Pattern recognition of fiber-reinforced plastic failure mechanism using computational intelligence techniques

• Main Authors: XuQin Li, Ramirez, C., Hines, E.L., Leeson, M.S., Purnell, P., Pharaoh, M.
• Format: Conference Proceeding
• Language: English
• Subjects: Clustering algorithms; Failure analysis; Glass; Materials; Neurons; Optical fiber sensors; Sensors
• ISSN: 2161-4393; 1522-4899; 2161-4407
• DOI: 10.1109/IJCNN.2008.4634122

Acoustic Emission (AE) can be used to discriminate the different types of damage occurring in composite materials, because any AE signal contains useful information about the damage mechanisms. A major issue in the use of the AE technique is how to discriminate the AE signatures which are due to the different damage mechanisms. Conventional studies have focused on the analysis of different parameters of such signals, say the frequency. But in previous publications where the frequency is employed to differentiate between events, only one frequency is considered and this frequency was not enough to thoroughly describe the behavior of the composite material. So we introduced the second frequency. A Fast Fourier Transform (FFT) is then applied to the signals resulting from the two frequencies to discriminate different failure mechanisms. This was achieved by using self-organizing map and Fuzzy C-means to cluster the AE data. The result shows that the two approaches have been very successful.