Y. R. Zhao - National Key Laboratory of Science and Technology on Aerodynamic Design and Research, Northwestern Polytechnical University, Xi’an, ۷۱۰۰۷۲, China
H. Y. Xu - National Key Laboratory of Science and Technology on Aerodynamic Design and Research, Northwestern Polytechnical University, Xi’an, ۷۱۰۰۷۲, China
Z. Y. Xie - National Key Laboratory of Science and Technology on Aerodynamic Design and Research, Northwestern Polytechnical University, Xi’an, ۷۱۰۰۷۲, China

A closed-loop control framework is developed for the co-flow jet (CFJ) airfoil by combining the numerical flow field environment of a CFJ۰۰۱۲ airfoil with a deep reinforcement learning (DRL) module called tensorforce integrated in Python. The DRL agent, which is trained through interacting with the numerical flow field environment, is capable of acquiring a policy that instructs the mass flow rate of the CFJ to make the stalled airfoil at an angle of attack (AoA) of ۱۸ degrees reach a specific high lift coefficient set to ۲.۰, thereby effectively suppressing flow separation on the upper surface of the airfoil. The subsequent test shows that the policy can be implemented to find a precise jet momentum coefficient of ۰.۰۴۹ to make the lift coefficient of the CFJ۰۰۱۲ airfoil reach ۲.۰۱ with a negligible error of ۰.۵%. Moreover, to evaluate the generalization ability of the policy trained at an AoA of ۱۸ degrees, two additional tests are conducted at AoAs of ۱۶ and ۲۰ degrees. The results show that, although using the policy gained under another AoA cannot help the lift coefficient of the airfoil reach a set target of ۲ accurately, the errors are acceptable with less than ۵.۵%, which means the policy trained under an AoA of ۱۸ degrees can also be applied to other AoAs to some extent. This work is helpful for the practical application of CFJ technology, as the closed-loop control framework ensures good aerodynamic performance of the CFJ airfoil, even in complex and changeable flight conditions.

Co-flow jet, Closed-loop control, Flow control, Lift enhancement, Deep reinforcement learning