Active Control of Separated Flows on 3D Wings Using Deep Reinforcement Learning (DRL)

In this work, deep reinforcement learning (DRL) is applied to active flow control (AFC) over a three-dimensional SD7003 wing at a Reynolds number of Re = 60,000 and angle of attack of AoA = 14◦. In the uncontrolled baseline case, the flow exhibits massive separation and a fully turbulent wake. Using a GPU-accelerated CFD solver and multi-agent training, DRL discovers control strategies that enhance lift (ΔCl = 79%), reduce drag (ΔCd = −39%), and improve aerodynamic efficiency (ΔE = 408%). Flow visualizations confirm reattachment of the separated shear layer, demonstrating the potential of DRL for complex and turbulent flows.