Steel-concrete Π -shaped cross-section is prevalent in the design of cable-stayed bridges. Nevertheless，the vortex-induced vibration（VIV）is prone to occur for this section type，which can cause structural safety problems and reduce the driving comfort. In this paper，the results of the wind tunnel tests of Π-shape prototype deck are used to certify the script in numerical simulation of vertical VIV. Then，the data set could consist of the results from wind tunnel tests and computational fluid dynamics（CFD），which is used to describe the relationship between the two shape parameters―aspect ratio and aperture ratio，and responses of vibration. The radial basis function artificial neural network is trained by the learning sample，and the setting parameters of artificial neural network should be optimized to improve the precision to study the mechanism of the VIV in the Π -shape prototype deck. The results indicate that the responses of VIV have non-linear relationship with two shape parameters. And the relationship can be used to guide the selection and optimization of the aerodynamics control measures.