基于多种Copula模型的地铁运营隧道结构可靠性分析

Reliability Analysis of Subway Operational Tunnel Structure Based on Multiple Copula Models

在地铁隧道结构运营管理中,精准评价运营隧道结构的安全性很有必要。本文考虑参数之间的关联性,提出了一种基于Copula理论的运营隧道可靠性评价方法。选用26种Copula函数,建立可靠度分析模型;分别通过局部优化算法和MCMC方法进行参数评估;利用最大似然法、赤池信息准则(AIC)和贝叶斯信息准则(BIC),识别出最优的Copula函数;通过蒙特卡罗模拟(MCS)计算出隧道结构的系统可靠性,并以武汉市地铁3号线的某段区间为案例进行分析。结果表明:不同Copula函数的拟合效果明显不同;在可靠性分析中采用传统Nataf模型或参数独立假设是不准确的;同时,现有评估Copula参数的研究中,广泛使用的局部优化算法的鲁棒性不足;有必要识别出最优Copula函数用于精确表征参数的相依关系。

It is necessary to evaluate operational tunnel structure safety precisely in the operational management of subway tunnel structure.In this paper,a reliability analysis method based on Copula theory was proposed for the operational tunnel,considering the correlation between parameters.Twenty-six Copula functions were selected to establish the reliability analysis model.The local optimization algorithm and the MCMC method were applied to evaluate the parameters.The optimal copula function was identified by maximum likelihood method,red pool information criterion(AIC)and Bayesian information criterion(BIC).The system reliability of tunnel structure was calculated by Monte Carlo simulation(MCS),and a section of Wuhan Metro line 3 was investigated.The results show that the fitting effects of different Copula functions are obviously different.It is inaccurate to use the traditional Nataf model or parameter independent hypothesis in reliability analysis.At the same time,the robustness of the widely used local optimization algorithm is insufficient in the existing research on the evaluation of Copula parameters;it is necessary to identify the optimal copula function to accurately characterize the dependence of parameters.