Key points are not available for this paper at this time.
Predicting noise from the actual train model by numerical simulation and wind tunnel testing requires appropriate corrections based on reduced model scale and airflow velocity. This paper utilises the Improved Delayed Detached Eddy Simulation method (IDDES) in conjunction with the Ffowcs Williams-Hawkings equation (FW-H) for simulating flow fields and predicting noise. The subdomain models of high-speed train head and leading bogie were used to investigate the effects of the model scale ratio (λ) and the airflow velocity ratio (γ) on the aerodynamic noise. The results show that compared with the baseline condition (case0), only reducing the model scale (case1) will lead to an increase in local viscous flow, thereby weakening the flow velocity and surface pressure fluctuation intensity in the bogie region, and the overall noise level is slightly reduced, with a difference of less than 1 dB. However, as expected, there is a significant frequency shift in the noise spectra. For case2, the model scale and the airflow velocity are reduced simultaneously to ensure the same Strouhal number, and the vortex structure and sound source distribution characteristics are closer to case0. The noise spectra have no frequency shift, but the noise level is reduced by 60 log10(γ).
Qin et al. (Fri,) studied this question.