Distortional elastic buckling for aluminium: Available prediction models versus design specifications
N. Kutanova1, T. Peköz2, F. Soetens1, 3
1 Eindhoven University of Technology, Eindhoven, the Netherlands
2 Cornell University, Ithaca, USA
3 TNO Built Environment and Geosciences, Delft, the Netherlands
Accurate prediction of buckling behaviour of aluminium structural elements with thin-walled cross-sectional shapes is important for efficient design. Many recent investigations in terms of cross-sectional instability have proved that distortional buckling has a substantial effect on the structural behaviour of aluminium thin-walled members. Aluminium design rules are limited regarding the distortional buckling phenomenon and simplified to meet the theoretical solutions of the plate-buckling problem. However, the plate buckling problem does not reflect the real cross-sectional instability behaviour because the edge restrained conditions of each plate element are not taken into account. In this respect, prediction models like the Direct Strength Method which is based on the elastic buckling solutions for the entire cross-section can achieve these requirements. In order to provide a reliable design approach specifically for aluminium, a new prediction model for distortional buckling of C-shaped thin-walled cross-sections has been developed (Kutanova model [2009]). This model is based on advanced non-linear finite element analyses that include the interaction of cross-sectional plate elements. In this study, the distortional bucking resistance of C-sections has been calculated according to the current design rules (AA Specification and Eurocode 9) and two prediction models: the Direct Strength Method and the Kutanova model. The calculations using these different approaches mainly show the inconsistencies in results provided by the aluminium design standards compare to the prediction models.
Key words: Distortional buckling, aluminium, thin-walled cross-sections, design specification, buckling