Fabric Simulation

Finite element simulation of the composite fabrics predicts mechanical behavior of the textile and the precipitation of defects. These information can be used to investigate variations of manufacturing parameters to minimize defects of cured composites therefore providing a link between manufacturing process parameters and part quality. Modeling of the composite reinforcement is accomplished at the mesoscopic scale using a discrete approach with a non-homogeneous material model. This simulation technique is therefore capable of representing a variety of fabric architectures. Standard element types can be used and the fabric behavior could be captured through user defined material subroutines. Specifically, beam elements incorporate the tensile and flexural properties of the fibers while shell elements define the shear properties of the composite reinforcement.

 

Video #1 – Forming

This video shows the deep-draw forming of a fiber-reinforced composite hemisphere. Modeling the composite forming process using the finite element method offers the advantage of incorporating complex boundary conditions and material behaviors into the mechanical analysis. In doing so, the model is able to predict the precipitation of defects. Therefore, the simulation can be used to link the quality of the structural composite to the manufacturing process, and a virtual design-build-test regimen can be employed to investigate variations of manufacturing parameters to minimize defects in the composites part.

 

Video #2 - Compression

This model was used to validate proper implementation of the textile's flexural behavior. The simulation replicates an experimental configuration designed to induce out-of-plane deformation of a fabric strip. The sample is continuously compressed along the axis of the fibers, and secondary compressive forces are applied normal to the fabric face once a certain wave amplitude is reached. Comparisons between FE and experimental results are used to confirm that the bending behavior is well represented in the simulation.