A Quantitative Review of Air-Coupled Ultrasonic Lamb Wave Analysis Based on Signal Transformations

Abstract

Lamb wave detection is increasingly being utilized in the industry due to its extensive coverage area, high signal detection efficiency, and ease of operation. This paper offers a quantitative review of eight signal transformation methods utilized for de-noising and time-frequency analysis of Lamb waves, which include Fourier transform (FT), singular value decomposition (SVD), short-time Fourier transform (STFT), Wigner–Ville distribution (WVD), wavelet transform (WT), S-transform, Hilbert–Huang transform (HHT), as well as empirical mode decomposition (EMD) and its improved algorithms. The performances of signal transformations on denoising and defect location are assessed quantitatively using the signal-to-noise ratio (SNR) and time-of-flight (ToF). The results demonstrate that the complete ensemble EMD with adaptive noise (CEEMDAN) is able to suppress noise effectively while maintaining the primary features of the signal in an adaptive manner. Additionally, the continuous WT can obtain a more accurate time-frequency distribution, thereby providing the superior analytical ability for dispersive lamb wave signals with respect to positioning on the time axis.