In order to get a deep understanding of composite failure mechanisms, the new advanced signal processing methodologies are established for the analysis of the large number of acoustic emission (AE) data obtained from the quasi-static tension test of carbon fiber twill weave composite. For this purpose, AE signals have been collected and post-processed for tension test, and are analyzed with three signal processing methods: Empirical Mode Decomposition (EMD), Hilbert-Huang Transform (HHT) and modified energy entropy algorithm. AE signals can be decomposed into a set of Intrinsic Mode Functions (IMF) components, results from this study reveal that the peak frequency of IMF components based on Fast Fourier Transform (FFT) corresponds to different damage mechanisms of composite. HHT of AE signals can clearly express the frequency distribution of IMF component in time-scale in different damage stages, and can calculate accurate instantaneous frequency for damage modes recognition. The energy entropy based on EMD is introduced to act as a new relevant descriptor of composite damage modes in order to improve the characterization and the discrimination of the damage mechanisms.
