Virtual testing against experiment for post-buckling behaviour of coldformed steel columns

Abstract

Cold-formed steel has already started to replace hot rolled companions in some structural applications. Advantages of cold-formed steel originate from its high strength over weight ratio and ease of manufacturing and construction compared to hot rolled heavy sections. Moreover, cold-formed columns have significant post-buckling reserve which has the potential to be exploited in design process. Therefore, it is essential to predict the response of cold-formed columns by means of high fidelity engineering techniques. Herein an in depth study which links experimental testing and non-linear computational capabilities is undertaken to address the failure behaviour of cold-formed columns. Experimental program comprises coupon tests to specify material properties and compression testing of fixed end cold-formed columns. Thereafter, measured material properties are utilized to generate a stress-strain curve for finite element models. Boundary conditions imposed into simulation models in such a way that would represent test conditions. Creating a suitable mesh for different cross sectional dimensions, different shapes of initial imperfections are introduced into models to compare contributions to performance of columns. Predicted collapse loads and modes via finite element models are assessed against test results. Mesh and initial imperfection sensitivities on failure characteristics are discussed. Finally a general assessment is made for the deployed testing and simulation to generate knowledge for the design evaluation of cold-formed steel columns. Key findings and discussions of present study have the potential to lead to develop promising cold-formed steel column virtual test models.