基于贝叶斯理论的高铁站房结构损伤分析识别方法

Analysis and identification method of railway station building structure damage based on Bayesian theory

由于工程实际损伤识别过程中建模和测量误差的影响,不可避免地带来结构力学行为的不确定性,利用贝叶斯方法来进行不确定结构损伤识别成为研究的一个热点。本文介绍了贝叶斯方法的基本理论,利用贝叶斯理论框架,以环境激励结构响应功率谱为观测量,建立了结构刚度参数在考虑测量信息下的条件概率密度,利用最大后验估计构造目标函数通过优化迭代技术进行刚度参数识别。最后,利用提出的损伤识别方法对某高铁站结构不同损伤工况下杆件材料刚度参数进行识别,取得较好的效果。

Due to the influence of modeling and measurement errors in the actual damage identification process, the uncertainty of structural mechanical behavior will inevitably be brought. The Bayesian method to identify uncertain structural damage has become a research hotspot. In this paper, the basic theory of Bayesian method is introduced. Based on the Bayesian theory framework, the conditional probability density of structural stiffness parameters considering the measurement information is established. The objective function is constructed by using the maximum a posteriori estimation, and the stiffness parameters are identified by optimization iteration technique. Finally, the proposed damage identification method is used to identify the stiffness parameters of members under different damage conditions of a high-speed railway station structure, and good results are obtained.