Research on the Application of Deep Learning Technology in urban wind field prediction
DOI:
https://doi.org/10.71222/4ref5v74Keywords:
Urban airflow, Deep learning, Turbulence modelAbstract
Rapid and accurate prediction of wind speed fields in urban environments is vital for understanding urban climates, promoting sustainability, and mitigating disasters. Existing prediction methods primarily rely on Computational Fluid Dynamics (CFD) numerical simulations. However, these traditional approaches face significant constraints regarding both accuracy and computational efficiency. With rapid advancements in computer technology, deep learning is gradually emerging as an efficient alternative to conventional CFD simulations for wind field prediction. This paper elaborates on the application of deep learning methods in wind field forecasting and comprehensively analyzes the current problems and challenges encountered in this domain. Finally, the study explores future prospects for the development of wind field prediction technologies.References
1. J. Yang, Y. Wang, B. Xue, et al., "Contribution of urban ventilation to the thermal environment and urban energy demand: Different climate background perspectives," Science of The Total Environment, vol. 795, p. 148791, 2021.
2. M. Kurppa, A. Hellsten, M. Auvinen, et al., "Ventilation and air quality in city blocks using large-eddy simulation---urban planning perspective," Atmosphere, vol. 9, no. 2, p. 65, 2018.
3. B. S. Han, S. B. Park, J. J. Baik, et al., "Large‐eddy simulation of vortex streets and pollutant dispersion behind high‐rise buildings," Quarterly Journal of the Royal Meteorological Society, vol. 143, no. 708, pp. 2714-2726, 2017.
4. Z. Zhai, "Application of computational fluid dynamics in building design: aspects and trends," Indoor and Built Environment, vol. 15, no. 4, pp. 305-313, 2006.
5. P. Xie, T. Li, J. Liu, et al., "Urban flow prediction from spatiotemporal data using machine learning: A survey," Information Fusion, vol. 59, pp. 1-12, 2020.
6. A. Kashefi, D. Rempe, L. J. Guibas, "A point-cloud deep learning framework for prediction of fluid flow fields on irregular geometries," Physics of Fluids, vol. 33, no. 2, p. 027104, 2021.
7. X. Guo, W. Li, F. Iorio, "Convolutional Neural Networks for Steady Flow Approximation," in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, California, USA: ACM, 2016, pp. 481-490.
8. V. Sekar, Q. Jiang, C. Shu, et al., "Fast flow field prediction over airfoils using deep learning approach," Physics of Fluids, vol. 31, no. 5, p. 057103, 2019.
9. M. D. Ribeiro, A. Rehman, S. Ahmed, et al., "DeepCFD: Efficient Steady-State Laminar Flow Approximation with Deep Convolutional Neural Networks," arXiv, 2020.
10. T. P. Miyanawala and R. K. Jaiman, "An Efficient Deep Learning Technique for the Navier-Stokes Equations: Application to Unsteady Wake Flow Dynamics," arXiv, 2017.
11. S. Xiang, X. Fu, J. Zhou, et al., "Non-intrusive reduced order model of urban airflow with dynamic boundary conditions," Building and Environment, vol. 187, p. 107397, 2021.
12. S. Wiewel, M. Becher, N. Thuerey, "Latent space physics: Towards learning the temporal evolution of fluid flow," in Computer Graphics Forum, vol. 38, Wiley Online Library, 2019, pp. 71-82.
13. M. Song, J. Huang, X. Shao, et al., "A three-dimensional dynamic wake prediction framework for multiple turbine operating states based on diffusion model," Energy, vol. 333, p. 137084, 2025.
14. T. Pfaff, M. Fortunato, A. Sanchez-Gonzalez, et al., "Learning Mesh-Based Simulation with Graph Networks," 2020.
15. M. Fortunato, T. Pfaff, P. Wirnsberger, et al., "MultiScale MeshGraphNets," arXiv, 2022.
16. R. Zhao, S. Liu, J. Liu, et al., "A two-stage CFD-GNN approach for efficient steady-state prediction of urban airflow and airborne contaminant dispersion," Sustainable Cities and Society, vol. 112, p. 105607, 2024.
17. X. Han, H. Gao, T. Pfaff, et al., "Predicting Physics in Mesh-reduced Space with Temporal Attention," arXiv, 2022.
18. C. Rao, H. Sun, Y. Liu, "Physics-informed deep learning for incompressible laminar flows," Theoretical and Applied Mechanics Letters, vol. 10, no. 3, pp. 207-212, 2020.
19. X. Shao, Z. Liu, S. Zhang, et al., "PIGNN-CFD: A physics-informed graph neural network for rapid predicting urban wind field defined on unstructured mesh," Building and Environment, vol. 232, p. 110056, 2023.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Chenyu Ma (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
