Active Flow Control

Active flow control (AFC) aims to manipulate fluid flows to improve efficiency and performance, often focusing on drag reduction. Current research heavily utilizes deep reinforcement learning (DRL), employing various architectures like proximal policy optimization and convolutional recurrent autoencoders, to optimize control strategies in complex scenarios such as flow around cylinders. This approach is proving effective in reducing computational costs through parallelization and optimized sensor placement, leading to significant drag reductions and improved control performance in both simulations and, increasingly, experimental settings. The advancements in AFC have implications for various engineering applications, including aerospace and marine design, by enabling more efficient and robust flow management.

Papers

November 8, 2024

Towards Active Flow Control Strategies Through Deep Reinforcement Learning
Ricard Montalà, Bernat Font, Pol Suárez, Jean Rabault, Oriol Lehmkuhl, Ivette Rodriguez
LeArning Abstract Deep Reinforcement Learning Deep Reinforcement Aerodynamic Force Flow Control Active Flow Control Actuation Strategy

February 18, 2024

Optimal Parallelization Strategies for Active Flow Control in Deep Reinforcement Learning-Based Computational Fluid Dynamics
Wang Jia, Hang Xu
Deep Reinforcement Learning Deep Reinforcement Computational Fluid Dynamic Large Scale Benchmark Parallelization Strategy Active Flow Control

September 4, 2023

Active flow control for three-dimensional cylinders through deep reinforcement learning
Pol Suárez, Francisco Alcántara-Ávila, Arnau Miró, Jean Rabault, Bernat Font, Oriol Lehmkuhl, R. Vinuesa
Deep Reinforcement Learning Aerodynamic Force Minutia Cylinder Code Active Flow Control

July 22, 2023

Active Control of Flow over Rotating Cylinder by Multiple Jets using Deep Reinforcement Learning
Kamyar Dobakhti, Jafar Ghazanfarian
Reinforcement Learning Deep Reinforcement Learning Aerodynamic Force Flow Mood Dynamic Control Minutia Cylinder Code Active Flow Control Multiple Jet

July 5, 2023

Dynamic Feature-based Deep Reinforcement Learning for Flow Control of Circular Cylinder with Sparse Surface Pressure Sensing
Qiulei Wang, Lei Yan, Gang Hu, Wenli Chen, Jean Rabault, Bernd R. Noack
Continuous Control Flow Control Minutia Cylinder Code Wake Model Active Flow Control

May 3, 2023

Gym-preCICE: Reinforcement Learning Environments for Active Flow Control
Mosayeb Shams, Ahmed H. Elsheikh
Reinforcement Learning Reinforcement Learning Environment MR iNet Gym Active Flow Control Multi Physic Simulation

April 23, 2023

How to Control Hydrodynamic Force on Fluidic Pinball via Deep Reinforcement Learning
Haodong Feng, Yue Wang, Hui Xiang, Zhiyang Jin, Dixia Fan
Deep Reinforcement Learning Fluid Dynamic Hydrodynamic Force Active Flow Control Fluidic Actuator

November 1, 2022

Combined space-time reduced-order model with 3D deep convolution for extrapolating fluid dynamics
Indu Kant Deo, Rui Gao, Rajeev Jaiman
Reduced Order Fluid Dynamic Convolutional Autoencoder 3D Convolutional Temporal Convolution Based Reduced Order Model Active Flow Control

June 19, 2022

An Embedded Feature Selection Framework for Control
Jiawen Wei, Fangyuan Wang, Wanxin Zeng, Wenwei Lin, Ning Gui
External Control Optimal Control Aerodynamic Force Dynamic Control Efficient Control Active Flow Control

February 23, 2022

Comparative analysis of machine learning methods for active flow control
Fabio Pino, Lorenzo Schena, Jean Rabault, Miguel A. Mendez
Machine Learning Comparative Study Genetic Programming Model Free Global Optimization Flow Control Active Flow Control

Active Flow Control

Papers

Towards Active Flow Control Strategies Through Deep Reinforcement Learning

Optimal Parallelization Strategies for Active Flow Control in Deep Reinforcement Learning-Based Computational Fluid Dynamics

Active flow control for three-dimensional cylinders through deep reinforcement learning

Active Control of Flow over Rotating Cylinder by Multiple Jets using Deep Reinforcement Learning

Dynamic Feature-based Deep Reinforcement Learning for Flow Control of Circular Cylinder with Sparse Surface Pressure Sensing

Gym-preCICE: Reinforcement Learning Environments for Active Flow Control

How to Control Hydrodynamic Force on Fluidic Pinball via Deep Reinforcement Learning

Combined space-time reduced-order model with 3D deep convolution for extrapolating fluid dynamics

An Embedded Feature Selection Framework for Control

Comparative analysis of machine learning methods for active flow control