Abstract The microstructure of short fiber‐reinforced composites is probabilistic due to the manufacturing process. The induced spatial fluctuation of the material properties cannot be covered by a homogeneous modeling concept since this technique leads to a deterministic representation of the material. In this study, a numerical approach is presented that includes the fluctuation of the material properties on the component level. The modeling approach consists of two steps. First, the probabilistic characteristics of the microstructure are derived by numerical simulation on the mesoscale. The obtained effective material properties and the correlation structure of the stiffness matrix elements are used to discretize second‐order Gaussian random fields for the material representation on the component level in a second step. To validate all results nanoindentation and tensile tests are conducted. It is shown that the suggested approach leads to excellent results comprising the overall structural response as well as the spatial fluctuation of the material properties. Hence, the introduced technique is suitable for modeling the linear‐elastic behavior of short fiber‐reinforced composites.
展开▼
