Identification of Different Acoustic Emission Signal Sources

Acoustic emission (AE) technique is one of the promising methods for structural health monitoring. However, the major difficulty in monitoring is the identification of the actual signal of interest in presence of noisy signal sources such as any impact and rubbing activity in the structure that also generate AE wave. Therefore, the discrimination of AE signals generated from different sources is an important task for successful monitoring. In this work several experiments have been performed to generate AE signals from three different test set up such as i) pencil lead break up test, ii) impact test, and iii) rubbing test. After detection and acquisition of AE signals from three different sources certain methods such as parameter analysis, short time Fourier transform, cross-correlation coefficients (CCC), magnitude squared coherence (MSC), and energy distribution of the AE signals have been compared from different sources for identification of each type of signal.