD%BD%E5%8A%9B%E7%A0%94%E7%A9%B6&rft.jtitle=%E4%B8%9C%E5%8D%97%E5%A4%A7%E5%AD%A6%E5%AD%A6%E6%8A%A5%EF%BC%88%E8%87%AA%E7%84%B6%E7%A7%91%E5%AD%A6%E7%89%88%EF%BC%89&rft.au=%E6%88%9A%E5%AE%B6%E5%8D%97&rft.au=%E6%9D%9C%E9%9B%A8%E8%BD%A9&rft.au=%E6%9D%8E%E7%AB%8B%E5%9D%A4&rft.au=%E9%9F%A9%E6%98%80%E8%8A%9D&rft.date=2025&rft.pub=%E4%B8%9C%E5%8D%97%E5%A4%A7%E5%AD%A6%E9%95%BF%E5%A4%A7%E6%A1%A5%E6%A2%81%E5%AE%89%E5%85%A8%E9%95%BF%E5%AF%BF%E4%B8%8E%E5%81%A5%E5%BA%B7%E8%BF%90%E7%BB%B4%E5%85%A8%E5%9B%BD%E9%87%8D%E7%82%B9%E5%AE%9E%E9%AA%8C%E5%AE%A4%2C%E5%8D%97%E4%BA%AC+211189&rft.issn=1001-0505&rft.volume=55&rft.issue=1&rft.spage=146&rft.epage=153&rft_id=info:doi/10.3969%2Fj.issn.1001-0505.2025.01.017&rft.externalDocID=dndxxb202501017
thumbnail_s http://utb.summon.serialssolutions.com/2.0.0/image/custom?url=http%3A%2F%2Fwww.wanfangdata.com.cn%2Fimages%2FPeriodicalImages%2Fdndxxb%2Fdndxxb.jpg