Lamb Wave-based Damage Indicator for Plate-Like Structures

Abstract

Structural health monitoring based on ultrasonics typically involves complex data analysis. Ultrasound monitoring based on Lamb waves techniques are extensively used nowadays due to their efficiency in exploring large areas with relatively small attenuation. In recent years, baseline based methods have been developed to identify structural damage based on the mismatch between the measured signal and the baseline one. To this end, complex time-frequency transformations are required to obtain signal features such as the time of arrival or the energy content, as indicators of damage onset and growth. Notwithstanding this, on-board applications require highly efficient processing techniques due to information storage and exchange limitations. This paper proposes a very high efficiency signal processing methodology to obtain a novel cumulative damage factor using Lamb wave raw data. The new methodology has been tested using ultrasonic and damage data from a fatigue test in carbon-epoxy composite laminates. The data is taken from NASA Prognostics data repository. In view of the results, the method is able to efficiently detect the onset and extent of damage from early stages of degradation. Moreover, the results demonstrate a remarkable agreement between the growth of delamination area and the predicted cumulative damage factor.