Simulation-Based Characterization of Mechanical Parameters and Thickness of Homogeneous Plates Using Guided Waves

Abstract

Properties of isotropic plates in terms of material constants and thickness are characterized by making use of dispersion characteristics of propagating Lamb waves. A numerical model is inversely optimized in order to match dispersion curves measured by laser vibrometry. This kind of material characterization has gained interest in recent research. Improvements in accuracy and efficiency of the optimization process are therefore important steps toward an industrial application of this technique to nondestructive testing and online monitoring. For this purpose, the use of a fast converging numerical model based on spectral collocation has been found to be well suited. Furthermore, we improved the signal-to-noise ratio by utilizing long-time broadband excitation signals for multimodal excitation of Lamb waves. The wavenumber spectrum up to 2.5 MHz is acquired by measurements with a laser-scanning vibrometer. In order to exploit the information contained in high-order modes, we present an algorithm to match the measured data to the calculated modes during the optimization process, leading to higher accuracy of the estimated model parameters. The characterization results are verified by comparison to measurements with a conventional ultrasonic method.

Daniel A. Kiefer

Researcher at Institut Langevin

Research in guided elastodynamic waves, fluid-structure interaction, simulation and signal processing for ultrasonic sensors and nondestructive testing.