Multi-scale framework for modelling carbon fibre weaves based on stochastic reinforcement geometry

The internal geometry of textile composites is subjected to a significant amount of variability. An improved assessment of the quality of any composite material is achieved by identification and simulation of the inherent uncertainty in the reinforcement geometry. By following this strategy, many random realistic representations of the textile reinforcement can be generated based on experimental data of only a few textile samples. This work presents such a roadmap with its successive steps to characterise the spatial scatter in the internal structure of any textile composite (short- and long-range) and simulate random models possessing the measured statistical information on average. Three main steps can be distinguished: (1) collection of experimental data and statistical analysis, (2) stochastic multi-scale modelling of the reinforcement, and (3) construction of virtual specimens in a geometrical pre-processor such as WiseTex. The methodology is demonstrated for a typical carbon-epoxy 2/2 twill woven composite produced by resin transfer moulding, but is applicable to any woven textile or other topology with minor modifications.status: publishe