This thesis aims to provide a kinematical representation of delamination in Fiber Reinforced Polymers (FRP) combining the accuracy and the computational efficiency necessary for the large structures.
Propagation of delamination cracks is usually treated numerically representing each one of the composite layers of the laminate and using connections in-between them which accurately models the interface degradation during delamination. This method results in as many elements through the thickness as there is layers in the composite. If the accuracy has been proven , the computational and pre-processing time required by this method is large. It makes it difficult to use in the development of large assemblies of complex components such as an automobile body-in-white. Aware of these necessities, several research teams proposed kinematic modelling using eXtended Finite Element Method (XFEM) or Phantom Nodes in order to detect delamination locally and enrich the model only where necessary. It represents a computationally efficient solution as the number of degrees of freedom increases only where delamination occurs. Moreover, it allows the initial use of one element through the thickness at the beginning of the simulation leading to no extra pre-processing cost with respect to classical modelling. Until now, such modelling has been compared to Beam-like problems in static load cases. This doctoral thesis aims to implement a similar model in a commercial FE software using in order to extend the use to larger crash simulations.
Docent vid Chalmers, Industri- och materialvetenskap, Material- och beräkningsmekanik
Finansierar Chalmers deltagande under 2018–2021