An Unsupervised Pattern Recognition Approach for AE Data Originating from Fatigue Tests on Polymer-Composite Materials
An Unsupervised Pattern Recognition Approach for AE Data Originating from Fatigue Tests on Polymer-Composite Materials
Dinh Dong Doan,E. Ramasso,2 Authors,N. Zerhouni
2014 · DOI: 10.1016/J.YMSSP.2015.04.011
67 Citations
TLDR
A Mahalanobis distance-based noise modeling has been proposed and an unsupervised classifier offline-learned from quasi-static data is used to classify the data to different AE sources with the possibility to dynamically accommodate with unseen ones.