Current browse context:
physics.flu-dyn
Change to browse by:
References & Citations
Physics > Fluid Dynamics
Title: Deep reinforcement learning-based active flow control of an elliptical cylinder: transitioning from an elliptical cylinder to a circular cylinder and a flat plate
(Submitted on 19 Apr 2024 (v1), last revised 29 Apr 2024 (this version, v2))
Abstract: This study investigates the effectiveness of active flow control (AFC) technology supported by deep reinforcement learning (DRL) applied to flows around elliptical cylinders at Re=100. We vary the aspect ratio (Ar) of the elliptical cylinder from an ellipsoid (Ar=2.0) to a circular shape (Ar=1.0), and ultimately to a flat plate (Ar=0). We utilize the proximal policy optimization (PPO) algorithm to precisely control the mass flow rates of synthetic jets located on both the upper and lower surfaces of the cylinder. The control objective focuses on reducing drag, minimizing lift and suppressing vortex shedding. We examine the robustness and adaptability of DRL-based control techniques across different geometric configurations. Our research findings indicate that, for elliptical cylinders with Ar between 1.75 and 0.75, the reduction in drag coefficient ranges from 0.9% to 15.7%, and the reduction in lift coefficient ranges from 95.2% to 99.7%. Notably, the DRL-based control strategy not only significantly reduces lift and drag, but also completely suppresses vortex shedding while using less than 1% of external excitation energy, demonstrating its efficiency and energy-saving capabilities. Additionally, for Ar from 0.5 to 0, the reduction in drag coefficient ranges from 26.9% to 43.6%, and the reduction in lift coefficient from 50.2% to 68.0%. This reflects the control strategy's significant reduction in both drag and lift coefficients, while also alleviating vortex shedding. Overall, our findings underscore the adaptability and potential of DRL-based AFC in controlling complex fluid dynamics across diverse geometric configurations.
Submission history
From: Wang Jia [view email][v1] Fri, 19 Apr 2024 16:59:50 GMT (46195kb,D)
[v2] Mon, 29 Apr 2024 13:05:08 GMT (41310kb,D)
Link back to: arXiv, form interface, contact.