SmartFlow: A CFD-solver-agnostic deep reinforcement learning framework for computational fluid dynamics on HPC platforms

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xiao, Maochao, Wang, Yuning, Rodach, Felix, Font, Bernat, Kurz, Marius, Suárez, Pol, Zhou, Di, Alcántara-Ávila, Francisco, Zhu, Ting, Liu, Junle, Montalà, Ricard, Chen, Jiawei, Rabault, Jean, Lehmkuhl, Oriol, Beck, Andrea, Larsson, Johan, Vinuesa, Ricardo, Pirozzoli, Sergio
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918110130339840
author Xiao, Maochao
Wang, Yuning
Rodach, Felix
Font, Bernat
Kurz, Marius
Suárez, Pol
Zhou, Di
Alcántara-Ávila, Francisco
Zhu, Ting
Liu, Junle
Montalà, Ricard
Chen, Jiawei
Rabault, Jean
Lehmkuhl, Oriol
Beck, Andrea
Larsson, Johan
Vinuesa, Ricardo
Pirozzoli, Sergio
author_facet Xiao, Maochao
Wang, Yuning
Rodach, Felix
Font, Bernat
Kurz, Marius
Suárez, Pol
Zhou, Di
Alcántara-Ávila, Francisco
Zhu, Ting
Liu, Junle
Montalà, Ricard
Chen, Jiawei
Rabault, Jean
Lehmkuhl, Oriol
Beck, Andrea
Larsson, Johan
Vinuesa, Ricardo
Pirozzoli, Sergio
contents Deep reinforcement learning (DRL) is emerging as a powerful tool for fluid-dynamics research, encompassing active flow control, autonomous navigation, turbulence modeling and discovery of novel numerical schemes. We introduce SmartFlow, a CFD-solver-agnostic framework for both single- and multi-agent DRL algorithms that can easily integrate with MPI-parallel CPU and GPU-accelerated solvers. Built on Relexi and SmartSOD2D, SmartFlow uses the SmartSim infrastructure library and our newly developed SmartRedis-MPI library to enable asynchronous, low-latency, in-memory communication between CFD solvers and Python-based DRL algorithms. SmartFlow leverages PyTorch's Stable-Baselines3 for training, which provides a modular, Gym-like environment API. We demonstrate its versatility via three case studies: single-agent synthetic-jet control for drag reduction in a cylinder flow simulated by the high-order FLEXI solver, multi-agent cylinder wake control using the GPU-accelerated spectral-element code SOD2D, and multi-agent wall-model learning for large-eddy simulation with the finite-difference solver CaLES. SmartFlow's CFD-solver-agnostic design and seamless HPC integration is promising to accelerate RL-driven fluid-mechanics studies.
format Preprint
id arxiv_https___arxiv_org_abs_2508_00645
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SmartFlow: A CFD-solver-agnostic deep reinforcement learning framework for computational fluid dynamics on HPC platforms
Xiao, Maochao
Wang, Yuning
Rodach, Felix
Font, Bernat
Kurz, Marius
Suárez, Pol
Zhou, Di
Alcántara-Ávila, Francisco
Zhu, Ting
Liu, Junle
Montalà, Ricard
Chen, Jiawei
Rabault, Jean
Lehmkuhl, Oriol
Beck, Andrea
Larsson, Johan
Vinuesa, Ricardo
Pirozzoli, Sergio
Fluid Dynamics
Computational Physics
Deep reinforcement learning (DRL) is emerging as a powerful tool for fluid-dynamics research, encompassing active flow control, autonomous navigation, turbulence modeling and discovery of novel numerical schemes. We introduce SmartFlow, a CFD-solver-agnostic framework for both single- and multi-agent DRL algorithms that can easily integrate with MPI-parallel CPU and GPU-accelerated solvers. Built on Relexi and SmartSOD2D, SmartFlow uses the SmartSim infrastructure library and our newly developed SmartRedis-MPI library to enable asynchronous, low-latency, in-memory communication between CFD solvers and Python-based DRL algorithms. SmartFlow leverages PyTorch's Stable-Baselines3 for training, which provides a modular, Gym-like environment API. We demonstrate its versatility via three case studies: single-agent synthetic-jet control for drag reduction in a cylinder flow simulated by the high-order FLEXI solver, multi-agent cylinder wake control using the GPU-accelerated spectral-element code SOD2D, and multi-agent wall-model learning for large-eddy simulation with the finite-difference solver CaLES. SmartFlow's CFD-solver-agnostic design and seamless HPC integration is promising to accelerate RL-driven fluid-mechanics studies.
title SmartFlow: A CFD-solver-agnostic deep reinforcement learning framework for computational fluid dynamics on HPC platforms
topic Fluid Dynamics
Computational Physics
url https://arxiv.org/abs/2508.00645