A probabilistic constitutive model for closed-cell foams

In the homogenization analysis of closed-cell foams for determination of their effective properties, the generation of numerical models which reproduce the real microstructure is very complex and time extensive. For this reason, the present study deals with the definition of a probabilistic constitutive model which makes it possible to compute the effective stiffness components and the corresponding scatter band widths without modeling the real microstructure. However, probability distributions of the relative density, the cell size, cell shape and orientation, as the most essential microstructural variables influencing the effective stiffness, are required as the input database. These material characteristics can be easily determined by computed tomography processes. Furthermore, the influence of these distributions for the input variables on the effective material properties is investigated by a systematic variation.