Synchrosqueezed wavelet transform-based method for characterizing the dispersive nature of laser-excited surface acoustic waves propagating through the coated or damaged medium

Abstract

Dispersion of surface acoustic waves (SAWs) can be used for determination of the mechanical parameters of the film or the damaged layer. The correct evaluation of the experimental dispersion curves is important for backcalculation. In this paper, we present a method to identify experimental dispersion curves from two-station time-varying signals based on time–frequency analysis. Synchrosqueezed wavelet transform (SWT) provides the quantitative description of the propagation of different frequency components of SAWs. Discrete phase velocity estimation is accomplished based on the cross-correlation function. Rényi entropy-based SWT spectrum noise reduction and cosine interpolation are applied for improving the quality of dispersion curves. An example is presented to illustrate the application of the proposed method. Finally, laser ultrasonic experiments are performed on the surface of three typical test samples. The comparison of the resulting dispersion curves shows the effectiveness of the proposed method and the necessity of noise reduction and interpolation.