Keywords: high-speed trains, ground modeling, aerodynamic noise, noise source, acoustic field, frequency band.

To investigate the impact of different ground modeling approaches on the aerodynamic noise characteristics of high-speed trains, this study adopts a numerical simulation method based on Delayed Detached Eddy Simulation and Ffowcs Williams-Hawkings equations. Aerodynamic noise sources and sound field characteristics of a simplified three-car high-speed train are compared between flat-ground modeling and refined bridge modeling. The results show that bridge modeling narrows the flow space beneath the train, leading to fragmentation and lateral overflow of vortex structures, thereby modifying the noise spectrum distribution. Under flat-ground modeling, sound energy is concentrated in the mid-frequency range (250-1600?Hz) with higher source intensity, while bridge modeling disperses sound energy toward both low and high frequencies, resulting in an overall reduction in noise level. The study further reveals that the underbody region serves as the main noise contributor. Although bridge modeling reduces overall noise by decreasing the underbody-to-rail gap, the fragmented vortex structures simultaneously enhance low- and high-frequency noise components. In terms of far-field noise, bridge modeling yields lower sound pressure levels, however, the difference between bridge and flat-ground modeling gradually widens downstream, reaching a maximum of approximately 4?dB. This variation is attributed to the combined effects of mid-frequency sound energy peaks and high-frequency sound energy contributions.

download the full-text of this paper (PDF, 19 pages, 1303 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description