Anchored-Branched Steady-state WInd Flow Transformer (AB-SWIFT): a metamodel for 3D atmospheric flow in urban environments

TL;DR

AB-SWIFT is a transformer-based metamodel designed for accurate 3D urban atmospheric flow prediction, overcoming challenges of high geometric variability and large mesh sizes in deep learning models.

Contribution

We introduce AB-SWIFT, a novel branched transformer architecture tailored for urban atmospheric flow modeling, trained on diverse simulations to outperform existing models.

Findings

Achieves superior accuracy over state-of-the-art models

Effectively handles high geometric variability in urban environments

Demonstrates robustness across different atmospheric stratifications

Abstract

Air flow modeling at a local scale is essential for applications such as pollutant dispersion modeling or wind farm modeling. To circumvent costly Computational Fluid Dynamics (CFD) computations, deep learning surrogate models have recently emerged as promising alternatives. However, in the context of urban air flow, deep learning models struggle to adapt to the high variations of the urban geometry and to large mesh sizes. To tackle these challenges, we introduce Anchored Branched Steady-state WInd Flow Transformer (AB-SWIFT), a transformer-based model with an internal branched structure uniquely designed for atmospheric flow modeling. We train our model on a specially designed database of atmospheric simulations around randomised urban geometries and with a mixture of unstable, neutral, and stable atmospheric stratifications. Our model reaches the best accuracy on all predicted fields compared to state-of-the-art transformers and graph-based models. Our code and data is available at https://github.com/cerea-daml/abswift.