Virtual characterization of composite materials for aero-engine components

University essay from Luleå tekniska universitet/Institutionen för teknikvetenskap och matematik

Abstract: Since its beginnings, the aerospace industry has been interested in lowering the weight of aircraft. Moving from performance and economic drivers to environmental design parameters, the weight has continuously been a major focus for this industry. A possible option to reduce weight is to use lighter materials such as fibre reinforced polymer composites (FRPC). This type of material has the potential to be used into cold or moderate high-temperature sections of aero-engines. One major obstacle that hinders composite insertion into aero-engines is the lack of predictive models. In recent years, there has been increasing interest in multiscale modelling as a possible approach to reliably predict composite behaviour. This modelling refers to the simulation of a material’s behaviour through multiple scales, passing on information from one scale to another. The purpose of the present work is to use a commercially available software tool (Altair Multiscale Designer™) to virtually characterize an FRPC made from a non-crimp fabric reinforcement based on its individual constituent properties. The studied composite was a carbon fibre and epoxy system developed by GKN Aerospace. In order to achieve this, a well-characterized unidirectional (UD) carbon fibre prepreg composite was used to calibrate the software. After calibration and verification, different repetitive unit cells were created to capture the non-crimp fabric (NCF) architecture where the effect of fibre waviness was studied. The calibration step allowed for fairly accurate and acceptable results when testing unidirectional or ±45 laminates with different tested UD prepreg material systems. The higher deviation from experimental values was up to 20% with these laminates’ configurations. When simulating more complex layups, such as quasi-isotropic ones, the simulations resulted in over-predicting up to 40% of the composite strength in comparison to experimental data. The study of NCF composites appeared to be more complicated than anticipated. Their complex architecture exhibits complicated failure modes, which could not be captured by the software tool. Large inaccuracy up to 100% were observed between simulation and experimental values of the laminate strengths. In spite of its limitations, the study of NCF composites allowed for a deeper understanding of the software functionalities and findings on the fibre waviness impact onto the predicted stiffness, while the strength of the laminate did not show dependency with the fibre waviness.

  AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)