Wavelet-based Lamb waves direction of arrival estimation in passive monitoring techniques

Abstract

Guided Wave (GW) testing for Structural Health Monitoring (SHM) is complicated by dispersion in wave mode propagation. In particular, the dispersive behavior of GWs hinders an accurate estimation of the direction of arrival of the wave by monitoring the Difference in Time of Arrival (DToA) associated to different sensors, as normally happens in bulk wave applications. In this work, a novel procedure for passive sensors which overcomes this limitation and precisely locates the wave direction of propagation is presented. The proposed procedure is based on the Continuous Wavelet Transform (CWT) decomposition of the non-stationary signals acquired by three closely-located piezoelectric transducers. Such transformation causes a resampling of the time-frequency domain which allows to separate the wave components traveling at different velocities and to highlight the frequency dependence in the DToA. In this way the DToA among two different transducers can be computed multiple times by applying the conventional cross-correlation method in the CWT domain for each computed scale (or corresponding frequency range). Finally, the GW direction of arrival can be estimated by applying an averaging procedure across scales to the arc-tangent of the ratio between the DToA among specific sensors.