基于深度强化学习的箱梁涡激振动智能流动控制

邓晓龙; 胡钢; 陈文礼

doi:10.7638/kqdlxxb-2025.0016

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于深度强化学习的箱梁涡激振动智能流动控制

研究专栏——大型结构风效应流固耦合与人工智能 | 更新时间：2026-05-28

- 基于深度强化学习的箱梁涡激振动智能流动控制
- Deep reinforcement learning based intelligent flow control of vortex-induced vibration for box girder
- 空气动力学学报 2025年43卷第5期页码：92-100
- 作者机构：
  
  1. 哈尔滨工业大学(深圳) 智能土木与海洋工程学院,深圳,518055
  2. 哈尔滨工业大学土木工程学院,哈尔滨,150096
- 作者简介：
  
  邓晓龙（1995—），男，哈尔滨人，博士生，研究方向：桥梁风工程. E-mail：21B954013@stu.edu.com
  胡钢*，教授，研究方向：风工程. E-mail：hugang@hit.edu.cn
- 基金信息：
  
  深圳市科技计划-稳定支持计划（GXWD20231129111527001）
- DOI：10.7638/kqdlxxb-2025.0016
  中图分类号： V214.3;TP181;TU311.3
- 纸质出版：2025
- 稿件说明：
移动端阅览
邓晓龙, 胡钢, 陈文礼. 基于深度强化学习的箱梁涡激振动智能流动控制[J]. 空气动力学学报, 2025, 43(5): 92−100.

DENG Xiaolong, HU Gang, CHEN Wenli. Deep reinforcement learning based intelligent flow control of vortex-induced vibration for box girder[J]. 2025, 43(5): 92-100.
邓晓龙, 胡钢, 陈文礼. 基于深度强化学习的箱梁涡激振动智能流动控制[J]. 空气动力学学报, 2025, 43(5): 92−100. DOI： 10.7638/kqdlxxb-2025.0016

DENG Xiaolong, HU Gang, CHEN Wenli. Deep reinforcement learning based intelligent flow control of vortex-induced vibration for box girder[J]. 2025, 43(5): 92-100. DOI： 10.7638/kqdlxxb-2025.0016.

摘要

为抑制桥梁涡激振动从而提升桥梁抗风性能，提出了一种基于深度强化学习的智能主动流动控制方法。研究利用合成射流对桥梁节段气动弹性模型尾流场的旋涡脱落进行扰动，从而实现对涡激振动的有效抑制。风洞试验结果验证了桥梁涡激振动在均匀稳定风场中的气动性能，并建立了合成射流激励器控制电压与射流流量的映射关系。通过对控制电压的系统性分析发现，控制电压与射流平均吹气速度近似线性正相关，且更高的控制电压能够显著提升抑制效果。随后，结合软演员-评论家（soft actor-critic

SAC）算法对合成射流的控制策略进行了优化，快速训练得出最优控制电压，并将振幅减少了83%。研究结果表明，合成射流结合深度强化学习算法能够高效抑制桥梁涡激振动，为桥梁风工程的抗风设计提供智能化解决方案。

Abstract

An intelligent active flow control method based on deep reinforcement learning (DRL) was proposed to suppress vortex-induced vibrations and enhance the wind resistance of bridges. This method employed synthetic jets to disturb the wake vortex shedding of an aeroelastic bridge section model

effectively suppressing vortex-induced vibrations. Wind tunnel tests were conducted to validate the aerodynamic performance of the model under uniform and steady wind conditions. Additionally

the relationship between the control voltage and synthetic jet flow rate was established. A systematic analysis showed that the control voltage was approximately linearly and positively correlated with the average jet velocity

with higher control voltages significantly improving the suppression effect. Subsequently

the synthetic jet control strategy was optimized using the Soft Actor-Critic (SAC) algorithm

which converged rapidly to the optimal control voltage

resulting in a maximum reduction of 83% in vibration amplitude. These findings demonstrate that combining synthetic jet technology with DRL algorithms provides an efficient and intelligent solution for suppressing bridge vortex-induced vibrations and offers an intelligent approach for wind-resistant bridge design.

关键词

Keywords

references

DENG X L, HU G, CHEN W L, et al. Active control of bridge aerodynamic flow driven by deep reinforcement learning[J]. China Journal of Highway and Transport, 2023, 36(8): 66−75(in Chinese).

XU F Y, DING W, JIANG F, et al. Development and prospect of study on vortex-induced vibration of long-span bridges[J]. Journal of Vibration and Shock, 2010, 29(10): 40−49,249(in Chinese). doi: 10.3969/j.issn.1000-3835.2010.10.009

ZHAO L, LI K, WANG C J, et al. Review on passive aerodynamic countermeasures on main girders aiming at wind-induced stabilities of long-span bridges[J]. China Journal of Highway and Transport, 2019, 32(10): 34−48(in Chinese).

ZHAO L, LIU C J, GE Y J. Vortex-induced vibration sensitivity of bridge girder structures[J]. Acta Aerodynamica Sinica, 2020, 38(4): 694−704(in Chinese). doi: 10.7638/kqdlxxb-2020.0105

LARSEN A, ESDAHL S, ANDERSEN J E, et al. Storebælt suspension bridge–vortex shedding excitation and mitigation by guide vanes[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2000, 88(2-3): 283−296. doi: 10.1016/S0167-6105(00)00054-4

LI Y L, HOU G Y, XIANG H Y, et al. Optimization of the vortex induced vibration for steel box girder of long span suspension bridges by wind tunnel test[J]. Acta Aerodynamica Sinica, 2011, 29(6): 702−708 (in Chinese). doi: 10.3969/j.issn.0258-1825.2011.06.004

REN F, WANG C L, TANG H. Bluff body uses deep-reinforcement-learning trained active flow control to achieve hydrodynamic stealth[J]. Physics of Fluids, 2021, 33(9): 093602. doi: 10.1063/5.0060690

RABAULT J, KUHNLE A. Accelerating deep reinforcement learning strategies of flow control through a multi-environment approach[J]. Physics of Fluids, 2019, 31(9): 094105. doi: 10.1063/1.5116415

WANG Q L, YAN L, HU G, et al. DRLinFluids: An open-source Python platform of coupling deep reinforcement learning and OpenFOAM[J]. Physics of Fluids, 2022, 34(8): 081801. doi: 10.1063/5.0103113

TANG H W, RABAULT J, KUHNLE A, et al. Roust active flow control over a range of Reynolds numbers using deep reinforcement learning[C]// 14th World Congress in Computational Mechanics and ECCOMAS Congress, Paris, France, 2020.

FAN D, YANG L, WANG Z, et al. Reinforcement learning for bluff body active flow control in experiments and simulations[J]. Proceedings of the National Academy of Sciences, 2020, 117(42): 26091−26098. doi: 10.1073/pnas.2004939117

LI K, GE Y J, ZHAO L, et al. Numerical simulation of feedback flutter control for a single-box-girder suspension bridge by twin-winglet system[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2017, 169: 77−93. doi: 10.1016/j.jweia.2017.07.013

DENG X L, HU G, CHEN W L. Intelligent active flow control of long-span bridge deck using deep reinforcement learning integrated transfer learning[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2024, 244: 105632. doi: 10.1016/j.jweia.2023.105632

DONG X H, WANG Z R, LIN P F, et al. An interactive platform of deep reinforcement learning and wind tunnel testing[J]. Physics of Fluids, 2024, 36(11): 115197. doi: 10.1063/5.0238959

ZHAN J, XIN D B, OU J P, et al. Experimental study on suppressing vortex-induced vibration of a long-span bridge by installing the wavy railings[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2020, 202: 104205. doi: 10.1016/j.jweia.2020.104205

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

不同尾流下斑海豹胡须涡激振动特性

影响振荡射流分离控制效果的关键因素

输电塔钢管涡激振动控制措施的风洞试验研究

背风面合成射流作用下方柱绕流统计特性

合成双射流控制水下圆柱绕流流动分离数值模拟研究