Studying buckling behavior of large shell structures through full-scale test articles can be complex and expensive. Therefore, reduced-scale structures are often preferred for investigating buckling behavior. However, designing reduced-scale structures that are representative of the full-scale structure can be difficult. An analytical scaling methodology for compression-loaded sandwich composite cylindrical shells based on the nondimensionalization of the buckling equations is presented herein. The methodology was used to develop scaled configurations that show similar buckling responses to the full-scale baseline configuration. Finite element analysis results showed that both a baseline and a scaled configuration buckled similarly, when the nondimensional stiffness, defined as the ratio between the nondimensional load and nondimensional displacement, was matched between the different scale models. Limitations of the methodology are discussed and are believed to be a result of neglecting the flexural anisotropy and the transverse shear compliance. A preliminary material failure assessment for the different scales is also considered.@en

The study of the buckling behavior of large shell structures through full-size tests can be complex and expensive. Therefore, scaled structures are often preferred to investigate the buckling behavior efficiently. However, it can be difficult to design scaled structures that are representative of the full-scale structures. Herein, an analytical scaling methodology for compression-loaded sandwich composite cylinders based on the nondimensionalization of the buckling equations is presented. The methodology is used to develop scaled configurations that show a similar buckling response. Both the baseline and the scaled configurations are verified by finite-element analysis. Limitations of the methodology are discussed and are a result of neglecting the flexural anisotropy and the transverse shear compliance.@en

The important role of geometric imperfections on the decrease of the buckling load for thin-walled cylinders had been recognized already by the first authors investigating the theoretical approaches on this topic. However, there are currently no closed-form solutions to take imperfections into account already during the early design phases, forcing the analysts to use lower-bound methods to calculate the required knock-down factors (KDF). Lower-bound methods such as the empirical NASA SP-8007 guideline are commonly used in the aerospace and space industries, while the approaches based on the Reduced Stiffness Method (RSM) have been used mostly in the civil engineering field. Since 1970s a considerable number of experimental and numerical investigations have been conducted to develop new stochastic and deterministic methods for calculating less conservative KDFs. Among the deterministic approaches, the single perturbation load approach (SPLA), proposed by Hühne, will be further investigated for axially compressed fiber composite cylindrical shells and compared with four other methods commonly used to create geometric imperfections: linear buckling mode-shaped, geometric dimples, axisymmetric imperfections and measured geometric imperfections from test articles. The finite element method using static analysis with artificial damping is used to simulate the displacement controlled compression tests up to the post-buckled range of loading. The implementation of each method is explained in details and the different KDFs obtained are compared. The study is part of the European Union (EU) project DESICOS, whose aim is to combine stochastic and deterministic approaches to develop less conservative guidelines for the design of imperfection sensitive structures.

@en