Quantitative identification of damages in beams using longitudinal guided wave: an experimental study

Abstract

An experimental study of identifying damages in beams using longitudinal guided wave is presented in this paper. In this study a probabilistic approach is employed to determine the location, length and depth of the damages by solving an inverse problem. The probabilistic approach is developed based on the Bayesian statistical framework. The damage identification is achieved by maximising the corresponding probability density function (PDF) of the damage location, length and depth for a set of measured longitudinal guided wave data. An advantage of the proposed approach is that it also quantifies the associated uncel1ainties of the predicted damage identification results, thus providing essential information for making decisions on necessary remedial work. The experimental study considers beams with a single damage. Damages having different locations, lengths and depths are used to demonstrate the performance of the proposed approach in damage identification. The experimental results show that the proposed approach is able to accurately identify the damages using longitudinal guided wave signals measured at a single location of the beams.