Damage mechanism identification and failure behavior of 2D needle-punched SiCf/SiC composites based on acoustic emission and digital image correlation

Abstract

In order to improve the reliability of damage analysis for SiC_f/SiC composites, an identification method of damage mechanism was established by combining in-situ acoustic emission (AE) and digital image correlation (DIC). The corresponding failure behavior of 2D needle-punched SiC_f/SiC composites during ambient-temperature tensile test was investigated in detail. Through a machine learning k-means algorithm, AE signals could be effectively divided into five clusters: friction and sliding, interface damage, matrix cracking, individual fiber breaks and collective fiber breaks. DIC results show that the surface strain of composites increased non-uniformly during the tensile process, and the architecture of the composites had a significant influence on the initiation and propagation of cracks. To summarize, the tensile process consisted of three stages: the elastic stage, the rapid propagation of matrix cracks, the coordinated fiber fracture within the large strain bands. The failure of composites was dominated by the limited load transferring ability of the interface.