Proceedings of the 20th International Conference on Composite Materials, Paper #4417-4, 2015

SIMULATION OF THE INTERMEDIATE VELOCITY IMPACT BEHAVIOR OF WOVEN COMPOSITE LAMINATE APPLYING PROGRESSIVE DAMAGE MODELS FOR PLIES AND INTERFACES

M. Schwab, M. Todt, H.E. Pettermann

Institute of Lightweight Design and Structural Biomechanics,
TU Wien, Vienna, Austria


Abstract - A strategy for modelling impact on woven fabric reinforced laminates at intermediate velocities using the framework of the explicit finite element method is presented such that the energy absorption of the laminate can be predicted and assigned to individual failure mechanisms. Thereby a "stacked shell" approach is utilized where each ply of the laminate is discretized by an individual layer of shell elements. The damage and failure behaviour of the plies is modelled using an energy based continuum damage mechanics approach. Tensile and compressive ply damage along the fibre directions, as well as shear damage are considered. Additionally plastic deformations due to shear loads are captured. Delamination is accounted for by modelling the interfaces between individual plies using cohesive zone elements in combination with a traction-separation based constitutive law. Element deletion is utilized to cope with fully damaged elements and, furthermore, contact definitions between delaminated plies are included. Hence, the approach is capable of predicting the impact response of laminated composites up to complete perforation. In order to show the predictive capabilities of the presented modelling approach, a rectangular carbon fabric/epoxy reinforced laminated plate with quasi-isotropic layup in a drop tower test setup is modelled. The predicted energy absorption of the plate, as well as the deformation behaviour and the damage and failure behaviour are compared to equivalent experimental results. Additionally, the contributions of individual failure mechanisms to the total energy absorption are predicted. The described modelling strategy is particularly advantageous in terms computational efficiency, while maintaining detailed and quantitative predictions.
(hjb,150727)