Abstract

Modeling the nonlinear material behaviour of long fiber reinforced thermoplastics (LFT) presents a challenging task since local inhomogeneities and nonlinear effects must be taken into account also on the microscale. We present a computational method with which we can predict the nonlinear material response of a composite material using only standard DMA measurements on the pure polymer matrix material. The material models considered include plasticity, damage, viscoelasticity, and viscoplasticity as described in [1]. These models can be combined similar to the model from [2] and extended to the composite by assigning linear elastic properties to the fibers. The mechanical response of the composite is computed using an FFT-based technique [3]. The geometry of the composite, in particular the fiber orientation, can be characterized using injection molding simulations or micro CT scans. We create virtual models of the composite using the algorithm of [4]. We show that with this method, the material behaviour of the composite can be predicted while the experimental complexity needed for the material characterization is low.