Application of machine learning models for designing CFCFST columns

Abstract

In this study, two machine learning (ML) algorithms including support vector regression (SVR) and artificial neural network (ANN) are employed to predict the ultimate strength of rectangular and circular concrete-filled cold-formed steel tubular (CFCFST) columns under concentric and eccentric loading. In total, 730 test results on CFCFST columns are compiled and used to train the algorithms. In addition, 720 rectangular and circular CFCFST columns subjected to concentric and eccentric loading are modelled and analysed using finite element (FE) method to expand the training data. The accuracy of the developed FE models is verified by comparing the simulation results with existing experimental..