Detecting Damages in Beams Using Hilbert-Huang Transform

This work illustrates the feasibility of the Hilbert-Huang transform (HHT) as a signal processing technique to locate and quantify the damages in beam elements. HHT is a powerful method for non-linear and non-stationary time series analysis. A key parameter of the method is the extraction of intrinsic mode functions that admit wellbehaved Hilbert transforms in magnitude and phase. The methodology to infer the damages is based on detecting the changes in the elements of HHT-intrinsic mode functions, instantaneous frequency and amplitude. Some researchers have shown that such structural modal parameters as mode shapes, natural frequencies etc. are not sensitive to local and small structural damage. Therefore, an index vector is defined in this study for damage identification. Dynamic model of beams with damage in the form of a crack is established using a MATLAB code developed in the study and accordingly the vibration responses of intact and damaged beams are numerically obtained. Then, the dynamic responses are processed using HHT to obtain the feature index vectors of beams with different degrees of damage. Based on the numerical data, the technique is shown to be robust in terms of the relationship between the extent of structural damage, location and the index vector Note that, however, the study presented here is purely analytical, thus, the results of the study needs to be tested using experimental data.