Multiscale Study on Compressive Failure Mechanism of Plain Woven Composites Considering Stochastic Waviness Defects

Jie Sun,Hongneng Cai,Yaping Huang,Erjin Yun
DOI: https://doi.org/10.1016/j.compstruct.2024.118600
IF: 6.3
2025-01-01
Composite Structures
Abstract:This work proposes a multiscale correlation algorithm based on structural level to study the compression damage behavior of plain woven composites. The multiscale algorithm includes a homogenization process that establishes the mesoscale constitutive relationship based the mechanical properties of microscopic constituent to derive the macro-mechanical response, and a localization process that constructs a mapping relationship among mesoscale stress, microscale stress and micro-component failure based on the macroscale load. The results of uniaxial compression experiments are used to validate the model. The multiscale model combined with micro-mechanics of failure (MMF) and 3D kinking model accurately predict the effective stiffness, ultimate compressive strength, damage modes and fracture characteristics. With increasing compression load, the matrix damage of yarn, fiber kinking and pure matrix damage successively initiate. Furthermore, the effect of yarn waviness defects deviating from the ideal design on compressive mechanical behavior is parametrically investigated based on the hypothesis of Gaussian distribution for defects. The increase of waviness misalignment angle (the difference between actual and ideal waviness angle) and its dispersion degree reduces the compression modulus and failure strength. Therefore, the influence of stochastic waviness defects on mechanical response cannot be ignored in structural level design and application of woven composites.
What problem does this paper attempt to address?