考虑变量相关性的正交异性板细节疲劳可靠性评估

传统的疲劳可靠性评估方法忽略了功能函数中变量的相关性,致使计算得到疲劳可靠度指标精度降低。针对这一问题,基于既有悬索桥主梁细节应变和环境温度大样本监测数据,讨论疲劳功能函数变量之间的相关性关系。提出基于Copula函数的相关性变量的联合分布模型,解决联合概率分布建模中多重积分求解困难的问题。研究表明,受环境温度的影响,靠近顶板焊接细节的日等效应力幅和应力循环次数存在较强的相关性,Gaussian Copula函数可作为相关性变量的最优连接函数,对于U肋-顶板细节,桥梁结构服役100年时,考虑与不考虑变量相关性的可靠度指标值分别5.3和6.9,两者之间相差约1.3倍。当年交通增长率分别等于3%和5%时,可靠度指标在服役年限为94.6年和67.1年时间到达目标可靠度指标。对于U肋-U肋对接细节,考虑和不考虑变量的相关性对可靠性指标影响相对顶板细节较小。

Fatigue reliability assessment of orthotropic steel deck weld details considering correlation between variables

The traditional fatigue reliability assessment method does not consider variables correlation.For this reason,the calculation value of fatigue reliability index has low accuracy.Concerning this issue,the orthotropic steel bridge deck(OSBD)detailed stress and ambient temperature data were collected.This study analyzed the correlation between the fatigue load effect variables.On this basis,the Copula function was introduced to establish a joint distribution model,and the issue of multiple integral calculations was solved.The research shows that the details which were close to the roof daily fatigue stress amplitude and numbers of variables have strong correlation.The Gaussian Copula is the optimal connection function for the correlation variables.For U-rib to deck details,after 100 years operation,the fatigue reliability indexes which considering and without considering the variables’correlation are 5.3 and 6.9.The value of considering variables’correlation is about 1.3 times to the value of without considering variables’correlation.When the traffic load growth rate is 3%and 5%,the reliability indexes are reach the target value in 94.6 years and 67.1 years.For U-rib to U-rib details,there have a few influences to consider variables’correlation.