Clustering of AE signals collected during torsional tests of 3D braiding composite shafts using PCA and FCM. (15th March 2019)

• Record Type: Journal Article
• Authors: Zhao, Guoqi; Zhang, Lu; Tang, Can; Hao, Wenfeng; Luo, Ying
• Is Part Of: Composites. Number 161(2019)
• Page Start: 547
• Page End: 554
• Publication Date: 2019-03-15
• Subjects: Acoustic emission -- 3D braiding composites shaft -- Torsional test -- Principal component analysis -- Fuzzy clustering method; Composite materials -- Periodicals; Materials science -- Periodicals; Composite materials; Periodicals; Electronic journals; 620.118
• DOI: 10.1016/j.compositesb.2018.12.145
• Languages: English
• ISSNs: 1359-8368

Abstract

Abstract: In this work, the acoustic emission (AE) signals collected during the torsional tests of the 3D braiding composite shafts were clustered using the principal component analysis (PCA) and the fuzzy clustering method (FCM). The 3D 4-directional braided composite shafts with various braiding angles were prepared using Vacuum Assisted Resin Transfer Moulding (VARTM). The torsional tests were performed on the static torsion machine and the AE was used to monitor the damage progression. The AE signals were analyzed by PCA and FCM. The accumulated damage was manifested by matrix cracking, fiber bundle-matrix debonding, and fiber breakage by the proposed method. The failure modes and the mechanisms of the 3D 4-directional braided composite shafts were characterized. The results will play important engineering roles in the design and the application of the 3D 4-directional braided composite shaft.