open access publication

Article, 2024

An accurate forming model for capturing the nonlinear material behaviour of multilayered binder-stabilised fabrics and predicting fibre wrinkling

Composites Part B Engineering, ISSN 1879-1069, 1359-8368, Volume 274, Page 111268, 10.1016/j.compositesb.2024.111268

Contributors

Broberg, Peter Hede 0000-0002-8115-7186 (Corresponding author) [1] Lindgaard, Esben 0000-0002-8253-2419 [1] Thompson, Adam J [2] Belnoue, Jonathan P-H 0000-0003-4477-3470 [2] Hallett, Stephen R 0000-0003-0751-8323 [2] Bak, Brian Lau Verndal [1]

Affiliations

  1. [1] Aalborg University
  2. [NORA names: AAU Aalborg University; University; Denmark; Europe, EU; Nordic; OECD];
  3. [2] University of Bristol
  4. [NORA names: United Kingdom; Europe, Non-EU; OECD]

Abstract

Prefabricated stacks of binder-stabilised dry fabrics are used to increase reinforcement deposition rates in the manufacturing of large composite structures such as wind turbine blades. To ensure optimal part quality, accurate simulation tools may be used to avoid costly trial-and-error experiments. This paper introduces a new modelling framework for predicting wrinkle defects that arise from forming thick binder-stabilised stacks of quasi-unidirectional non-crimp fabric. In this context, quasi-unidirectional non-crimp fabrics are unidirectional rovings stabilised with a backing layer. The non-constant bending stiffness of the fabric is modelled using an asymmetric compressive modulus, while a mixed-mode cohesive formulation with user-defined damage evolution is used for modelling the initial non-constant elastic binder interface stiffness and interface degradation. The model is verified against cantilever bending tests of single fabric layers and transverse shear experiments of bindered fabric layers. Comparison with experimental results on the forming of preforms shows that the model accurately predicts wrinkle location and geometry during the forming of thick binder-stabilised preforms. The error between the wrinkle geometry predicted by the new modelling approach and the experimentally measured wrinkle geometry is reduced by 35% to 94% compared to wrinkles predicted by conventional modelling approaches. The modelling framework is freely available at https://doi.org/10.5281/zenodo.10077428.

Keywords

accurate simulation tools, approach, back, backing layer, behavior, bending stiffness, bending tests, blade, cantilever, cantilever bending test, cohesive formulation, comparison, composite structures, compressive modulus, context, damage, damage evolution, defects, degradation, deposition rate, dry fabrics, error, evolution, experimental results, experimentation, experiments, fabric layers, fabrication, fibre wrinkling, formulation, framework, geometry, interface, interface degradation, interface stiffness, layer, location, manufacturing, material behavior, model, modeling approach, modeling framework, modulus, non-crimp fabrics, nonlinear material behavior, optimal part quality, part quality, preform, quality, quasi-unidirectional non-crimp fabric, rate, results, roving, shear experiments, simulation tool, stiffness, structure, test, tools, turbine blades, wind, wind turbine blades, wrinkle defects, wrinkle geometry, wrinkles, wrinkling location

Funders

  • Engineering and Physical Sciences Research Council

Data Provider: Digital Science