A novel imaging technique for structural health monitoring using sparse and compact arrays

Abstract

In the present paper, a technique called Excitelet is presented for the imaging of damages in thin-walled structures using the correlation of the measured signals with dispersed versions of the excitation signal. Piezoceramic (PZT) actuators are used to generate burst Lamb waves which interact with defects in metallic structures and the measurement is taken using sparse and compact array configurations of PZT sensors. The sparse sensing configuration consists of individual circular PZT elements distributed over the plate while the compact array configuration consists of a linear arrangement of sensors micro-machined on a single piece of bulk PZT wafer. This approach is presented as an extension of the classical imaging techniques and takes advantage of the chirplet-based matching pursuit algorithm. The approach is investigated experimentally on a 1.54 mm thick aluminum plate and comparison with existing Embedded Ultrasonic Structural Radar (EUSR) algorithm is performed for A0 and S0 modes for two frequency ranges of interest (centered at 150 kHz and 550 kHz). Damages are simulated using stacked magnets at different locations on the plate. Significant improvement of imaging quality is demonstrated with respect to existing imaging techniques based on group velocity and Time-of-Flight (ToF), for both sparse and compact PZT array configurations. Multimodal imaging strategies are presented to improve the imaging results. Moreover, it is shown that the proposed imaging technique provide accurate results in the case of dispersive propagation, while existing imaging techniques are no longer applicable.