基于条件扩散的变形结构气动热预测方法

何纪云; 吕宏强; 李旭东; 虞建; 许冉; 张俊龙; 刘学军

doi:10.7638/kqdlxxb-2024.0176

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 >

基于条件扩散的变形结构气动热预测方法

研究论文 | 更新时间：2026-05-28

- 基于条件扩散的变形结构气动热预测方法
- An aeroheating prediction method for deformed structures based on conditional diffusion model
- 空气动力学学报 2025年43卷第8期页码：22-35
- 作者机构：
  
  1. 南京航空航天大学人工智能学院，脑机智能技术教育部重点实验室,南京,211106
  2. 南京航空航天大学航空学院,南京,210016
  3. 北京航天长征飞行器研究所,北京,100071
- 作者简介：
  
  何纪云（1994—），男，安徽人，硕士研究生，研究方向：机器学习与人工智能. E-mail：hejiyun@nuaa.edu.cn
  刘学军*（1976—），女，教授，研究方向：机器学习. E-mail：xuejun.liu@nuaa.edu.cn
- 基金信息：
  
  国家自然科学基金面上项目（12472236）
- DOI：10.7638/kqdlxxb-2024.0176
  中图分类号： TP181;V211.3
- 纸质出版：2025
- 稿件说明：
移动端阅览
何纪云, 吕宏强, 李旭东, 等. 基于条件扩散的变形结构气动热预测方法[J]. 空气动力学学报, 2025, 43(8): 22−35.

HE Jiyun, LYU Hongqiang, LI Xudong, et al. An aeroheating prediction method for deformed structures based on conditional diffusion model[J]. 2025, 43(8): 22-35.
何纪云, 吕宏强, 李旭东, 等. 基于条件扩散的变形结构气动热预测方法[J]. 空气动力学学报, 2025, 43(8): 22−35. DOI： 10.7638/kqdlxxb-2024.0176

HE Jiyun, LYU Hongqiang, LI Xudong, et al. An aeroheating prediction method for deformed structures based on conditional diffusion model[J]. 2025, 43(8): 22-35. DOI： 10.7638/kqdlxxb-2024.0176.

摘要

伸缩变形结构飞行器外形复杂，气动热数据分布差异较大，代理模型难以准确捕获其物面气动热数据的分布。为了对变形结构飞行器物面气动热进行准确预测，本文基于条件扩散模型提出了一种变形结构气动热预测方法（heating-MLP diffusion

HMD）。该方法包括前向扩散与逆向去噪两个过程。在前向扩散过程中，对原始气动热数据进行逐步加噪，直至成为纯高斯噪声；在逆向去噪过程中，将变形结构外形和工况参数作为条件，利用全连接神经网络预测扩散过程中每步添加的噪声，从而学习隐含的气动热数据分布特性，最终实现飞行器伸缩变形机翼物面网格点的气动热预测。基于数值仿真数据的模型验证结果表明，相较于高斯过程、神经过程和全连接神经网络，基于条件扩散模型的气动热预测方法能够取得更好的预测效果，平均绝对百分比误差在10%左右。该方法可为伸缩变形结构高速飞行器机翼物面气动热计算提供一种精确预测模型。

Abstract

The shape of aircraft with ‌telescopic deformed structures‌ is complex

and the distribution of ‌aerothermal‌ data varies ‌significantly‌. Traditional surrogate models to capture the ‌aerothermal‌ heating distribution of ‌telescopic structures‌

hindering effective prediction struggle on structural surfaces. Based on ‌the conditional diffusion model

‌the Heating-MLP Diffusion (HMD) method for deformable structures‌ was proposed

comprising two processes: forward diffusion and reverse denoising. In the forward diffusion process

the original ‌aerothermal‌ data is gradually ‌corrupted‌ until it becomes pure Gaussian noise. In the reverse denoising process

‌using‌ the shape and operating conditions of the deformed structure as ‌conditional inputs‌

a fully connected neural network predicts the noise added at each diffusion step

thereby learning the implicit ‌distribution characteristics of aerothermal‌ heating data

thus enabling ‌the prediction of aerothermal heating‌ on the surface grid points of ‌aircraft telescopic deformed wings‌. Numerical simulation data validated the proposed model. Experimental results demonstrate that compared with Gaussian processes

neural processes

and neural networks

the ‌conditional diffusion model-based‌ aerothermal prediction method achieves higher accuracy

with a mean absolute percentage error of ～10%. This evidence proves its effectiveness in predicting ‌aerothermal heating‌ on ‌high-speed aircraft‌ wings with telescopic deformed structures providing an accurate prediction model for engineering applications.

关键词

Keywords

references

PENG Z Y, SHI Y L, GONG H M, et al. Hypersonic aeroheating prediction technique and its trend of development[J]. Acta Aeronautica et Astronautica Sinica, 2015, 36(1): 325−345 (in Chinese).

LUO C T, HU Z M, LIU Y F, et al. Research progress on ground-to-flight correlation of aerodynamic force and heating data from hypersonic wind tunnels[J]. Journal of Experiments in Fluid Mechanics, 2020, 34(3): 78−89 (in Chinese). doi: 10.11729/syltlx20200006

YU C Z, LIU W H. Research status of aeroheating prediction technology for hypersionic aircraft[J]. Aeronautical Science Technology, 2021, 32(2): 14−21 (in Chinese).

YAN C, YU J J, LI J Z. Scheme effect and grid dependency in CFD computations of heat transfer[J]. Acta Aerodynamica Sinica, 2006, 24(1): 125−130 (in Chinese). doi: 10.3969/j.issn.0258-1825.2006.01.023

ZHAO R, WEN C Y, TIAN X D, et al. Numerical simulation of local wall heating and cooling effect on the stability of a hypersonic boundary layer[J]. International Journal of Heat and Mass Transfer, 2018, 121: 986−998. doi: 10.1016/j.ijheatmasstransfer.2018.01.054

LEE J H, RHO O. Accuracy of AUSM+ scheme in hypersonic blunt body flow calculations[R]. AIAA Aerospace Research Central, 1998. doi: 10.2514/6.1998-1538

KLOPFER G, YEE H. Viscous hypersonic shock-on-shock interaction on blunt cowl lips[C]//26th Aerospace Sciences Meeting, Reno, NV, USA. AIAA, 1988: AIAA 1988-233. doi: 10.2514/6.1988-233

WANG Z Y, ZHANG W W, LIU L. Review of simulation methods of hypersonic aerothermoelastic problems[J]. Physics of Gases, 2020, 5(6): 1−15 (in Chinese).

TAO Y, ZONG Q, ZENG F L. Research on data fitting method of aerodynamic model of hypersonic vehicle[C]//Proceedings of the 31st Chinese Control Conference. Beijing: Technical Committee on Control Theory, Chinese Association of Automation, 2012: 1275−1278 (in Chinese).

DOWELL E H. Eigenmode analysis in unsteady aerodynamics - reduced-order models[J]. AIAA Journal, 1996, 34(8): 1578−1583. doi: 10.2514/3.13274

SILVA W A. Identification of linear and nonlinear aerodynamic impulse responses using digital filter techniques[C]// 22nd Atmospheric Flight Mechanics Conference, New Orleans, LA, USA. AIAA, 1997: AIAA 1997-3712. doi: 10.2514/6.1997-3712

SIMPSON T W, KORTE J J, MAUERY T M, et al. Kriging models for global approximation in simulation-based multidisciplinary design optimization[J]. AIAA Journal, 2001, 39(12): 2233−2241. doi: 10.2514/2.1234

NIE C S, HUANG J D, WANG X, et al. Fast aeroheating prediction method for complex shape vehicles based on proper orthogonal decomposition[J]. Acta Aerodynamica Sinica, 2017, 35(6): 760−765 (in Chinese).

YANG F, LIN M Y, HU Z M, et al. Prediction method of aero-heating of hypersonic vehicle bi-curvature leading edge based on machine learning[J]. Journal of Beijing University of Aeronautics and Astronautics, 2024, 50(9): 2826−2834 (in Chinese). doi: 10.13700/j.bh.1001-5965.2022.0746

KOU J Q, ZHANG W W. Research on the effects of basis function widths of aerodynamic modeling based on recursive RBF neural network[J]. Advances in Aeronautical Science and Engineering, 2015, 6(3): 261−270 (in Chinese). doi: 10.3969/j.issn.1674-8190.2015.03.001

ZHANG Z C, GAO T Y, ZHANG L, et al. Aeroheating agent model based on radial basis function neural network[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(4): 524167 (in Chinese).

YUAN J C, ZONG W G, ZENG L, et al. CNN-based method for predicting aerodynamic heating[J]. Acta Aerodynamica Sinica, 2024, 42(1): 13−25 (in Chinese). doi: 10.7638/kqdlxxb-2023.0072

KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84−90. doi: 10.1145/3065386

SOHL-DICKSTEIN J, WEISS E A, MAHESWARANATHAN N, et al. Deep unsupervised learning using nonequilibrium thermodynamics[C]// Proceedings of the 32nd International Conference on International Conference on Machine Learning - Volume 37, Lille, France. ACM, 2015: 2256−2265. doi: 10.5555/3045118.3045358

HO J, JAIN A, ABBEEL P, et al. Denoising diffusion probabilistic models[C]// 34th International Conference on Neural Information Processing Systems, Vancouver, BC, Canada. ACM, 2020: 6840-6851. doi: 10.5555/3495724.3496298

DHARIWAL P, NICHOL A, DHARIWAL P, et al. Diffusion models beat GANs on image synthesis[C]// 35th International Conference on Neural Information Processing Systems, ACM, 2021: 8780−8794. doi: 10.5555/3540261.3540933

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

基于深度强化学习的三维变形机翼反设计方法

一种数据驱动的气动热预示模型

直升机旋翼翼型需求分析及技术发展展望

大展弦比飞行器突风响应数值计算

热防护材料动态烧蚀形貌结构光视觉测量技术