Quantitative analysis of acoustic emission signals

Abstract
Acoustic emission (AE) signals emanating from various deformation processes in materials are of a broadband nature. It is also well established that the ultrasonic attenuation caused by grain boundary scattering in polycrystalline materials is strongly frequency dependent. It is worth noting that distortions of AE signals due to grain boundary scattering (Rayleigh or stochastic type) have been inadequately treated in some previous quantitative AE studies. We propose a quantitative approach to estimating AE source parameters with proper care taken of the frequency-dependent media attenuation. Without getting involved in the complete measurement of media attenuation, the present approach is designed to find some characteristic quantities in both the frequency and time domains that are least sensitive to attenuation. The node frequency in the amplitude spectra and the peak time in real-time signals are found to be the most suitable for the purpose. They can be used to infer the source parameters such as the final size and the average velocity of source expansion.

This publication has 13 references indexed in Scilit: