Bayesian modelling for determining material properties

Abstract

Sound wave propagation in materials has been used extensively for nondestructive testing of materials, studying the internal structure of Earth or for oil/gas/mineral exploration. The propagation characteristics have been exploited and studied using signal processing tools for analyzing the material properties or for examining the sub-surface features. The signal processing tools can be innovated in accordance with the characteristics of the sound wave propagation in the media for improvement of the signal to noise ratio. A synthetic model of one dimensional chain of spherical particles is used to generate space time responses when impulse moves along the chain, it gives information about the longitudinal wave propagation (P-wave). The space time responses obtained from the synthetic model are used by the Bayesian inference technique to obtain the properties of the media (particle size/mass distribution of the chain). Finally, the importance of Bayesian inference technique as a signal processing tool is discussed upon.