On efficient modelling of progressive damage in composite laminates using an equivalent single-layer approach

Licentiatavhandling, 2016

In order to reduce the weight and subsequently the fuel consumption of their vehicles, the automotive industry is currently very active in research to incorporate laminated composites of Carbon Fibre Reinforced Plastics (CFRP) into structural components. This effort is a reaction to the increasing demands from regulatory bodies aimed at reducing the CO2 emissions from the transportation sector. Historically composite materials have mainly been used in the aerospace industry, whereby CAE-based design and development tools for composite structures have been developed primarily to the specific needs and requirements in this industry. Even if many of the methodologies used to develop aerospace composite structures are directly transferable to the automotive industry, the assessment of crashworthiness of automotive vehicles has no clear equivalence within aerospace. Thus, it is crucial to develop numerical tools which are able to asses the crashworthiness performance of vehicles made of composite materials. However, in order to be usable in the automotive product development process these tools must be both computationally efficient and be able to make an accurate prediction of the crash response. For an accurate prediction, predominant failure mechanisms such as delaminations must be captured by the simulation models. In this thesis, we will present a route to full scale vehicle crash simulations based on a computationally efficient adaptive methodology. To model the delamination propagation in the laminates, we propose an Equivalent Single-Layer (ESL) shell formulation with adaptively refined delamination enrichments using the eXtended Finite Element Method (XFEM). By combining the formulation with a stress recovery technique, we can increase the accuracy of the transverse stress distributions of the shell model, and thus the prediction of delaminations. In this way, we balance the delicate issues of model accuracy and efficiency in the context of progressive delamination failure analyses in laminated composites. The work presented in this thesis lays the ground work for achieving computationally efficient and accurate predictions of the crashworthiness performance of composite components in simulations. This is an absolute requirement in the automotive development process today if these materials are to be have a widespread use in future automotive vehicles.