Finite Element Modeling of Composite Materials using Kinematic Constraints

The purpose of this article is to present simulations of the behavior of composite materials based on kinematic restrictions among the fibers themselves and among fibers and the surrounding resine. In the literature review the authors have found that the kinematic restrictions have not been fully exploited for modeling composite materials, probably due to their high computational expense. The purpose of this article is to show the implementation and results of such a model, by using a Finite Element Analysis of geometric restrictions prescribed to the resine and fiber nodes. Closed analytic descriptions on behavior of layered composite materials are very rare. Many approaches to describe layered composite material are based on the theory of functions [formula] and [fórmula], such as the Classic Layer Theory (CLT). These theories of functions contain strong simplifications of the material, in particular for woven composites. A hybrid approximation to model composite materials with Finite Elements (FEA) was developed by Sidhu and Averill and adapted by Li and Sherwood for composite materials woven with glass polypropylene. The present article presents a method to obtain values for the properties of the composed materials. Such values are used to simulate the reinforced woven fibers by applying layered elements in the ANSYS software. Our model requires less simplifications as compared with the theories [formula] and [fórmula]. In the present article, differing from the model Li and Sherwood, the weaving model is geometrically simulated. A Boundary Representation (B-Rep "Hand" model) with genus 1 (with complex geometry) was used to apply geometric restrictions to the layers resine, fiber, etc., showing to be appropriate to simulate complex structures. In the future, non linear properties of materials are to be considered, as well as to perform the required experimental work