Dynamic mode decomposition of noisy flow data
Fluid Dynamics
2025-01-30 v2
Abstract
Dynamic mode decomposition (DMD) is a popular approach to analyzing and modeling fluid flows. In practice, datasets are almost always corrupted to some degree by noise. The vanilla DMD is highly noise-sensitive, which is why many algorithmic extensions for improved robustness exist. We introduce a flexible optimization approach that merges available ideas for improved accuracy and robustness. The approach simultaneously identifies coherent dynamics and noise in the data. In tests on the laminar flow past a cylinder, the method displays strong noise robustness and high levels of accuracy.
Cite
@article{arxiv.2411.04868,
title = {Dynamic mode decomposition of noisy flow data},
author = {Andre Weiner and Janis Geise},
journal= {arXiv preprint arXiv:2411.04868},
year = {2025}
}