A new method for generating random fibre distributions for fibre reinforced composites

This paper presents a new method to generate the random distribution of fibres in the transversal cross-section of fibre reinforced composites with high fibre volume fraction. The algorithm, named random sequential expansion (RSE) algorithm by the authors, is based on the hard-core model, but overcomes the jamming limit of hard-core model and is able to generate random distributions for high values of fibre volume fractions. By adjusting the inter-fibre distance parameters, it can generate microstructures with different fibre volume fractions expediently. Statistical analysis is performed on the generated distributions and a good agreement with the completely spatial random (CSR) pattern is found in all statistics analysed. Finite element analysis (FEA) is conducted to predict the effective properties of the generated microstructures. The predicted properties show reasonable agreement with the experimental results. The present algorithm is simple, highly efficient and provides a useful alternative to generate random numerical models that can be used in micromechanical analysis of composites.