Detecting the Presence of Crack in a Beam with Compressive Sensing (CS)

One of the first characteristics of a damaged structure is a change in the local stiffness of the structure and the consequent change in its natural frequencies. In recent years, Compressive Sensing (CS) has achieved remarkable success compared to the Nyquist sampling rate. Compressive Sensing is based on the fact that most natural signals are sparse when displayed on a suitable basis (such as wavelet, Fourier, etc.), so that when sampling signals, sampling of unnecessary parts of the signal can be omitted. As a result, it significantly reduced the number of samples compared to the Nyquist rate. In this research, for the first time, an Orthogonal Matching Pursuit algorithm (OMP) and an Iterative Hard-threshold algorithm (IHT) to detect the presence of cracks in a beam with the lowest signal sampling rate (SNR) are used. The results of this method indicate that when the Gaussian signal-to-noise ratio (SNR) is low, the proposed method performs better.