Physics-informed Neural-operator Predictive Control for Drag Reduction in Turbulent Flows
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916987815329792 |
|---|---|
| author | Zhao, Zelin Li, Zongyi Hassibi, Kimia Azizzadenesheli, Kamyar Yan, Junchi Bae, H. Jane Zhou, Di Anandkumar, Anima |
| author_facet | Zhao, Zelin Li, Zongyi Hassibi, Kimia Azizzadenesheli, Kamyar Yan, Junchi Bae, H. Jane Zhou, Di Anandkumar, Anima |
| contents | Assessing turbulence control effects for wall friction numerically is a significant challenge since it requires expensive simulations of turbulent fluid dynamics. We instead propose an efficient deep reinforcement learning (RL) framework for modeling and control of turbulent flows. It is model-based RL for predictive control (PC), where both the policy and the observer models for turbulence control are learned jointly using Physics Informed Neural Operators (PINO), which are discretization invariant and can capture fine scales in turbulent flows accurately. Our PINO-PC outperforms prior model-free reinforcement learning methods in various challenging scenarios where the flows are of high Reynolds numbers and unseen, i.e., not provided during model training. We find that PINO-PC achieves a drag reduction of 39.0\% under a bulk-velocity Reynolds number of 15,000, outperforming previous fluid control methods by more than 32\%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_03360 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Physics-informed Neural-operator Predictive Control for Drag Reduction in Turbulent Flows Zhao, Zelin Li, Zongyi Hassibi, Kimia Azizzadenesheli, Kamyar Yan, Junchi Bae, H. Jane Zhou, Di Anandkumar, Anima Machine Learning Artificial Intelligence Optimization and Control Fluid Dynamics Assessing turbulence control effects for wall friction numerically is a significant challenge since it requires expensive simulations of turbulent fluid dynamics. We instead propose an efficient deep reinforcement learning (RL) framework for modeling and control of turbulent flows. It is model-based RL for predictive control (PC), where both the policy and the observer models for turbulence control are learned jointly using Physics Informed Neural Operators (PINO), which are discretization invariant and can capture fine scales in turbulent flows accurately. Our PINO-PC outperforms prior model-free reinforcement learning methods in various challenging scenarios where the flows are of high Reynolds numbers and unseen, i.e., not provided during model training. We find that PINO-PC achieves a drag reduction of 39.0\% under a bulk-velocity Reynolds number of 15,000, outperforming previous fluid control methods by more than 32\%. |
| title | Physics-informed Neural-operator Predictive Control for Drag Reduction in Turbulent Flows |
| topic | Machine Learning Artificial Intelligence Optimization and Control Fluid Dynamics |
| url | https://arxiv.org/abs/2510.03360 |