Computational fluid dynamics prediction of the aerodynamic difference between stationary and moving trains

Abstract

Moving model simulations have been a key method of predicting the aerodynamic performances of High-Speed Trains (HSTs). Ideally, the aerodynamic characteristics of a train moving or being blown by the wind are the same with appropriate ground configurations. In a numerical simulation, there are differences due to interpolation errors and errors caused by model movement. The impact of the error caused by the movement on the result is not known. Therefore, in this study, stationary and moving cases were used to assess the magnitude of the movement’s effect using the Improved Detached Eddy Simulation (IDDES) method. A wind tunnel test validated the numerical algorithm at 60 m/s and a common yaw angle of 0°. Moreover, the spatial and time discretization satisfied the high accuracy requirements, as determined through a mesh independence study and convective Courant number testing. The time-averaged drag coefficients predicted by the moving case were similar to those of the stationary case, especially the total drag coefficients. In contrast, differences were determined in the stationary and moving cases in terms of the flow structure and slipstream. The motion encouraged the streamwise vortices around the tail car and the wake vortices to expand along the spanwise direction and the wall-normal direction, and the vortex cores shifted away from the outer surface of the vehicle. As a consequence, the average value and the standard deviation of the slipstream increased. Therefore, moving model simulations require more caution. These findings can help researchers make directional corrections in the numerical simulation of train-tunnel systems.