Recent advances in applying deep reinforcement learning for flow control: Perspectives and future directions

C Vignon, J Rabault, R Vinuesa - Physics of fluids, 2023 - pubs.aip.org
Deep reinforcement learning (DRL) has been applied to a variety of problems during the
past decade and has provided effective control strategies in high-dimensional and non …

A review on deep reinforcement learning for fluid mechanics: An update

J Viquerat, P Meliga, A Larcher, E Hachem - Physics of Fluids, 2022 - pubs.aip.org
In the past couple of years, the interest of the fluid mechanics community for deep
reinforcement learning techniques has increased at fast pace, leading to a growing …

Deep reinforcement learning for flow control exploits different physics for increasing Reynolds number regimes

P Varela, P Suárez, F Alcántara-Ávila, A Miró… - Actuators, 2022 - mdpi.com
The increase in emissions associated with aviation requires deeper research into novel
sensing and flow-control strategies to obtain improved aerodynamic performances. In this …

Effective control of two-dimensional Rayleigh–Bénard convection: Invariant multi-agent reinforcement learning is all you need

C Vignon, J Rabault, J Vasanth, F Alcántara-Ávila… - Physics of …, 2023 - pubs.aip.org
Rayleigh–Bénard convection (RBC) is a recurrent phenomenon in a number of industrial
and geoscience flows and a well-studied system from a fundamental fluid-mechanics …

Dynamic feature-based deep reinforcement learning for flow control of circular cylinder with sparse surface pressure sensing

Q Wang, L Yan, G Hu, W Chen, J Rabault… - Journal of Fluid …, 2024 - cambridge.org
This study proposes a self-learning algorithm for closed-loop cylinder wake control targeting
lower drag and lower lift fluctuations with the additional challenge of sparse sensor …

[HTML][HTML] Thermodynamics-informed neural network for recovering supercritical fluid thermophysical information from turbulent velocity data

N Masclans, F Vázquez-Novoa, M Bernades… - International Journal of …, 2023 - Elsevier
Recent research has highlighted the potential of supercritical fluids under high-pressure
transcritical conditions to achieve microconfined turbulence as a result of the thermophysical …

Machine-learning flow control with few sensor feedback and measurement noise

R Castellanos, GY Cornejo Maceda, I De La Fuente… - Physics of …, 2022 - pubs.aip.org
A comparative assessment of machine-learning (ML) methods for active flow control is
performed. The chosen benchmark problem is the drag reduction of a two-dimensional …

Unsteady cylinder wakes from arbitrary bodies with differentiable physics-assisted neural network

S Brahmachary, N Thuerey - Physical Review E, 2024 - APS
This work describes a hybrid predictive framework configured as a coarse-grained surrogate
for reconstructing unsteady fluid flows around multiple cylinders of diverse configurations …

Active flow control for bluff body drag reduction using reinforcement learning with partial measurements

C **a, J Zhang, EC Kerrigan, G Rigas - Journal of Fluid Mechanics, 2024 - cambridge.org
Active flow control for drag reduction with reinforcement learning (RL) is performed in the
wake of a two-dimensional square bluff body at laminar regimes with vortex shedding …

Establishment and validation of a relationship model between nozzle experiments and CFD results based on convolutional neural network

T Yu, X Wu, Y Yu, R Li, H Zhang - Aerospace Science and Technology, 2023 - Elsevier
The acquisition of experimental data in a supersonic wind tunnel often faces challenges of
complexity and high costs. Furthermore, there are limitations in the control of experimental …