Two-scale damage failure analysis of 3D braided composites considering pore defects

Lei Ge, Huimin Li*, Huayong Zheng, Chun Zhang, Daining Fang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

41 Citations (Scopus)

Abstract

Pore defects tend to form in the manufacturing process, and significantly influence the mechanical properties of 3D braided composites. To predict the failure modes and strength properties considering pore defects, a two-scale progressive damage method coupled with representative volume cell (RVC) has been developed due to the periodic feature across different scales of 3D braided composites. Different void generating methods across two scales are established, including the element-model by elements chosen and the void-model by voids explicitly constructed. The user-defined material subroutine (UMAT) is coupled in the nonlinear finite element analysis software ABAQUS to implement the damage model for micro and meso numerical models. Also, the Chamis model considering pore defects are employed and verified. It shows that results by void-model agree well with the Chamis model and the experimental data. And, the void-model is superior to the element-model when predicting the mechanical properties of braided composites. The pore defects weaken the matrix properties, and therefore largely determine the strength and damage mechanism of 3D braided composites with large braiding angle. This two-scale damage failure analysis method by void-model is effective to predict strength and damage failure mechanism of 3D braided composites considering pore defects.

Original languageEnglish
Article number113556
JournalComposite Structures
Volume260
DOIs
Publication statusPublished - 15 Mar 2021

Keywords

  • 3D braided composites
  • Damage failure
  • Pore defects
  • Strength
  • Two-scale analysis

Fingerprint

Dive into the research topics of 'Two-scale damage failure analysis of 3D braided composites considering pore defects'. Together they form a unique fingerprint.

Cite this