Artificial Neural Networks trained through Deep Reinforcement Learning discover control strategies for active flow control

TL;DR

This paper demonstrates that deep reinforcement learning can be used to train neural networks to discover effective active flow control strategies in fluid dynamics simulations, achieving vortex stabilization and drag reduction.

Contribution

It introduces the first application of deep reinforcement learning to active flow control, enabling neural networks to learn control strategies from interaction with the flow.

Findings

Neural network stabilizes vortex street and reduces drag by about 8%.

Control achieved with minimal actuation, around 0.5% of incoming flow.

Method opens new avenues for active flow control techniques.

Abstract

We present the first application of an Artificial Neural Network trained through a Deep Reinforcement Learning agent to perform active flow control. It is shown that, in a 2D simulation of the Karman vortex street at moderate Reynolds number (Re = 100), our Artificial Neural Network is able to learn an active control strategy from experimenting with the mass flow rates of two jets on the sides of a cylinder. By interacting with the unsteady wake, the Artificial Neural Network successfully stabilizes the vortex alley and reduces drag by about 8%. This is performed while using small mass flow rates for the actuation, on the order of 0.5% of the mass flow rate intersecting the cylinder cross section once a new pseudo-periodic shedding regime is found. This opens the way to a new class of methods for performing active flow control.