基于试验设计的结构损伤概率诊断方法

Statistical Damage Assessment Method of Structures Based on Experimental Design

基于结构损伤识别问题的求解特点,将试验设计引入损伤识别过程,提出基于均匀试验设计的结构参数识别及损伤概率诊断方法。首先通过响应面拟合获得结构响应和待识别参数之间的显式表达式,然后据此进行参数识别问题的求解,以提高参数识别问题的求解效率并降低其求解难度。在此基础上提出结构损伤的概率诊断方法。根据一斜拉桥结构的损伤识别算例对本文方法和传统损伤识别方法进行对比研究。研究结果表明,所提出的方法能够显著提高损伤识别的求解效率,并能够在观测噪声水平较高的条件下实现损伤的准确定位及损伤程度识别。本文的方法和思路可为结构损伤识别提供一种可行途径。

Based on the theory of uniformly experimental design, a new parameter estimation and statistical damage assessment method is developed. The uniform design algorithm is introduced to alleviate the solution difficulty of structural damage detection in consideration of the characteristics of damage assessment. The ex- plicit relation formula between structural response and parameters is obtained by response surface fitting, and parameter estimation can be solved based on such explicit expressions. Then the solution difficulty of the parameter estimation problem is reduced and its solution efficiency is improved notably. The statistical damage assessment method is established and the damage detection of a cable-stayed bridge shows that the method presented has outstanding efficiency. This advantage over the traditional approaches is enhanced by its ability of accessing damage exactly in the case of test data polluted by measurement noises.