AI-enhanced large-eddy simulation for urban wind flows: accelerating spin-up with Fourier neural operator initialization
Bibliographic record
Abstract
Urban wind simulations based on large-eddy simulation (LES) offer high fidelity in capturing turbulent dynamics across complex building geometries. However, their practical application remains limited due to the high computational cost associated with the prolonged spin-up period required to reach statistically steady turbulence. This study introduces a hybrid deep learning–CFD framework that leverages the Fourier Neural Operator (FNO) to generate physically consistent initial flow fields, thereby significantly reducing LES initialization time. A customized FNO model, Flume-FNO, is trained on 30 LES-generated flow cases over real urban geometries worldwide using GPU-accelerated simulations. To enhance geometric representation, the model incorporates multi-directional distance features (MDDF) and a localization strategy, enabling accurate flow field prediction from building morphology alone. The predicted fields are then used to restart LES, yielding mean velocity profiles, turbulence statistics, and mean pressure coefficients that closely match those of fully developed LES results. Quantitative evaluations show a velocity profile error below 5 % and accurate reproduction of coherent structures such as vorticity and shear layers. Furthermore, the FNO-initialized simulations achieve up to a 48.8 % reduction in spin-up time. This work demonstrates a scalable, efficient, and generalizable approach to accelerating urban LES using deep learning priors, with applications in microclimate modeling, pedestrian wind comfort, eVTOL safety, and digital twins. • Develops a hybrid AI-CFD framework to accelerate high-fidelity urban wind simulations. • Accurately predicts flow structures, turbulence statistics, and aerodynamic behavior in complex urban settings. • Reduces LES simulation time by nearly 50 % while retaining high-resolution detail and physical fidelity. • Accurately resolves key aerodynamic features around complex urban geometries, including pressure zones, vortices, and wake dynamics.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".