Machine-learned flow estimation with sparse data -- exemplified for the rooftop of a UAV vertiport

TL;DR

This paper introduces a physics-informed machine learning framework for urban wind estimation that generalizes well to unseen conditions, demonstrated on UAV vertiport flow data.

Contribution

It develops a novel normalized manifold approach with a double encoder-decoder architecture for sparse sensor-based wind estimation, incorporating Reynolds number independence.

Findings

Accurately estimates wind conditions beyond training data.

Generalizes well to rare and unseen wind scenarios.

Reduces computational load via manifold clustering.

Abstract

We propose a physics-informed data-driven framework for urban wind estimation. This framework validates and incorporates the Reynolds number independence for flows under various working conditions, thus allowing the extrapolation for wind conditions far beyond the training data. Another key enabler is a machine-learned non-dimensionalized manifold from snapshot data. The velocity field is modeled using a double encoder-decoder approach. The first encoder normalizes data using the oncoming wind speed, while the second encoder projects this normalized data onto the isometric feature mapping manifold. The decoders reverse this process, with $k$-nearest neighbor performing the first decoding and the second undoing the normalization. The manifold is coarse-grained by clustering to reduce the computational load for de- and encoding. The sensor-based flow estimation is based on the estimate of the oncoming wind speed and a mapping from sensor signal to the manifold latent variables. The proposed machine-learned flow estimation framework is exemplified for the flow above an Unmanned Aerial Vehicle vertiport. The wind estimation is shown to generalize well for rare wind conditions, not included in the original database.