Computational Modeling and Comparison with Experiments on 3D Permeability in Vacuum Assisted Resin Infusion of Glass-Fabric Reinforced Composite Laminates

Abstract

Listen

Translate article

Vacuum-assisted resin infusion molding has been increasingly used in manufacturing large composite structures, such as MW-scale wind turbine blades, due to structural size, cost-effectiveness, and out-of-autoclave curing requirements. Among various material and processing parameters, 3D permeability of composites with thicksection laminate construction is critical in controlling resin flow in an efficient manufacturing process development by accurate numerical modeling and simulation. Reactive resin flow in a mold with complex geometry and material systems, such as a wind turbine blade with a foam core and multiple fabric orientations, is strongly affected by the 3D permeability of the fabric composite. An experimental method to measure the orientation and the magnitude of individual principal component of permeability tensor for thick-section fabric reinforcements has been addressed in a previous study. The objectives of this research are to: (1) use the 3D permeability measurements from the previous experimental study as inputs into ANSYS Fluent’s volume-of-fluid model to simulate flow front progressions, (2) compare the simulated flow front progression with experimental results, and (3) identify the sources in the experimental and numerical methods that contribute to the discrepancies between experimental and numerical results. A brief review is provided on the 3D permeability measurement theory and the experimental method and results from the previous study. Numerical simulations of the flow front progression of a Newtonian fluid into an isotropic porous media are first compared to analytical results to select the computational models and discretization schemes used in this study. The choice of time-step and mesh discretization is discussed along with a convergence study. Insight into the limitations of the 3D permeability measurement theory and numerical methods is provided by comparing simulation results with circular and spherical inlets. Suggestions are made on methods to overcome incompatibilities of assumptions on inlet geometry between the in-plane and throughthickness permeability measurement theories.