A Neural Network Model to Predict Ultimate Strength of Rectangular Concrete Filled Steel Tube Beam – Columns


In this study, a model for predicting the ultimate strength of rectangularconcrete filled steel tube (RCFST) beam-columns under eccentric axial loads hasbeen developed using artificial neural networks (ANN). The available experimentalresults for (111) specimens obtained from open literature were used to build theproposed model. The predicted strengths obtained from the proposed ANN modelwere compared with the experimental values and with unfactored design strengthspredicted using the design procedure specified in the AISC and Eurocode 4 forRCFST beam-columns. Results showed that the predicted values by the proposedANN model were very close to the experimental values and were more accuratethan the AISC and Eurocode 4 values. As a result, ANN provided an efficientalternative method in predicting the ultimate strength of RCFST beam-columns.