☆ 4.5 Article

Gym-preCICE: Reinforcement learning environments for active flow control

SOFTWAREX (2023)

Related references

Note: Only part of the references are listed.

Article Thermodynamics

Deep reinforcement learning for turbulence modeling in large eddy simulations

Marius Kurz et al.

Summary: In recent years, supervised learning has become the state-of-the-art approach for data-driven turbulence modeling. However, this approach is not feasible for implicitly filtered large eddy simulation (LES) due to the unknown filter form. To address this issue, reinforcement learning is applied to directly interact with the LES environment and incorporate the implicit LES filter into the training process. The trained models demonstrate long-term stability, outperform analytical models, and generalize well to different resolutions and discretizations.

INTERNATIONAL JOURNAL OF HEAT AND FLUID FLOW (2023)

Add to Collection

Article Chemistry, Physical

Deep reinforcement learning for turbulent drag reduction in channel flows

Luca Guastoni et al.

Summary: This research introduces a reinforcement learning environment for designing and evaluating control strategies to reduce drag in turbulent fluid flows in a channel. The environment provides a computationally efficient, parallelized, high-fidelity fluid simulation framework that can interface with established RL agent programming interfaces. This allows for testing existing DRL algorithms and advancing our understanding of complex turbulent physical systems.

EUROPEAN PHYSICAL JOURNAL E (2023)

Add to Collection

Article Thermodynamics

Turbulence control in plane Couette flow using low-dimensional neural ODE-based models and deep reinforcement learning

Alec J. Linot et al.

Summary: In this study, we successfully applied deep reinforcement learning to control turbulent flows by using a data-driven low-dimensional model. Compared to traditional control methods, this approach significantly reduces training costs and improves control performance.

INTERNATIONAL JOURNAL OF HEAT AND FLUID FLOW (2023)

Add to Collection

Article Physics, Fluids & Plasmas

Recent advances in applying deep reinforcement learning for flow control: Perspectives and future directions

C. Vignon et al.

PHYSICS OF FLUIDS (2023)

Add to Collection

Article Mathematics, Applied

The deal.II library, Version 9.4

Daniel Arndt et al.

Summary: This paper mainly provides an overview of the new features of the finite element library deal.II, version 9.4.

JOURNAL OF NUMERICAL MATHEMATICS (2022)

Add to Collection

Article Mechanics

DRLinFluids: An open-source Python platform of coupling deep reinforcement learning and OpenFOAM

Qiulei Wang et al.

Summary: This is an open-source Python platform for deep reinforcement learning (DRL) applications in fluid mechanics. The platform, called DRLinFluids, enables simulations for flow control and optimization problems using DRL techniques, utilizing popular solvers and DRL packages in the field. The platform's reliability and efficiency are demonstrated in wake stabilization benchmark problems, offering substantial reduction in application effort for DRL in fluid mechanics.

PHYSICS OF FLUIDS (2022)

Add to Collection

Article Mechanics

Reinforcement-learning-based control of confined cylinder wakes with stability analyses

Jichao Li et al.

Summary: This research utilizes Reinforcement Learning for flow control by setting up synthetic jets on both sides of a cylinder to suppress vortex shedding. Through global linear stability and sensitivity analyses, and incorporating physical results in the design of control policies, the study finds that the controlled wake converges to the unstable base flow, requiring persistent oscillating control for maintaining this state. Additionally, embedding flow stability information in the reward function can lead to a more stable controlled flow.

JOURNAL OF FLUID MECHANICS (2021)

Add to Collection

Article Computer Science, Software Engineering

FEniCS-preCICE: Coupling FEniCS to other simulation software

Benjamin Rodenberg et al.

Summary: The new software FEniCS-preCICE acts as a middle layer facilitating coupling between existing FEniCS application codes and other simulation software via preCICE. It provides conversion between mesh and data structures, easy-to-use coupling conditions, and data checkpointing for implicit coupling, requiring only a few lines of additional code. Illustrative examples show good agreement with other simulation software.

SOFTWAREX (2021)

Add to Collection

Review Mechanics

Active flow control using machine learning: A brief review

Feng Ren et al.

JOURNAL OF HYDRODYNAMICS (2020)

Add to Collection

Article Multidisciplinary Sciences

Mastering Atari, Go, chess and shogi by planning with a learned model

Julian Schrittwieser et al.

NATURE (2020)

Add to Collection

Article Mechanics

Artificial neural networks trained through deep reinforcement learning discover control strategies for active flow control

Jean Rabault et al.

JOURNAL OF FLUID MECHANICS (2019)

Add to Collection

Article Mechanics

Accelerating deep reinforcement learning strategies of flow control through a multi-environment approach

Jean Rabault et al.

PHYSICS OF FLUIDS (2019)

Add to Collection

Article Multidisciplinary Sciences

Mastering the game of Go without human knowledge

David Silver et al.

NATURE (2017)

Add to Collection

Review Mechanics

Actuators for Active Flow Control

Louis N. Cattafesta et al.

ANNUAL REVIEW OF FLUID MECHANICS, VOL 43 (2011)

Add to Collection

Review Engineering, Aerospace

Issues in active flow control: theory, control, simulation, and experiment

SS Collis et al.

PROGRESS IN AEROSPACE SCIENCES (2004)

Add to Collection

© Peeref 2019-2024. All rights reserved.