Exploring Frequency Based Analysis Methods for Damage Identification in Fiberglass-Epoxy Composite Systems.
Acoustic Emission (AE) monitoring has been shown to provide valuable information about the damage progression in materials under an applied load. For AE measurements in glass fiber reinforced composites (GFRC), the analysis of the AE signals and their correlation to specific damage mechanisms has shown significant promise; however, many details of the process have yet to be completely understood. Peak-frequency analysis, using the Fast Fourier Transform (FFT) was among the first and simplest techniques to be utilized and identified frequency ranges that were related to specific damage mechanisms. More complicated signal processing methods including artificial neural networks and data clustering techniques have become more prevalent recently. The research presented in this paper demonstrates an alternative signal-processing technique applied at the waveform level through the use of the Short Term Fourier Transform. The STFT allows individual waveforms to be evaluated in frequency and time, resulting in a detailed view of the mixed-mode waves present in an AE signal. For GFRC laminates undergoing quasi-static tensile tests, the addition of the time domain analysis resulted in a shift in peak frequencies compared to conventional FFT approach.
Authors: Michael M. Voth, David A. Miller, Jake D. Nunemaker, Paul Murdy, Daniel D. Samborsky, and Douglas S. Cairns
Conference: SAMPE Seattle 2017