Prediction of high-speed train equivalent passing noise based on numerical analysis of aerodynamic noise

In the train shape design stage, train passing noise is usually used in engineering to evaluate line noise to guide aerodynamic appearance design. It is expensive and challenging to measure passing noise in practical vehicle tests. Therefore, the aerodynamic characteristics were analyzed based on numerical model in this work, and subsequently the moving sound source delay model was used to predict the passing noise, which provided theoretical and technical support for the evaluation of the external aerodynamic noise in stage of the train shape design. The calculation grid of train aerodynamic noise was relatively large for capture the fluid disturbance on the train surface, so a three-car train grid model was established to simplify the calculation. The aerodynamic characteristics of high-speed train were investigated by using software ANSYS FLUENT, turbulence model based on the steady-state and large eddy simulation model. Then the radiated aerodynamic noise of the standard point along the line outside the train was obtained through the Ffowcs Williams-Hawkins (FW-H) equation. According to the moving sound source delay time model, a train equivalent passing noise conversion model was established to convert the train radiated noise in the frequency domain at standard points into the equivalent passing noise in the time domain. The passing noise level of the train with different formations was further predicted by the superposition of sound sources. It was revealed that the increase in the number of formations will increase the noise level. To comprehensively evaluate the passing noise level during train operation, the wheel-rail noise was considered. The passing noise level was represented by the maximum A sound level (LAmax), the sound exposure level (SEL) and the short-term equivalent continuous sound pressure level (LAeq, T), which demonstrated that the wheel-rail noise had a significant impact on the passing noise. The prediction model of passing noise can initially evaluate the noise level of high-speed train operation lines, and provide guidance for optimizing the shape of the train to reduce noise. © 2024, Central South University Press. All rights reserved.