Next Article in Journal
Hydrogen Production from Methane Cracking in Dielectric Barrier Discharge Catalytic Plasma Reactor Using a Nanocatalyst
Previous Article in Journal
Quad-Trapezoidal-Leg Orthoplanar Spring with Piezoelectric Plate for Enhancing the Performances of Vibration Energy Harvester
Previous Article in Special Issue
Insights into the Aeroacoustic Noise Generation for Vertical Axis Turbines in Close Proximity
Open AccessArticle

Deep Reinforcement Learning Control of Cylinder Flow Using Rotary Oscillations at Low Reynolds Number

1
Institute of Thermophysics SB RAS, Lavrentyev ave. 1, 630090 Novosibirsk, Russia
2
Physics and Mathematics Departments, Novosibirsk State University, Pirogov str. 1, 630090 Novosibirsk, Russia
*
Author to whom correspondence should be addressed.
Energies 2020, 13(22), 5920; https://doi.org/10.3390/en13225920
Received: 30 September 2020 / Revised: 26 October 2020 / Accepted: 27 October 2020 / Published: 13 November 2020
(This article belongs to the Special Issue Machine-Learning Methods for Complex Flows)
We apply deep reinforcement learning to active closed-loop control of a two-dimensional flow over a cylinder oscillating around its axis with a time-dependent angular velocity representing the only control parameter. Experimenting with the angular velocity, the neural network is able to devise a control strategy based on low frequency harmonic oscillations with some additional modulations to stabilize the Kármán vortex street at a low Reynolds number Re=100. We examine the convergence issue for two reward functions showing that later epoch number does not always guarantee a better result. The performance of the controller provide the drag reduction of 14% or 16% depending on the employed reward function. The additional efforts are very low as the maximum amplitude of the angular velocity is equal to 8% of the incoming flow in the first case while the latter reward function returns an impressive 0.8% rotation amplitude which is comparable with the state-of-the-art adjoint optimization results. A detailed comparison with a flow controlled by harmonic oscillations with fixed amplitude and frequency is presented, highlighting the benefits of a feedback loop. View Full-Text
Keywords: flow control; ANN; DRL flow control; ANN; DRL
Show Figures

Figure 1

MDPI and ACS Style

Tokarev, M.; Palkin, E.; Mullyadzhanov, R. Deep Reinforcement Learning Control of Cylinder Flow Using Rotary Oscillations at Low Reynolds Number. Energies 2020, 13, 5920.

AMA Style

Tokarev M, Palkin E, Mullyadzhanov R. Deep Reinforcement Learning Control of Cylinder Flow Using Rotary Oscillations at Low Reynolds Number. Energies. 2020; 13(22):5920.

Chicago/Turabian Style

Tokarev, Mikhail; Palkin, Egor; Mullyadzhanov, Rustam. 2020. "Deep Reinforcement Learning Control of Cylinder Flow Using Rotary Oscillations at Low Reynolds Number" Energies 13, no. 22: 5920.

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Search more from Scilit
 
Search
Back to TopTop