基于深度神经网络的高阶非线性激波判别式

杨华; 陈树生; 李猛; 庞超; 高正红; 向星皓

doi:10.7638/kqdlxxb-2022.0082

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 >

基于深度神经网络的高阶非线性激波判别式

研究论文 | 更新时间：2026-05-28

- 基于深度神经网络的高阶非线性激波判别式
- Higher-order nonlinear shock wave discriminant based on deep neural network
- 空气动力学学报 2023年41卷第7期页码：56-63
- 作者机构：
  
  1. 西北工业大学航空学院,西安,710072
  2. 中航工业第一飞机设计研究院,西安,710089
  3. 空气动力学国家重点实验室,绵阳,621000
- 作者简介：
  
  杨华（1996-），男，陕西西安人，博士研究生，研究方向：飞行器设计. E-mail：hcyanghua@mail.nwpu.edu.cn
  陈树生*（1991-），男，福建泉州人，副教授，研究方向：计算流体力学、飞行动力学及飞行器设计. E-mail：sshengchen@nwpu.edu.cn
- 基金信息：
  
  中国科协青年人才托举工程(2022QNRC001)
- DOI：10.7638/kqdlxxb-2022.0082
  中图分类号： V211.3
- 纸质出版：2023
- 稿件说明：
移动端阅览
杨华, 陈树生, 李猛, 等. 基于深度神经网络的高阶非线性激波判别式[J]. 空气动力学学报, 2023, 41(7): 56−63.

YANG Hua, CHEN Shusheng, LI Meng, et al. Higher-order nonlinear shock wave discriminant based on deep neural network[J]. 2023, 41(7): 56-63.
杨华, 陈树生, 李猛, 等. 基于深度神经网络的高阶非线性激波判别式[J]. 空气动力学学报, 2023, 41(7): 56−63. DOI： 10.7638/kqdlxxb-2022.0082

YANG Hua, CHEN Shusheng, LI Meng, et al. Higher-order nonlinear shock wave discriminant based on deep neural network[J]. 2023, 41(7): 56-63. DOI： 10.7638/kqdlxxb-2022.0082.

摘要

超声速流动数值模拟中准确识别激波位置对求解飞行器气动特性十分关键，传统的人工构造激波识别方法严重依赖于经验参数，在实际应用中存在一定局限性。为了建立更加通用的激波识别方法，采用深度神经网络构建了高阶非线性激波判别式。在生成了训练数据集与测试数据集的基础上，构建深度前馈神经网络并训练该模型使其满足激波识别预测精度，并通过一维Lax、Sod激波问题及二维超声速无黏圆柱绕流算例验证了该判别式的准确性。结果表明，本文构建的高阶非线性激波判别式相较于人工构造的激波识别器能更加精准地判别光滑区域和激波区域，在不同来流马赫数、不同网格分布条件下均能准确识别出激波位置，但激波识别精度依赖于网格分辨率。

Abstract

In the numerical simulation of supersonic flow

accurately identifying shock wave position is essential to solve an aircraft's aerodynamic and thermal characteristics. The traditional artificial shock identification method seriously depends on empirical coefficients

which has some limitations in practical application. A higher-order nonlinear shock wave discriminant is constructed by using a deep neural network to establish a more general shock identification method. Based on the training data set and the test data set

the depth Feedforward Neural Network is constructed

and the model is trained to meet the prediction accuracy of shock wave identification; it is applied to one-dimensional Lax and Sod shock problems and two-dimensional supersonic inviscid circular cylinder to verify the accuracy of the discriminant. The results show that this paper's high-order nonlinear shock discriminant can more accurately distinguish the smooth region and shock region than the artificial shock discriminant. Secondly

the shock discriminant can accurately identify the shock position under different Mach numbers and grid distributions

but the accuracy of shock identification depends on the grid resolution.

关键词

Keywords

references

YAN C. Computational fluid dynamic's methods and applications [M]. Beijing: Beijing University of Aeronautics Astronautics Press, 2006 (in Chinese).

侯中喜. 超声速复杂流场并行数值分析及高阶格式研究[D]. 长沙: 中国人民解放军国防科学技术大学, 2000. HOU Z X. Supersonic complex flowfield parallel analysis and high order schemes study[D]. Changsha: National University of Defense Technology, 2000 (in Chinese).

肖丰收. 若干典型高超声速激波干扰流动特性研究[D]. 合肥: 中国科学技术大学, 2016. XIAO F S. Research on flow characteristics of some typical hypersonic shock wave interactions[D]. Hefei: University of Science and Technology of China, 2016(in Chinese).

LARSSON J, COOK A, LELE S K, et al. Computational issues and algorithm assessment for shock/turbulence interaction problems[J]. Journal of Physics: Conference Series, 2007, 78: 012014. DOI:10.1088/1742-6596/78/1/012014[LinkOut]

常思源, 白晓征, 刘君. 一种基于聚类分析的二维激波模式识别算法[J]. 航空学报, 2020, 41(8): 123626. CHANG S Y, BAI X Z, LIU J. A two-dimensional shock wave pattern recognition algorithm based on cluster analysis[J]. Acta Aeronautica et Astronautica Sinica, 2020, 41(8): 123626(in Chinese). doi: 10.7527/S1000-6893.2019.23626

RAY D, HESTHAVEN J S. An artificial neural network as a troubled-cell indicator[J]. Journal of Computational Physics, 2018, 367: 166-191. DOI:10.1016/j.jcp.2018.04.029[LinkOut]

RAY D, HESTHAVEN J S. Detecting troubled-cells on two-dimensional unstructured grids using a neural network[J]. Journal of Computational Physics, 2019, 397: 108845. DOI:10.1016/j.jcp.2019.07.043[LinkOut]

ABGRALL R, HAN VEIGA M. Neural network-based limiter with transfer learning[J]. Communications on Applied Mathematics and Computation, 2020: 1-41. DOI:10.1007/s42967-020-00087-1[LinkOut]

SUN Z, WANG S Y, CHANG L B, et al. Convolution neural network shock detector for numerical solution of conservation laws[J]. Communications in Computational Physics, 2020, 28(5): 2075-2108. DOI:10.4208/cicp.oa-2020-0199[LinkOut]

HUYNH H. Accurate upwind schemes for the Euler equations[C]//Proc of the 12th Computational Fluid Dynamics Conference, San Diego, CA. Reston, Virginia: AIAA, 1995: 1737. doi: 10.2514/6.1995-1737

GEROLYMOS G A, SÉNÉCHAL D, VALLET I. Very-high-order WENO schemes[J]. Journal of Computational Physics, 2009, 228(23): 8481-8524. DOI:10.1016/j.jcp.2009.07.039[LinkOut]

QIU X P. Neural networks and deep learning[M]. Beijing: China Machine Press, 2020(in Chinese).

李彦苏, 阎超, 屈峰, 等. 三阶中心无振荡格式性能分析[J]. 北京航空航天大学学报, 2014, 40(11): 1576-1580. LI Y S, YAN C, QU F, et al. Analysis for centered nonoscillatory scheme of third order[J]. Journal of Beijing University of Aeronautics and Astronautics, 2014, 40(11): 1576-1580(in Chinese). [知网中文][知网英文]

CHEN B, SHEN Y Q. Research on adaptive artificial viscosity and shock detecting method for hyperbolic systems of conservation laws[C]// LHD 2013 Summer Academic Symposium, Pingshan, 2013: 111-117 (in Chinese).

CHEN S S, CAI F J, XUE H C, et al. An improved AUSM-family scheme with robustness and accuracy for all Mach number flows[J]. Applied Mathematical Modelling, 2020, 77: 1065-1081. DOI:10.1016/j.apm.2019.09.005[LinkOut]

CHEN S S, YANG H, CAI F J, et al. Multi-dimensional dissipation strategy within advection upstream splitting methods in hypersonic flows[J]. Journal of Physics: Conference Series, 2021, 1786(1): 012050. DOI:10.1088/1742-6596/1786/1/012050[LinkOut]

CHEN S S, YAN C, LIN B X, et al. Affordable shock-stable item for Godunov-type schemes against carbuncle phenomenon[J]. Journal of Computational Physics, 2018, 373: 662-672. DOI:10.1016/j.jcp.2018.07.022[LinkOut]

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

无人机气动/风能获取一体化设计

基于机器学习的房颤病人左心耳血流动力学研究

基于射流流动控制的飞翼布局飞行器阵风载荷减缓研究

基于风雷软件的壁面距离场的快速迭代并行求解方法

基于多支微射流的超声速射流主动控制