Next meeting 14th UFM Meeting · 23 Mar 2027 · Bristol Save the date →
Urban Fluid MechanicsSpecial Interest Group

feilian-3d: deep-learning prediction of 3D wind fields in cities

Simulation

Deep learning model code predicting 3D wind speed fields in cities from building geometry.

Get the code

Zenodo DOI 10.5281/zenodo.18252581
Licence
CC BY 4.0 Free to use and share, including commercially, as long as you credit the creators.
Size
38 kB

How to cite

Green, S., Deeptimahanti, D., Leong, I. H., Mann, T., Abdurakhmanov, I., Lu, J., & Nazarian, N. (2026). feilian-3d [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.18252581

Check the repository for the citation its authors prefer.

Details

Resource
Software & tools
Method
ML / surrogate
Geometry
City scale
Topic
Wind loads & pedestrian wind
Quantities
Velocity
Year
2026
Creators
Green, Samuel; Deeptimahanti, Deva; Leong, Iao Hoi; Mann, Tristyn; Abdurakhmanov, Ilkhom; Lu, Jiachen; Nazarian, Negin
Institution
Pawsey Supercomputing Research Centre; UNSW; UNSW Sydney

Related resources

  1. FLICK: fast urban wind-field prediction from building geometry

    112 MB

    Open-source Python package for fast urban wind-field prediction from geometry.

    Simulation Software & tools ML / surrogate

    2026 · AGPL-3.0 · Paper: Building a general and data-efficient convolutional neural network-based model for fast urban flow estimation ↗

  2. UrbanFlowNet-MFSS: multi-fidelity surrogates for urban wind reconstruction

    UrbanFlowNet-MFSS, a physics-regularised multi-fidelity surrogate for urban flow reconstruction.

    Simulation Software & tools ML / surrogate

    2025 · MIT

  3. Flow simulations around randomly generated urban geometries (Random Buildings)

    3.6 GB

    Dataset of microscale atmospheric flow simulations around randomly generated urban geometries for several stability classes. Useful for training and testing ML surrogates of urban flow.

    Simulation Simulation output RANSML / surrogate

    2026 · MIT

  4. Building-resolved wind fields over Camden, London, for scientific machine learning

    508 MB

    41 PALM NetCDF files of building-resolved wind fields for Camden, London, for scientific machine learning.

    Simulation Simulation output LESML / surrogate

    Camden, London, UK · 2026 · CC BY 4.0

  5. Physics-informed surrogate of urban wind from sparse measurements

    30 MB

    Field and CFD data for physics-informed surrogate modelling of urban wind under sparse measurements.

    Field MeasurementsSimulation output RANSML / surrogate

    2026 · CC BY 4.0

  6. Agentic AI workflow for automated wind CFD of buildings and cities

    Paper release of an agentic AI workflow that automates CFD wind simulations for buildings and cities. Useful for automated case setup.

    Simulation Software & tools ML / surrogate

    2026 · CC BY 4.0

All 36 resources on pedestrian wind & wind loads →

From Zenodo, added 2026-11-07. Listed there as “feilian-3d”. Tags proposed by AI and not yet checked by hand. Spotted a mistake? Tell the portal contacts.