Single-step deep reinforcement learning for open-loop control of laminar and turbulent flows

TL;DR

This paper demonstrates the potential of a novel single-step deep reinforcement learning approach, based on a modified PPO algorithm, for optimizing complex fluid flow control problems in computational fluid dynamics.

Contribution

It introduces a new single-step PPO algorithm combined with a stabilized finite elements environment for effective flow control optimization.

Findings

Single-step PPO effectively optimizes fluid flow control.

Method outperforms traditional direct and adjoint methods in certain cases.

Applicable to complex unsteady and high-dimensional flow problems.

Abstract

This research gauges the ability of deep reinforcement learning (DRL) techniques to assist the optimization and control of fluid mechanical systems. It combines a novel, "degenerate" version of the proximal policy optimization (PPO) algorithm, that trains a neural network in optimizing the system only once per learning episode, and an in-house stabilized finite elements environment implementing the variational multiscale (VMS) method, that computes the numerical reward fed to the neural network. Three prototypical examples of separated flows in two dimensions are used as testbed for developing the methodology, each of which adds a layer of complexity due either to the unsteadiness of the flow solutions, or the sharpness of the objective function, or the dimension of the control parameter space. Relevance is carefully assessed by comparing systematically to reference data obtained by canonical direct and adjoint methods. Beyond adding value to the shallow literature on this subject, these findings establish the potential of single-step PPO for reliable black-box optimization of computational fluid dynamics (CFD) systems, which paves the way for future progress in optimal flow control using this new class of methods.