局部损伤所引起的结构特性改变将导致不同自由度之间的传递特性发生变化,而传递性函数的改变可用于结构损伤识别。但是,基于Fourier变换的传递性函数对结构轻微局部损伤不够敏感。本文利用小波变换的时-频多分辨率特性,在小波域内构造了一种新的传递性函数,并且结合统计学中的离群值分析,建立了基于小波传递性函数的结构损伤探测方法。最后通过一个斜拉桥模型试验对方法的可行性予以验证。研究表明,所提出的小波传递性函数指标对结构损伤十分敏感,利用本文的损伤探测方法可以有效地对结构的轻微损伤进行识别。
The changes in the structural properties caused by the local damage often lead to the change of transmissibility between the different degrees of the freedom,which can be utilized to identify the structural damage. However,the Fourier transform based transmissibility functions is not sensitive to the slight local damage in the structures. In this paper,a wavelet transform based transmissibility function is proposed to achieve the refined analysis performance by using the multi-resolution property in the time-frequency domain. The methodology is employed with the outlier analysis to detect the structural damage. A small scale cable-stayed bridge model is constructed and used to verify the effectiveness of the proposed method. The results show that the wavelet transmissibility function is very sensitive to the local damage and the proposed approach can detect the slight damage in the structures.