Timelapse ultrasonic tomography for measuring damage localization in geomechanics laboratory tests.

Erika Tudisco, Philippe Roux, Stephen Hall, Giulia M.B Viggiani, Gioacchino Viggiani

Forskningsoutput: TidskriftsbidragArtikel i vetenskaplig tidskriftPeer review

264 Nedladdningar (Pure)

Sammanfattning

Variation of mechanical properties in materials can be detected non-destructively using ultrasonic measurements. In particular, changes in elastic wave velocity can occur due to damage, i.e., micro-cracking and particles debonding. Here the challenge of characterizing damage in geomaterials, i.e., rocks and soils, is addressed. Geomaterials are naturally heterogeneous media in which the deformation can localize, so that few measurements of acoustic velocity across the sample are not sufficient to capture the heterogeneities. Therefore, an ultrasonic tomography procedure has been implemented to map the spatial and temporal variations in propagation velocity, which provides information on the damage process. Moreover, double beamforming has been successfully applied to identify and isolate multiple arrivals that are caused by strong heterogeneities (natural or induced by the deformation process). The applicability of the developed experimental technique to laboratory geomechanics testing is illustrated using data acquired on a sample of natural rock before and after being deformed under triaxial compression. The approach is then validated and extended to time-lapse monitoring using data acquired during plane strain compression of a sample including a well defined layer with different mechanical properties than the matrix.
Originalspråkengelska
Sidor (från-till)1389-1400
TidskriftJournal of the Acoustical Society of America
Volym137
Nummer3
DOI
StatusPublished - 2015

Ämnesklassifikation (UKÄ)

  • Geoteknik

Fingeravtryck

Utforska forskningsämnen för ”Timelapse ultrasonic tomography for measuring damage localization in geomechanics laboratory tests.”. Tillsammans bildar de ett unikt fingeravtryck.

Citera det här