Actuation sensitivity analysis of cylinder wake via automatic differentiation
Physics of Fluids · 01.11.2026 · 38 (11)
DOI: 10.1063/5.0360686 ↗Kurz gefasst
Fachartikel in der Zeitschrift Physics of Fluids, online erschienen am 12.11.2026. Verfasst von Changrui Mao und 1 weiteren Autor:innen. Die Arbeit nennt 28 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Jiangsu Province, Priority Academic Program Development of Jiangsu Higher Education Institutions.
Abstract (Originalsprache)
The response of a bluff-body wake to actuation is confined to specific flow regions and frequency bands tied to vortex shedding. A precise, quantitative measure of this receptivity to arbitrary, time-dependent actuation remains challenging. We use the differentiable solver JAX-Fluids to compute the gradient of the aerodynamic objective with respect to the actuation. The gradient is obtained by reverse-mode automatic differentiation. The test case is a confined cylinder wake with synthetic-jet actuation. This gradient is effective as an open-loop directive only when combined with periodic actuation parameterization over long time windows. When the gradient is instead supplied to a reinforcement-learning agent as a temporary physics-informed prior, it stabilizes the learning process in the runs reported without altering final performance. Both policies consistently align with the loss gradient in the high-gradient region. That is, the action update has the same sign as the loss gradient. In the low-gradient region, the alignment dissolves and the policies rely on their own learned control rhythm. These results quantify when and how a cylinder wake can be influenced by actuation. They also indicate the limits of automatic differentiation gradients in unsteady flow control.
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Jiangsu Province
- Priority Academic Program Development of Jiangsu Higher Education Institutions
Zitieren
APA
Mao, C., & Wu, J. (2026). Actuation sensitivity analysis of cylinder wake via automatic differentiation. Physics of Fluids, 38(11). https://doi.org/10.1063/5.0360686
Vancouver
Mao C, Wu J. Actuation sensitivity analysis of cylinder wake via automatic differentiation. Physics of Fluids. 2026;38(11). doi:10.1063/5.0360686
BibTeX
@article{Mao2026,
author = {Mao, Changrui and Wu, Jie},
title = {Actuation sensitivity analysis of cylinder wake via automatic differentiation},
journal = {Physics of Fluids},
year = {2026},
volume = {38},
number = {11},
doi = {10.1063/5.0360686}
}
EUVIMED