Study on Liutex-driven reward for intelligent flow control by dynamic feature-based deep reinforcement learning

The issue of flow separation over an airfoil under weak turbulent conditions is addressed and resolved through the deep reinforcement learning (DRL) strategy. To suppress the generation of separation flow and the instability of the vortex shedding alley over an airfoil, a novel reward function named...

Full description

Saved in:
Bibliographic Details
Main Authors: Qi Wang, Xiangrui Dong, Sunyu You, Xiaoshu Cai
Format: Article
Language:English
Published: AIP Publishing LLC 2025-05-01
Series:AIP Advances
Online Access:http://dx.doi.org/10.1063/5.0271616
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850209023843893248
author Qi Wang
Xiangrui Dong
Sunyu You
Xiaoshu Cai
author_facet Qi Wang
Xiangrui Dong
Sunyu You
Xiaoshu Cai
author_sort Qi Wang
collection DOAJ
description The issue of flow separation over an airfoil under weak turbulent conditions is addressed and resolved through the deep reinforcement learning (DRL) strategy. To suppress the generation of separation flow and the instability of the vortex shedding alley over an airfoil, a novel reward function named RLiutex, considering the suppression of the rigid rotation intensity and core size of the vortex, is first proposed in this paper. The great potential of this Liutex-driven reward for effective large eddy elimination and aerodynamic optimization is verified in this work. Furthermore, a dynamic feature-based DRL (DF-DRL) framework is redeveloped to markedly enhance the learning efficiency and convergence speed. The combination of the Liutex-driven reward function with the DF-DRL framework results in an exceptional aerodynamic performance, with minimal fluctuations in both drag and lift coefficients, highlighting the potential of this approach for advanced flow control strategies.
format Article
id doaj-art-694d3c08c97f4eaebbacc740d8d0bd13
institution OA Journals
issn 2158-3226
language English
publishDate 2025-05-01
publisher AIP Publishing LLC
record_format Article
series AIP Advances
spelling doaj-art-694d3c08c97f4eaebbacc740d8d0bd132025-08-20T02:10:07ZengAIP Publishing LLCAIP Advances2158-32262025-05-01155055117055117-1010.1063/5.0271616Study on Liutex-driven reward for intelligent flow control by dynamic feature-based deep reinforcement learningQi Wang0Xiangrui Dong1Sunyu You2Xiaoshu Cai3School of Energy and Power Engineering, University of Shanghai for Science and Technology, Shanghai 200093, ChinaSchool of Energy and Power Engineering, University of Shanghai for Science and Technology, Shanghai 200093, ChinaSchool of Energy and Power Engineering, University of Shanghai for Science and Technology, Shanghai 200093, ChinaSchool of Energy and Power Engineering, University of Shanghai for Science and Technology, Shanghai 200093, ChinaThe issue of flow separation over an airfoil under weak turbulent conditions is addressed and resolved through the deep reinforcement learning (DRL) strategy. To suppress the generation of separation flow and the instability of the vortex shedding alley over an airfoil, a novel reward function named RLiutex, considering the suppression of the rigid rotation intensity and core size of the vortex, is first proposed in this paper. The great potential of this Liutex-driven reward for effective large eddy elimination and aerodynamic optimization is verified in this work. Furthermore, a dynamic feature-based DRL (DF-DRL) framework is redeveloped to markedly enhance the learning efficiency and convergence speed. The combination of the Liutex-driven reward function with the DF-DRL framework results in an exceptional aerodynamic performance, with minimal fluctuations in both drag and lift coefficients, highlighting the potential of this approach for advanced flow control strategies.http://dx.doi.org/10.1063/5.0271616
spellingShingle Qi Wang
Xiangrui Dong
Sunyu You
Xiaoshu Cai
Study on Liutex-driven reward for intelligent flow control by dynamic feature-based deep reinforcement learning
AIP Advances
title Study on Liutex-driven reward for intelligent flow control by dynamic feature-based deep reinforcement learning
title_full Study on Liutex-driven reward for intelligent flow control by dynamic feature-based deep reinforcement learning
title_fullStr Study on Liutex-driven reward for intelligent flow control by dynamic feature-based deep reinforcement learning
title_full_unstemmed Study on Liutex-driven reward for intelligent flow control by dynamic feature-based deep reinforcement learning
title_short Study on Liutex-driven reward for intelligent flow control by dynamic feature-based deep reinforcement learning
title_sort study on liutex driven reward for intelligent flow control by dynamic feature based deep reinforcement learning
url http://dx.doi.org/10.1063/5.0271616
work_keys_str_mv AT qiwang studyonliutexdrivenrewardforintelligentflowcontrolbydynamicfeaturebaseddeepreinforcementlearning
AT xiangruidong studyonliutexdrivenrewardforintelligentflowcontrolbydynamicfeaturebaseddeepreinforcementlearning
AT sunyuyou studyonliutexdrivenrewardforintelligentflowcontrolbydynamicfeaturebaseddeepreinforcementlearning
AT xiaoshucai studyonliutexdrivenrewardforintelligentflowcontrolbydynamicfeaturebaseddeepreinforcementlearning