A simplified method for free vibration analysis of cable-stayed bridges

Abstract

A simplified method for the accurate prediction of the natural frequencies of cable-stayed bridges is proposed in this paper. In the proposed method artificial neural network (ANN) is applied to derive a simple formula to predict the natural frequencies of cable-stayed bridges based on existing natural frequency data. Unlike in the existing empirical methods, no functional relationship among the variables is assumed before we can develop an ANN model. ANN automatically constructs the relationships and adapts based on the training data presented to them. Also, the proposed method takes into account the wide range of parameters which may have a significant effect on the natural frequencies of cable-stayed bridges. The proposed method is particularly useful for the preliminary design stage of cable-stayed bridges where free vibration analysis needs to be carried out. The proposed method is compared with two existing empirical methods. It is found that the simplified method proposed in this study can produce a more accurate prediction of the natural frequencies of cable-stayed bridges than the existing empirical methods.