Generalization of Urban Wind Environment Using Fourier Neural Operator Across Different Wind Directions and Cities
Cheng Chen, Geng Tian, Shaoxiang Qin, Senwen Yang, Dingyang Geng, Dongxue Zhan, Jinqiu Yang, David Vidal, Liangzhu Leon Wang
TL;DR
The paper tackles the expensive nature of CFD-based urban wind simulations and their limited generalizability across wind directions and city layouts. It introduces an autoregressive Fourier Neural Operator (FNO) framework, trained on CityFFD LES data, that uses patch-based training and signed distance function (SDF) geometry encoding to predict near-term wind fields in urban environments. The authors show that patch-based training with SDF yields the best accuracy and dramatic speedups over CityFFD, and that wind-direction generalization improves when test data are aligned with training orientations; cross-city transfer, however, depends on geometric similarity between cities. These findings suggest that the proposed FNO approach can enable near real-time urban wind predictions for planning and energy applications, while highlighting the need for physics-informed training and broader urban datasets to improve long-term robustness and cross-city applicability.
Abstract
Simulation of urban wind environments is crucial for urban planning, pollution control, and renewable energy utilization. However, the computational requirements of high-fidelity computational fluid dynamics (CFD) methods make them impractical for real cities. To address these limitations, this study investigates the effectiveness of the Fourier Neural Operator (FNO) model in predicting flow fields under different wind directions and urban layouts. In this study, we investigate the effectiveness of the Fourier Neural Operator (FNO) model in predicting urban wind conditions under different wind directions and urban layouts. By training the model on velocity data from large eddy simulation data, we evaluate the performance of the model under different urban configurations and wind conditions. The results show that the FNO model can provide accurate predictions while significantly reducing the computational time by 99%. Our innovative approach of dividing the wind field into smaller spatial blocks for training improves the ability of the FNO model to capture wind frequency features effectively. The SDF data also provides important spatial building information, enhancing the model's ability to recognize physical boundaries and generate more realistic predictions. The proposed FNO approach enhances the AI model's generalizability for different wind directions and urban layouts.