Layered material characterization using ultrasonic transmission. An inverse estimation methodology

Abstract

This paper presents an inverse methodology with the aim to characterize a layered material through the identification of acoustical and mechanical properties of its layers. The framework to accomplish this objective is provided by the Inverse Problems (IPs) theory. Material characterization refers to the detection and localization of discontinuities, as well as to the identification of physical properties, in order to predict the material behaviour. In this particular case, the IP is solved in the form of a parameter estimation problem, in which the goal is the estimation of the characteristic acoustic impedance, transit time, and attenuation of each layer. These parameters are directly related to relevant material properties, such as the speed of sound, density, elastic modulus and elastic energy dissipation constants. The IP solution is obtained by minimizing a cost functional formulated as the least squares error between the waveform calculated using an equivalent model, and the measured waveform obtained from ultrasonic transmission tests. The applied methodology allowed the accurate estimation of the desired parameters in materials composed of up to three layers. As a second contribution, a power law frequency dependence of the wave attenuation was identified for several homogeneous materials, based on the same ultrasonic transmission experiments.