定常来流下钝体二维组合断面气动特性与流场的智能预测方法

李珂; 王路路; 陈增顺; 赵文卓; 秦煜; 李少鹏

doi:10.7638/kqdlxxb-2025.0036

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 >

定常来流下钝体二维组合断面气动特性与流场的智能预测方法

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

- 定常来流下钝体二维组合断面气动特性与流场的智能预测方法
- Intelligent prediction method of flow field and aerodynamic characteristics for two-dimensional blunt body combined sections in steady wind
- 空气动力学学报 2025年43卷第5期页码：112-123
- 作者机构：
  
  重庆大学土木工程学院,重庆,400045
- 作者简介：
  
  李珂（1989—），男，四川人，副教授，研究方向：智能风工程. E-mail：keli-bridge@cqu.edu.cn
  陈增顺*，教授，研究方向：结构风工程. E-mail：zengshunchen@cqu.edu.cn
- 基金信息：
  
  重庆市杰出青年基金（2022NSCQ-JQX2377）
- DOI：10.7638/kqdlxxb-2025.0036
  中图分类号： TU312.1;V211.3;TP183
- 纸质出版：2025
- 稿件说明：
移动端阅览
李珂, 王路路, 陈增顺, 等. 定常来流下钝体二维组合断面气动特性与流场的智能预测方法[J]. 空气动力学学报, 2025, 43(5): 112−123.

LI Ke, WANG Lulu, CHEN Zengshun, et al. Intelligent prediction method of flow field and aerodynamic characteristics for two-dimensional blunt body combined sections in steady wind[J]. 2025, 43(5): 112-123.
李珂, 王路路, 陈增顺, 等. 定常来流下钝体二维组合断面气动特性与流场的智能预测方法[J]. 空气动力学学报, 2025, 43(5): 112−123. DOI： 10.7638/kqdlxxb-2025.0036

LI Ke, WANG Lulu, CHEN Zengshun, et al. Intelligent prediction method of flow field and aerodynamic characteristics for two-dimensional blunt body combined sections in steady wind[J]. 2025, 43(5): 112-123. DOI： 10.7638/kqdlxxb-2025.0036.

摘要

结构断面气动性能的高效精准获取对于设计优化至关重要。特别是在处理结构断面设计参数多样性和优化方向不确定性的迭代计算任务时，为了解决效率低下的问题，本文提出一种基于深度学习代理模型的智能预测方法，对定常风下钝体二维组合断面的绕流流场及三分力系数进行快速精准预测。具体而言，该方法采用类似图片的一致化形状表达来描述钝体组合断面的气动外形，具有不受断面形式限制的通用性；通过融合卷积注意力机制模块与残差模块构建神经网络架构，并利用均方误差来捕捉神经网络预测误差，实现了从气动外形到绕流流场及三分力系数的强非线性映射。研究结果表明，该方法在定常风条件下，对钝体二维组合断面的时均绕流流场、表面压力分布以及三分力系数的预测误差分别控制在3.7%、0.35%和6.25%以内，满足精度要求，且与传统CFD模拟计算相比，计算效率提升了4个数量级。该方法为定常风下钝体断面气动性能的快速预测提供了一种高效、实用的技术手段。

Abstract

In the design and optimization of structural cross-sections

the efficient and accurate evaluation of aerodynamic performance is of significant importance. To address the inefficiency in iterative computations involving diverse design parameters‌

this study proposed an intelligent prediction method based on a deep learning surrogate model

focusing on the rapid and accurate prediction of flow fields around two-dimensional combined bluff body cross-sections and static force coefficients under steady wind conditions. Specifically

this method employed a unified image-like shape representation to characterize the aerodynamic shapes of bluff body combined cross-sections

ensuring broad applicability without being limited by specific cross-section configurations. By integrating convolutional attention mechanisms and residual modules to construct the neural network architecture and using mean squared error to capture prediction errors

the method achieved a highly nonlinear mapping from aerodynamic shapes to flow characteristics and static force coefficients. The model achieves prediction errors within ‌3.7% for velocity fields‌

‌0.35% for surface pressure‌

and ‌6.25% for force coefficients‌ under steady wind

meeting all accuracy requirements. Additionally

the computational efficiency was improved by four orders of magnitude compared to conventional CFD. This method provides an efficient and practical solution for the rapid prediction of aerodynamic performance for bluff body cross-sections under steady wind environments.

关键词

Keywords

references

XIE Z N, GU M. Wind-induced interference effects of arbitrary arrangement of twin cylindrical bodies[J]. Journal of Civil Engineering, 2005, 38(10): 32−38 (in Chinese). doi: 10.3321/j.issn:1000-131X.2005.10.006

KIM Y C, JUN K D, TAMURA Y. Wind-induced coupled motion of tall buildings with varying square plan with height[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2011, 99(5): 638−650. doi: 10.1016/j.jweia.2011.03.004

HE X H, LI H, WANG H F, et al. Effects of geometrical parameters on the aerodynamic characteristics of a streamlined flat box girder[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2017, 170: 56−67. doi: 10.1016/j.jweia.2017.08.009

HU C X, ZHOU Z Y, SUN Q. Effects of bridge decks on stability of long-span cable-stayed bridge[J]. Bridge Construction, 2018, 48(6): 53−57 (in Chinese). doi: 10.3969/j.issn.1003-4722.2018.06.010

WU Q Y, SUN Y S, LIU X B. Study on interference effect of aerodynamic characteristics of two parallel columns[J]. Engineering Mechanics, 2020, 37(S1): 265−269 (in Chinese). doi: 10.6052/j.issn.1000-4750.2019.04.S007

YANG J, XU F Y, ZENG D L. Experimental study on the aerodynamic characteristics of a rectangular cross-section with a width-to-height ratio of 10[J]. Vibration and Shock, 2020, 39: 109–115, 129 (in Chinese).

MADSEN J I, SHYY W, HAFTKA R T. Response surface techniques for diffuser shape optimization[J]. AIAA Journal, 2000, 38: 1512−1518. doi: 10.2514/3.14576

GAO Y Q, WANG H, XU Z D, et al. Numerical simulation of static tri-component coefficients of girder sections of a long-span high-speed railway continuous beam bridge[J]. Journal of Nanjing Tech University (Natural Science Edition), 2020, 42(3): 358−365 (in Chinese).

ZUO Y. Study on aerodynamic characteristics of composite beams with 10∶1 aspect-to-height ratio of side girders with different Liang Kuan[J]. Sichuan Architecture, 2024, 44(2): 182−185 (in Chinese). doi: 10.3969/j.issn.1007-8983.2024.02.062

CELIK I, KLEIN M, FREITAG M, et al. Assessment measures for URANS/DES/LES: An overview with applications[J]. Journal of Turbulence, 2006, 7: 48. doi: 10.1080/14685240600794379

CHEN X C, WANG Y. The unsteady numerical simulation of separated flow around bluff bodies[J]. Acta Aerodynamica Sinica, 2007, 25(4): 513−520 (in Chinese). doi: 10.3969/j.issn.0258-1825.2007.04.019

CHEN H, QIAN W Q, HE L. Aerodynamic coefficient prediction of airfoils based on deep learning[J]. Acta Aerodynamica Sinica, 2018, 36(2): 294−299 (in Chinese).

MIYANAWALA T P, JAIMAN R K. A novel deep learning method for the predictions of current forces on bluff bodies[C]// ASME 2018 37th International Conference on Ocean, Offshore and Arctic Engineering. Madrid, Spain, 2018. doi: 10.1115/OMAE2018-78338

GUO X X, LI W, IORIO F. Convolutional neural networks for steady flow approximation[C]// 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco California USA. New York: ACM, 2016: 481–490. doi: 10.1145/2939672.2939738

ABBAS T, KAVRAKOV I, MORGENTHAL G, et al. Prediction of aeroelastic response of bridge decks using artificial neural networks[J]. Computers Structures, 2020, 231: 106198. doi: 10.1016/j.compstruc.2020.106198

LI K, LI H, LI S P, et al. Fully convolutional neural network prediction method for aerostatic performance of bluff bodies based on consistent shape description[J]. Applied Sciences, 2022, 12(6): 3147. doi: 10.3390/app12063147

LIN P F, YOU C Y, WANG Q L, et al. Deep learning-based fast prediction of flow field around multiple bluff bodies[J]. Advances in Wind Engineering, 2024, 1(2): 100020. doi: 10.1016/j.awe.2024.100020

BERNARDINI E, SPENCE S M J, WEI D, et al. Aerodynamic shape optimization of civil structures: A CFD-enabled Kriging-based approach[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2015, 144: 154−164. doi: 10.1016/j.jweia.2015.03.011

CHEN D L, GAO X, XU C F, et al. FlowGAN: A conditional generative adversarial network for flow prediction in various conditions[C]// Proc of the 2020 IEEE 32nd International Conference on Tools with Artificial Intelligence (ICTAI), Baltimore, MD, USA. IEEE, 2020: 315-322. doi: 10.1109/ictai50040.2020.00057

WOO S, PARK J, LEE J Y, et al. CBAM: Convolutional block attention module[C]// Computer Vision-ECCV 2018. Cham: Springer International Publishing, 2018: 3-19. doi: 10.1007/978-3-030-01234-2_1

ALVAREZ A J, NIETO F, KWOK K C S, et al. Aerodynamic performance of twin-box decks: A parametric study on gap width effects based on validated 2D URANS simulations[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2018, 182: 202−221. doi: 10.1016/j.jweia.2018.09.004

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

基于U-Net的磁浮飞行风洞气动力干扰剥离算法

基于迁移学习的点云机翼气动性能预测

基于条件扩散模型的火箭气动外形快速设计方法

1∶5矩形断面速度场降阶动力学模态智能预测模型

基于物理约束深度学习的大跨柔性光伏阵列绕流场重构