桥梁支座温致位移概率建模性能预警方法

Performance early warning method for probabilistic modeling of thermal displacement of bridge bearing

提出了一种监测数据驱动的大跨桥梁支座服役性能概率预警方法。首先,基于正常服役状态下获取的长期监测数据,对桥梁的结构温度场及其支座位移的联合概率密度进行建模;其次,利用该模型对未知服役状态下的温致支座位移的条件概率密度进行预测;最后,根据退化支座轴承纵向位移退化规律模拟相应支座的纵向位移,以达到对支座性能劣化预警的目的。通过某大跨斜拉桥为期12个月的连续监测数据研究分析马氏平方距离指标和欧氏平方距离指标对本文所提方法的有效性进行验证,结果表明:所提方法能有效建立结构温度场和支座位移的联合概率密度模型,且能准确对支座退化性能有效预警;马氏平方距离指标法优于欧氏平方距离指标。

A probabilistic warning method for service performance of long-span bridge bearingss driven by monitoring data is proposed in this paper.Firstly,based on the long-term monitoring data obtained in normal service,the joint proba⁃bility density of the structural temperature field and the bearings displacement of the bridge are modeled.Secondly,the model is used to predict the conditional probability density of the temperature-induced bearings displacement under un⁃known service conditions.Finally,the longitudinal displacement of the corresponding bearing is simulated according to the degradation law of the longitudinal displacement of the degraded bearing to achieve the purpose of warning the deteri⁃oration of the bearing performance.The validity of the proposed method has been verified by analyzing the mahalanobis square distance index and Euclidean square distance index from the continuous monitoring data of a long-span cablestayed bridge for 12 months.The results show that the proposed method can effectively establish structure temperature field and the displacement of the joint probability density model,and can accurately for bearing performance degradation effective early warning;Markov square distance control chart is superior to Euclidean square distance control chart.Ma⁃halanobis square distance index method is better than Euclidean square distance index.