The property study of concrete-filled steel rib and tube column based on artificial neural networks

In this paper, a new style of concrete-filled steel rib and tube column was made and though a series of nonlinear full-range numerical calculation, some important factors were analyzed, such as the strength of the core concrete, the steel tube and the angle steel, the volume ratio of the steel tube and the angle steel, position coefficient. RBFNNs were employed for calculating the strength of the concrete-filled steel rib and tube column, and the prediction results were compared with theoretical formula calculation results. The maximum and minimum error ratio of prediction were 12.56% and 4.62%, respectively. © 2011 ISSN 1881-803X.