An integrated computational framework for simulating the failure response of carbon fiber reinforced polymer composites

A new computational framework is introduced for the automated finite element (FE) modeling of fiber reinforced composites and simulating their micromechanical behavior. The proposed methodology relies on a new microstructure reconstruction algorithm that implements the centroidal Voronoi tessellation (CVT) to generate an initial uniform distribution of fibers with desired volume fraction and size distribution in a repeating unit cell of the composite. The genetic algorithm (GA) is then employed to optimize locations of fibers such that they replicate the target spatial arrangement. We also use a non-iterative mesh generation algorithm, named conforming to interface structured adaptive mesh refinement (CISAMR), to create FE models of the CFRPC. The CVT-GA-CISAMR framework is then employed to investigate the appropriate size of the composite's representative volume element. We also study the strength and failure mechanisms in the CFRPC subject to varying uniaxial and mixed-mode loadings.