A FRF-based damage detection method utilising wavelet decomposition

Damage in a structure causes deviation in dynamic responses of the structure either in frequency or time domain in comparison with its healthy status. The purpose of this study is to present a new damage detection method in order to detect and localize the structural damage. This novel algorithm is based on the directly-measured frequency response functions (FRFs). The approach is composed of three major steps: first, developing the curvature of FRFs which produces spatially distributed shape functions at each frequency coordinate, secondly, normalization of FRFs' curvature in order to enhance the influence of the lower-frequency-band data; finally decomposition of the obtained profiles (normalized version of FRFs' curvature) by conducting wavelet analysis to detect any possible structural abnormality through structure. The combination of these three steps leads to a robust algorithm in detection and localisation of any damage in the structure even at small levels which other FRF-based methods were unable to detect. There are some benefits with the presented method: first, this method does not need higher-frequency-range data which is hard to obtain in most civil applications; second, there is no need for baseline data from the intact structure; This is particularly attractive for practical applications as it opens an opportunity for online monitoring of the structural integrity without demanding for any previous data records of the structure. The performance of the method is evaluated on a numerical model and the effect of different parameters such as the location of the excitation point, the level and the location of the damage was studied; the results demonstrated that the method can efficiently identify the location of the damage in the structure even for damage at small levels.