Research on predicting axial load capacity of concrete filled steel tubular columns based on support vector machine method

The restriction between steel tubular and core concrete in CFSTs is complex and the relationship between geometric and material properties and axial compression behavior is highly nonlinear. These challenges have prompted the use of soft computing methods to predict the ultimate bearing capacity under axial compression. This Research investigates a soft computing tool using Support Vector Machine Learning method (SVM) and Experimental studies on the prediction of axial load capacity of concrete filled steel tubular columns (CFST). A large number of databases were collected from previously reported studies conducted by various researchers from various parts of globe on CFST columns and author’s experimental data of ninety specimens self-consolidating fibre reinforced CFST columns were considered for deriving two companion equations for the prediction of the axial load strength of CFST columns using Excel Program and using SVM technique. The main parameters considered in the Equation One were – Cube Strength of Concrete, Concrete Area, Steel Area, Tensile strength of Steel and in the Equation Two – same parameters with additional Non Dimensional parameter (D/t) was considered. The results from experimental data were correlated with the predicted values and Validations were made. It is also observed that the SVM method gives a better prediction than the other researcher’s derived equations. Validation showed that SVM has strong potential as a feasible tool for predicting axial load carrying capacity of CFST columns.