Micro-geometry modeling based on Monte Carlo and permeability prediction of yarn

In order to simulate the resin filling process of Liquid Composite Modeling (LCM) such as ResinTransfer Molding (RTM) effectively, the permeability tensor field of preform is needed to be predicted accurately in advance. In the several existing micro-permeability prediction models, the fibers are assumed to be regularly arranged (quadratic or hexagonal) to simplify the flow analysis, they all ignore the randomness of the fiber distribution in yarn. The micro-permeability is a preform property depending strongly on the preform's micro-geometry, so building the micro-geometry close to the real structure is significant to predict the micro-permeability of preform. In this work, we present a Monte Carlo moving method to obtain the fiber random distribution on the front and rear cross sections of yarn segment, then build the geometry model of the yarn segment, adjust the center path line of fiber model by Bezier method, establish the micro-geometry model of yarn close to the real structure containing both axial and radial random factors. The micro-geometry model consisting of ∼100 fibers is built using the mentioned method, the fiber volume fraction varies from 0.15 to 0.90, and CFD simulations are carried out to determine their permeability values. The results show that compared to the regular fiber arrays, the disordered fiber arrays have a greater axial permeability, and two special transverse flow effects emerged as the disorder of fiber distribution, named local channeling effect and local obstructing effect. Relative to the theory predicted transverse permeability, the simulated permeability increases in the high porosity yarn as the channeling effect dominates the liquid flow, vice versa, the simulated permeability decreases in the low porosity yarn since the obstructing effect dominates the liquid flow, the simulation results agree well with the experimental ones, indicating the validity of the micro-geometry modeling method.