Drag Reduction in Flows Past 2D and 3D Circular Cylinders Through Deep Reinforcement Learning

TL;DR

This paper demonstrates that deep reinforcement learning can discover effective and transferable control strategies for reducing drag in flows past 2D and 3D cylinders, even beyond training conditions, with implications for unsteady flow management.

Contribution

It introduces a deep reinforcement learning approach to control flow separation on cylinders, revealing transferable policies and analyzing drag reduction mechanisms across dimensions.

Findings

Discovered control policies generalize to higher Reynolds numbers.

Two-dimensional controls transfer effectively to three-dimensional flows.

Trade-offs between drag reduction and energy input are characterized.

Abstract

We investigate drag reduction mechanisms in flows past two- and three-dimensional cylinders controlled by surface actuators using deep reinforcement learning. We investigate 2D and 3D flows at Reynolds numbers up to 8,000 and 4,000, respectively. The learning agents are trained in planar flows at various Reynolds numbers, with constraints on the available actuation energy. The discovered actuation policies exhibit intriguing generalization capabilities, enabling open-loop control even for Reynolds numbers beyond their training range. Remarkably, the discovered two-dimensional controls, inducing delayed separation, are transferable to three-dimensional cylinder flows. We examine the trade-offs between drag reduction and energy input while discussing the associated mechanisms. The present work paves the way for control of unsteady separated flows via interpretable control strategies discovered through deep reinforcement learning.