Wavelet Transform Based Damage Detection in a Plate Structure

Abstract

A 2-dimensional wavelet transform algorithm employing isotropic Pet Hat wavelet is used to locate the area of damage in a numerically simulated aluminium plate model. To study the advantages and limitations of proposed method, signals of vibrational mode shapes of the plate are reduced by integer numbers to study the performance of algorithm at various sensor densities. Results suggest that the lowest scale of selected wavelet function yields the best damage identification results with a damage location likelihood of more than 98%.