Practical artificial neural network tool for predicting the axial compression capacity of circular concrete-filled steel tube columns with ultra-high-strength concrete

Abstract

This paper aims to develop a practical artificial neural network tool for predicting the axial compression capacity of circular concrete-filled steel tube columns with ultra-high-strength concrete. For this purpose, a nonlinear finite element analysis of circular concrete-filled steel tube columns with ultra-high-strength concrete was conducted and verified with experiments in the literature. Accordingly, a database of 768 finite element models was generated to use for developing the artificial neural network models. In this regard, the column length, the diameter of steel tube, the thickness of steel tube, yield and ultimate strength of steel tube, and compressive strength of concrete were considered as the input variables while the axial compression capacity was considered as an output variable. The performance of the proposed artificial neural network model was compared with the current structural design codes including AS/NZS 5100.6, Eurocode 4, AISC, and GB 50936. The comparative study indicated that the proposed artificial neural network model achieved a superior prediction compared to others. Ultimately, a graphical user interface tool was developed based on the proposed artificial neural network model to predict the axial compression capacity of circular concrete-filled steel tube columns with ultra-high-strength concrete for practical engineering design. • The proposed ANN model can accurately predict the axial compression capacity of CCFST columns with UHSC. • The proposed ANN model shows more accuracy than four notable existing design code formulas. • A GUI tool is developed based on the proposed ANN model for practical engineering design.