Non-Contact Lamb Wave Defect Detection Based Solely on Air-Coupled Ultrasonic Phased Arrays

• Main Authors: Laub, Felix, Haugwitz, Christoph, Allevato, Gianni, Seiler, Julian, Findeisen, Rolf, Kupnik, Mario
• Format: Conference Proceeding
• Language: English
• Subjects: Acoustics; air-coupled; deconvolution; Direction-of-arrival estimation; DoA estimation; Estimation; Lamb wave; Monte Carlo methods; Non-contact NDT; Phased arrays; Production; ultrasonic imaging; Ultrasonic variables measurement
• ISSN: 2168-9229
• DOI: 10.1109/SENSORS56945.2023.10324898

Air-coupled ultrasonic non-destructive testing (NDT) enables the inspection of composite materials for defects. In this paper, we investigate non-contact defect detection based on Lamb waves by using two dedicated air-coupled ultrasonic phased arrays, i.e. one for transmission and one for reception. The major challenge is the signal detection of the reflected leaky Lamb wave originating from the defect, despite a strong interfering signal, i.e. the direct air-path wave, causing a so-called blind zone. Therefore, we exploit the angular difference of the reflected leaky Lamb wave and the direct air-path wave by applying high-resolution direction of arrival (\text{DoA}) estimation algorithms, i.e. MVDR beamformer (Capon) and MUSIC algorithms, as well as image deconvolution methods, i.e. CLEAN and Richardson-Lucy algorithms. First, we assess the signal detection capability by using Monte Carlo simulations considering realistic non-ideal array elements. Second, we validate the simulations by conducting testbed measurements, which show that conventional beamforming (CBF), MVDR and MUSIC are capable of fully air-coupled defect detection, even when the direct air-path wave is impinging. We conclude that the blind zone can be eliminated and show that the MUSIC algorithm excels in highlighting the reflected leaky Lamb wave.