Damage identification on composite materials: A Neural Networking process for fiber reinforced composite materials micro mechanical acoustic emission signatures

This book presents a reliable correlation between micro-mechanical failure events on fibre reinforced composite materials and their acoustic emission (AE) signature. Several fields of knowledge were involved on the research behind this book, for instance, composites theory was applied on the manufacturing of the testing coupons and the understanding of the micro-mechanical failure modes, ultrasound theory for the understanding of generation, transmission and recording of waveforms on composite materials. Finally, artificial neural networks for the classification and clustering of AE events recorded from the material. During pilot testing, the characterization of the recording system was defined, along with the study of the testing settings and physical factors that influence frequency response from fibre reinforced composite materials. The post analysis for each test involved the extraction and classification of the first and second highest frequencies by means of a self-organizing map algorithm. The resulting plots from this analysis showed 5 different classes of acoustic emission events; each class was successfully correlated to a different micro-mechanical event.