Transformer-Based Deep Learning Framework for Efficient Prediction of Automotive Wind Noise

Aerodynamic wind noise is a critical challenge in modern automotive development, particularly with the rise of vehicle electrification and intelligent mobility, where cabin acoustic comfort is a key quality metric. While reliable, traditional methods like wind tunnel experiment and CFD simulations are both costly and time-consuming. To address this, we propose a novel transformer-based framework for rapid and accurate wind noise prediction. Several model improvements, including the physical attention, geometry wave number embedding, hybrid FPS-random down sampling method and frequency separation output heads are properly employed to reduce the GPU mermory cost and improve the prediction accuracy. This framework is pre-trained on a large-scale transient flow field dataset of 1,000 diverse vehicles generated using Delayed Detached Eddy Simulation (DES). From a vehicle's point-cloud coordinates, the model directly predicts the surface pressure spectrum on the driver-side window and the corresponding in-cabin Sound Pressure Level (SPL). Subsequently, the model is fine-tuned using experimental wind tunnel data for enhanced accuracy. The validated model demonstrates exceptional performance across various vehicle types, including sedans, SUVs, and MPVs, achieving a mean absolute error of less than 1 dBA and a maximum error of less than 5 dBA in under one second on the validation and test sets. This approach significantly enhances efficiency, offering a possible solution that reduces development costs and accelerates the design cycle in automotive wind noise engineering.