Clustering of acoustic emission signals collected during tensile tests on unidirectional glass/polyester composite using supervised and unsupervised classifiers

Acoustic Emission (AE) can be used to discriminate the different types of damage occurring in a constrained composite. However, the main problem associated with data analysis is the discrimination between the different acoustic emission sources. The objective of the cluster analysis is to separate a set of data into several classes that reflect the internal structure of the data. Indeed, cluster analysis is an important tool for investigating and interpreting data. In this paper we use two kinds of classifiers: a supervised classifier and also an unsupervised one (Kohonen's map). We combine two techniques: the k-means algorithm and the k nearest neighbours. Glass/polyester model specimens were used for the validation of the proposed methodology. We worked on polyester resin and glass/polyester unidirectional specimens, subjected to tensile loading within different configurations, awaiting preferential damage modes in the material. Moreover, single fibre composites have been tested to produce fibre breakage acoustic emission events under conditions closely approximating those encountered in a real composite.