Numerical and experimental post-buckling analysis of slender thin-walled GLARE members subjected to compressive loading

Abstract

This study deals with a post-buckling analysis of thin-walled GLARE members subjected to axial compressive loading. Considered slender GLARE samples are a hybrid composite that consists of alternating thin layers of aluminium alloy sheets and unidirectional glass fibrereinforced prepregs. Comparative study was performed for different thin-walled, open cross–section members that included channel, Z-shape and top-hat-shaped slender sections. This study is mainly focused on assessing the performance of top-hat-shaped sections which provided the highest strength when subjected to the compressive test. Various layer arrangements of GLARE samples were investigated based on the fibres alignment in the composite layer. GLARE laminates were manufactured by autoclaving technique which provided the high-quality of multi-layered composite. Composite specimens were axially compressed in laboratory tests by means of static testing unit of Instron upgraded with Zwick/Roel control software. Electromechanical strength testing machine had a maximum capacity of 200kN for which applied screw type testing machine provided a displacement control loading. During experimental tests, deformations were measured by Aramis 3D non-contact optical equipment based on digital image correlation (DIC) which allowed investigating columns behaviour in full load range until fracture. The behaviour of thin-walled GLARE members was analysed with the main attention to postbuckling response. Simultaneously, numerical simulations by FEM were performed to predict the post-buckling equilibrium paths of the multi-layered GLARE structures modelled by SHELL element type. The non-linear problem was solved initially by the incremental Newton– Raphson procedure whereby the stiffness matrix was updated with each iteration and equilibrium paths were calculated based on nominal stress state in non-degraded structure. Secondly, the damage evolution law was introduced by the material property degradation method (MPDG). Hence, based on the damage variables the stiffness was gradually reduced in the area where failure was initiated which had a significant impact on the samples’ post-buckling response. Therefore, the influence of damage variables on the laminate's post-buckling equilibrium paths was also considered. For various GLARE lay-ups, numerical results were found to be in a good agreement with experimental tests. Acknowledgments The author has obtained funding as part of financing a doctoral scholarship from the National Science Center, No. UMO-2019/32/T/ST8/00605