Experimental validation of differential evolution indicators for ultrasonic imaging in unknown backgrounds

TL;DR

This study validates differential evolution indicators for ultrasonic imaging of damage progression in solids, demonstrating their effectiveness in tracking fracture development in granite under controlled laboratory conditions.

Contribution

It provides experimental evidence that differential evolution indicators can accurately image and monitor damage evolution in scattering solids, extending their application to unknown backgrounds.

Findings

Successfully reconstructed damage support and discontinuities.

Indicators reveal crack fragmentation and coalescence.

Effective even with partial and reduced data sets.

Abstract

The differential evolution indicators, recently introduced for imaging mechanical evolution in highly scattering solids, is examined in a laboratory setting with the focus on spatiotemporal tracking of an advancing damage zone in an elastic specimen. To this end, a prismatic slab of charcoal granite is quasi-statically fractured in the three-point-bending (3PB) configuration, while ultrasonic shear waves are periodically generated in the sample at certain time steps $t_\kappa$, $\kappa = \circ, 1, 2, ..., 4$. The interaction of probing waves with the propagating damage give rise to transient velocity responses measured on the plate's boundary by a 3D scanning laser Doppler vibrometer. Thus obtained sensory data are then carefully processed to retrieve the associated spectra of scattered displacement fields $v^\kappa$ at every $t_\kappa$. On deploying consecutive pairs of multi-frequency data $(v^\kappa, v^{\kappa+1})$, the differential indicators are computed exposing the progress of 3PB-induced damage in the specimen. Verified with in-situ observations, each indicator map successfully reconstructs (a) the support of newborn fractures, and (b) the loci of discontinuities in the process zone that undergo interfacial evolution in the designated timeframe $[t_\kappa \,\,\, t_{\kappa+1}]$. Further, it is shown that the evolution indicators help better understand the damage mechanism e.g., by shining light on the fragmented nature of induced cracks and their coalescence. For completeness, data inversion via reduced and partial-aperture data is investigated, including the one-sided reconstruction.