Non-deterministic approach for identification and isolation of ultrasonic guided wave modes for structural health monitoring

Gullapalli, Anirudh, Featherston, Carol ORCID: https://orcid.org/0000-0001-7548-2882 and Kundu, Abhishek ORCID: https://orcid.org/0000-0002-8714-4087 2026. Non-deterministic approach for identification and isolation of ultrasonic guided wave modes for structural health monitoring. Mechanical Systems and Signal Processing 250 , 114148. 10.1016/j.ymssp.2026.114148

Full text not available from this repository.

Abstract

Ultrasonic inspection techniques have shown great promise for monitoring progressive damage in thin-walled structures. The ultrasonic signals contain damage fingerprints that can be used for assessment of structural damage and degradation. The signal features are inherently linked to the physical behaviour of fundamental guided wave modes. This study presents a novel signal reconstruction and modal identification approach for experimentally measured ultrasonic signals with composite waveguide dispersion models and harmonic wave propagation functions. The modal amplitudes and dispersion characteristics have been calibrated accurately using both a deterministic approach and a Bayesian joint parameter estimation technique. The latter quantifies the uncertainties in both experimental measurements and latent dispersion parameters. The modal identification is regularized by physics-informed models of waveguide dispersion. The reconstructed signals show excellent agreement with the experimental measurements over a broad frequency range. The calibrated parameters were subsequently used to investigate progressive structural degradation arising from displacement-controlled compressive fatigue loading. A probabilistic Bayesian joint parameter estimation framework effectively captured direction-specific signatures and quantified uncertainty in parameter estimation, revealing distinct directional and modal sensitivities to fatigue damage. This achievement underscores the efficacy and reliability of the calibrated ultrasonic guided wave modes as reliable identifiers of damage with potential for further description, characterization, and sentencing.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Schools > Engineering
Publisher:	Elsevier
ISSN:	0888-3270
Date of Acceptance:	10 March 2026
Last Modified:	23 Mar 2026 14:45
URI:	https://orca.cardiff.ac.uk/id/eprint/185965

Actions (repository staff only)

Edit Item