On the basis of earthquake-resistant behavior experiments of high strength concrete column reinforced with concrete filled steel tube, two methods for calculating flexural strength of the column, one is based upon the method in building code for calculating the flexural strength of reinforced concrete member while the other is based upon the method of superposition, are discussed, and the momentę¢xial load relationship is also analyzed using finite element method. The results are in good agreement with the experimental ones. According to the test results, the limited values of axial compression ratio and limited values of axial force for high strength concrete column reinforced with concrete filled steel tube are proposed based on the condition of displacement ductility needed for earthquake-resistance. In the final part of the paper, the feasibility of applying neural networks to evaluate the performance of the columns is investigated. A three-layer back-propagation network is trained using the earthquake-resistant behavior experimental data of the columns to predict the ductility of the columns. The predicted results agree well with the test results. |