Lamb wave based damage identification in structures with complex geometry

Structural health monitoring (SHM) plays a significant role in terms of fatigue life and damage accumulation prognostics. SHM for structures with complex geometry are much more practical in engineering applications. In this paper, complex aluminium alloy structures with "U" shape section were evaluated in terms of both finite element method (FEM)- and experiment-based Lamb wave analysis for the purpose of damage detection and identification. In the FEM-based analysis, three-dimensional finite element model was established to simulate the propagation behavior of Lamb wave in the structures. On the other hand, in the experiments, piezoelectric (PZT) wafers, functioning as both actuator and sensor, were used to generate Lamb waves propagating in the structures and collect the Lamb wave signals from the complex structures. Quantitative relationship between crack location and the reflection coefficient was constructed by taking advantage of continuous wavelet transform (CWT) and Hilbert transform (HT), which are based on the collected Lamb wave signals. Furthermore, the differences between simulated and experimental results in respect of crack severity evaluation and the reasons were discussed.