Analysis of acoustic emission signals in machining
Conference Paper
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
In this paper, we present a novel approach, combining chaos theory with wavelets, for Acoustic Emission (AE) signal analysis. We perform thorough characterization of AE signals collected from extensive experimentation on the turning process and develop time-frequency representation - called Suboptimal Wavelet Packets (SWP) - to compactly model AE signals. We use the features extracted from SWP representation for on-line flank wear estimation. The results show that the developed techniques perform better than the existing analytical methods for AE signal representation and AE signal based state estimation.