Critical Assessment of Various Guided Wave Based Imaging Damage Detection Algorithms

Lamb wave imaging is a promising technique for the structural health monitoring of composite plates, but the widely used Delay-and-Sum (DAS) method often suffers from poor resolution and high levels of imaging artifacts, i.e. high sensitivity to noise. To overcome these limitations, advanced algorithms adapted from the field of Direction-of-Arrival (DOA) estimation can be employed. This paper provides a detailed comparison of the principles and performance of DAS, Minimum Variance Distortionless Response (MVDR), and Multiple Signal Classification (MUSIC) for localizing simulated damage on a quasi-isotropic composite plate. Since all three methods rely on the common ground of back-propagated scattered signals, this analysis aims to provide insights for developing novel damage imaging techniques. The key to improvement lies in finding effective ways to suppress noise and artifacts, a primary strength of methods like MVDR, and this work suggests that extending other state-of-the-art DOA estimation methods to the damage imaging field is a promising avenue for future research.