Numerical study of snow accumulation on the bogies of a high-speed train using URANS coupled with discrete phase model

Abstract

The snow accumulation on bogies of a three-car grouping high-speed train has been investigated using unsteady Reynolds-Averaged Navier-Stokes simulations (URANS) coupled with Discrete Phase Model (DPM). The accuracy of mesh resolution and methodology of computational fluid dynamics (CFD) are validated against experimental results. The characteristics of flow fields around bogie regions, the snow particles' motion and snow accumulation are analysed herein. Also the computational costs of numerical simulations using URANS and detached eddy simulation (DES) in the present study are compared. The results show that snow particles begin to flow out from the high-speed train bottom as they come to the first bogie region. After that they whirl around, forming a wind-snow flow in the wake region. Bogie 2 experiences the most snow accumulation. Subsequently, the snow covering on the bogie regions decreases along the train length. A comparison of snow accumulation between URANS and DES methods is made, showing the URANS is sufficient to predict this snow accumulation, which only takes up 2/3 computational cost of the DES. The geometry model consisting of a leading vehicle and a 1/3 s vehicle is recommended to investigate the snow accumulation issue on the bogies of the train.