Toward real-time assessment of material properties using elastic guided waves

Elastic guided waves are one of the most promising quantitative ultrasound techniques for the assessment of elastic properties in plate-like structures. Measurements of guided waves, associated with suitable waveguide modeling, can yield accurate estimates of waveguide properties like thickness and stiffnesses. Such a model-based approach requires solving a multi-parametric inverse problem to match experimental data with guided modes. Sensitivity studies suggest that isolated areas of the dispersion curves have predominant influence on specific model parameters. In particular, guided waves in a free plate exhibit a resonant behavior at frequencies where their group velocity vanishes while their phase velocity remains finite (so-called zero-group velocity Lamb modes). In this study, we investigate the feasibility of exploiting targeted data in the vicinity of these particular points, along with data associated with the cut-off frequencies. To this end, guided waves measurements are performed on a series of materials (anisotropic plates and tri-layer structures) using a linear multi-element transducer array and dedicated signal processing. A genetic algorithm-based inverse problem is then solved to recover the waveguide properties. Preliminary results indicate that, owing to the measurements versatility in terms of acquisition speed, this approach has the potential for inferring reliable structural and material properties in real-time.