Integration of Distributed Optical Fibres and Piezoelectric Sensors for SHM of Composite Tubes

The aim of the study is to detect, localise, and attempt to classify damage mechanisms in a carbon fibre/epoxy composite tube from an early stage of loading until final failure. Parametric analysis is completed on acoustic emission (AE) data to attempt damage classification, while experimental wave velocities are used to estimate damage locations. This is complemented by damage index calculations based on the amplitude reduction observed in guided wave signals. The application of an unsupervised clustering algorithm enables observation of the evolution of damage, as separated into groups. When overlaid with the cumulated energy obtained from AE signals, these “damage profiles” for each cluster of data can be linked more closely to the known damage processes.