斜拉桥结构损伤识别的概率可靠度法

从概率可靠度的角度出发,提出并发展了基于参数识别和假设检验的斜拉桥结构损伤识别方法.将正则化理论引入参数识别过程,综合利用蒙特卡罗方法和最优化理论识别结构的主要参数并获得其概率分布特性,进而采用假设检验确定损伤的位置和程度,实现损伤的概率诊断.为验证方法的准确性和有效性,将其应用于斜拉桥的损伤识别问题,提出了斜拉索损伤识别的斜拉索索力指标,得出了有益的结论.

Statistical damage assessment method of cable-stayed bridge

A damage assessment algorithm of cable-stayed bridge based on parameter identification and hypothesis test is developed.The Monte Carlo method and optimization theory are employed jointly to obtain main parameters of the objective structure and corresponding statistical distributions.Furthermore,hypothesis test is utilized to locate the damage and assess its severity.To reflect the effect of measurement noise and alleviate difficulty of the damage identification problem,the data perturbation method and regularization technique are introduced.The validity and feasibility of the proposed method is demonstrated by a numerical simulation case based on a cable-stayed bridge,and cable tension index for damage assessment of cable is presented.