Ultimate capacity prediction of axially loaded CFST short columns

Composite columns have superior strength and ductility performance, and they have become more widely accepted in the engineering applications. However, the filled tubular columns require more attention. This study aims to present a new formulation for the axial load carrying capacity (N (u) ) of circular concrete filled steel tubular (CFST) short columns having various geometrical and material properties. Although there have been some empirical relations for predicting N (u) in the literature, genetic algorithm based explicit formulation is not available. In the current study, 314 comprehensive experimental data samples presented in the previous studies were examined to prepare a data set for training and testing of the prediction model. The prediction parameters were selected as outer diameter of column (D), wall thickness (t), length of column (L), compressive strength of concrete (f (c) ), and yield strength of steel (f (y) ). The prediction model was obtained by means of gene expression programming (GEP). The proposed model was compared with available ones presented in the current design codes (ACI, Australian Standards, AISC, AIJ, Eurocode 4, DL/T, and CISC) and some existing empirical models proposed by researchers. The prediction performance of all models were also evaluated by the statistical parameters. The results indicated that the GEP model was much better than the available formulae, yielding higher correlation coefficient and lower error.