中国安全科学学报 ›› 2019, Vol. 29 ›› Issue (8): 164-171.doi: 10.16265/j.cnki.issn1003-3033.2019.08.026

• 公共安全 • 上一篇    下一篇

基于Copula相依模型的地铁结构安全可靠性分析

刘文1, 刘文黎**1, 翟世鸿2 教授级高工   

  1. 1 华中科技大学 土木工程与力学学院,湖北 武汉 430074;
    2 中交第二航务工程局有限公司,湖北 武汉 430040
  • 收稿日期:2019-04-25 修回日期:2019-06-24 发布日期:2020-10-21
  • 通讯作者: ** 刘文黎(1989—),男,湖北洪湖人,博士研究生,主要研究方向为地铁及隧道安全管理。E-mail:newheny2014@hotmail.com。
  • 作者简介:刘 文 (1985—),男,湖北恩施人,博士研究生,主要研究方向为建筑施工技术及质量管理,建筑施工安全技术,建筑设施安全管理。E-mail:379047775@qq.com。
  • 基金资助:
    国家自然科学基金资助(71801101);教育部博士后基金资助(2018M632880)。

Metro structural safety and reliability analysis based on Copula dependency model

LIU Wen1, LIU Wenli1, ZHAI Shihong2   

  1. 1 School of Civil Engineering & Mechanics, Huazhong University of Science and Technology, Wuhan Hubei 430074, China;
    2 Second Harbor Engineering of China Communications Construction Limited Liability Company, Wuhan Hubei 430040, China
  • Received:2019-04-25 Revised:2019-06-24 Published:2020-10-21

摘要: 为提高地铁运营期的安全性,利用Copula相依性建模理论,研究时变效应下临近环境变化对地铁结构安全的影响。采用结构可靠度指标描述运营期地铁结构的安全性,通过Copula相依性建模理论描述地铁运营期监测指标参数之间的相关性,构建净空收敛(CD)和差异沉降(DS)二元离散时变相依模型,并以武汉地铁2号线某区间为工程背景,验证该模型的准确性。结果表明:Weibull分布和Frank Copula函数分别为最优边缘分布函数和Copula函数;Gaussian Copula并非最优Copula函数,传统的Gaussian Copula函数描述的参数相关性并不太准确;利用构建的Copula模型分析地铁CD和DS值,实现时间效应下盾构地铁结构安全性变化的敏感捕捉;针对分析结果可制定相应运营期风险控制专项方案。

关键词: 运营期地铁, Copula函数, 时间效应, 相依性建模, 可靠性分析

Abstract: In order to improve metro operation safety, Copula dependency modeling theory is used to study influences of adjacent environmental changes under time-varying effects on metro structural safety. Firstly, structural reliability indicators were adopted to describe its safety during operation. Then based on Copula theory, the correlation between parameters of monitoring indexes during metro operation was described and binary discrete time-varying dependence model of convergence deformation (CD) and differential settlement (DS) was constructed. Finally, the model's accurancy was verified with a section of Wuhan Metro Line 2 as a project backgrround. The results show that Weibull distribution and Frank Copula are the optimal edge distribution function and Copula function respectively while Gaussian Copula is not optimal since traditional Gaussian Copula is not accurate in describing parameters' correlation. Copula model in this study can sensitively capture structure safety changes under time effect by analyzing CD and DS values, thus helping formulate corresponding special operation risk control scheme based on analysis results.

Key words: operational metro, Copula function, time-varying effect, dependency modeling, reliability analysis

中图分类号: