Numerical study of the trailing vehicle length on train aerodynamics under crosswind

Limited by the wind tunnel size, a one-and-a-half train model is generally adopted during the test to represent the realistic long tandem. Thus, the reasonable arrangement of the trailing vehicle length is of significance for experimental accuracy. In this work, the aerodynamic performance of trains with different trailing vehicle lengths subjected to crosswind was studied using the improved detached eddy simulation method combined with the shear–stress–transport k–ω turbulence model. The 1/8th scaled high-speed trains with five trailing lengths were proposed, and the aerodynamic differences were evaluated at yaw angles ranging from β = 0° to β = 60°. The numerical method was validated by the previous wind tunnel test. Results show that the aerodynamics and flow patterns of the La = 0.50 and the benchmark are highly relevant. The downstream dummy vehicle length has the greatest effect on aerodynamic coefficients at a higher yaw angle (β > 30°), especially at β = 60°, whereas the effects become insignificant at a lower yaw angle (β < 30°). For various La lengths, notable discrepancy appears at the leeward and top side where large vortexes shed off from the roof. The larger contributions to the lateral force and lift force coefficients are mainly due to these areas. A suitable length of La = 0.50 is therefore recommended to obtain more accurate aerodynamics of a long train set.