Using microscale SPH-simulations to parameterize macroscopic fiber orientation models for discontinuous fiber reinforced polymers

Abstract The orientation of discontinuous fibers in injection or compression molded parts has crucial impact on the thermo-mechanical properties of the part. Therefore, the prediction of these orientation states is of high importance and has been the focus of research for several years. Today’s models often represent the orientation evolution by a semi-empirical tensor evolution equation, which needs at least one empirical parameter. More complex models need more parameters. The determination of these parameters often entails a high experimental effort, especially if data over time is needed. This work presents a multiscale approach, where microscale simulations of individual and interacting fibers in a simple shear flow are used to parameterize macroscopic orientation models. The microscale simulations are performed with the smoothed particle hydrodynamics method. The macroscopic models, being the Folgar-Tucker and ARD-RSC model, are fitted according to the microscale results and are compared to state-of-the-art approaches for parameter determination. For experimental validation, injection molding trials of a 20 weight-% short glass fiber reinforced phenolic are used, showing the benefits of the micro–macro coupling especially when predicting the orientation evolution over time. The main advantage of the multiscale approach is the parameterization of the ARD-RSC approach without the need of experimental data for a wide range of fiber-polymer materials.