面向三维飞行器的气动性能智能预测模型及迁移学习框架研究

张鹏; 徐丹; 孙国鹏; 王岳青; 刘雨霜; 陈一; 陈坚强

doi:10.7638/kqdlxxb-2025.0049

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 >

面向三维飞行器的气动性能智能预测模型及迁移学习框架研究

研究论文 | 更新时间：2026-07-11

- 面向三维飞行器的气动性能智能预测模型及迁移学习框架研究
- An intelligent prediction model and transfer learning framework for aerodynamic performance of 3D aircraft
- 空气动力学学报 2026年44卷第6期页码：14-24
- 作者机构：
  
  1. 空天飞行空气动力科学与技术全国重点实验室,绵阳,621000
  2. 军事科学院战略评估咨询中心,北京,100091
- 作者简介：
  
  张鹏，研究方向：气动智能建模. E-mail：zhangpeng14f@nudt.edu.cn
  王岳青*，研究方向：气动智能建模. E-mail：qywang2013@163.com
- 基金信息：
  
  国家自然科学基金（12472300）
- DOI：10.7638/kqdlxxb-2025.0049
  中图分类号： V11;TP183
- 纸质出版：2026
- 稿件说明：
移动端阅览
张鹏, 徐丹, 孙国鹏, 等. 面向三维飞行器的气动性能智能预测模型及迁移学习框架研究[J]. 空气动力学学报, 2026, 44(6): 14−24.

Zhang Peng, Xu Dan, Sun Guopeng, et al. An intelligent prediction model and transfer learning framework for aerodynamic performance of 3D aircraft[J]. 2026, 44(6): 14-24.
张鹏, 徐丹, 孙国鹏, 等. 面向三维飞行器的气动性能智能预测模型及迁移学习框架研究[J]. 空气动力学学报, 2026, 44(6): 14−24. DOI： 10.7638/kqdlxxb-2025.0049

Zhang Peng, Xu Dan, Sun Guopeng, et al. An intelligent prediction model and transfer learning framework for aerodynamic performance of 3D aircraft[J]. 2026, 44(6): 14-24. DOI： 10.7638/kqdlxxb-2025.0049.

摘要

人工智能技术的飞速发展为飞行器气动性能计算模式的创新带来了机遇。本文面向三维飞行器发展了一种基于点云表征和Transformer架构的气动性能智能预测模型，对气动参数进行端到端预测，加速了气动问题求解过程。为解决智能模型的大样本需求与气动领域小样本现实之间的矛盾，本文基于模型参数复用方法和域对抗学习方法发展了面向气动性能预测任务的迁移学习框架，有效减小了模型对训练样本的需求。数据集构建上，基于自有三维飞行器仿真性能数据，构建了共包含上万样本的外形泛化、工况泛化及外形、工况双泛化测试数据集。实验结果表明，气动性能预测模型对单个三维飞行器的气动性能预测时间仅为3.096 s，计算效率较传统CFD方法提升2个数量级，预测结果的非标准平均相对误差仅为5.6%，此外，该模型在跨外形、跨工况等测试数据集上均表现出良好的泛化性；提出的迁移学习框架在保持模型同等精度水平（10%）的同时对训练样本量的需求降低了70%，可有效应对空气动力学中三维仿真数据稀缺的挑战，有望为飞行器设计优化、快速研制提供关键技术支撑。

Abstract

The rapid advancement of artificial intelligence technology has brought opportunities for innovation in the computational modeling of aircraft aerodynamic performance. This paper develops an intelligent prediction model for aerodynamic performance based on point cloud representation and Transformer architecture for three-dimensional aircraft

achieving end-to-end prediction of aerodynamic parameters and accelerating the process of solving aerodynamic problems. Furthermore

to address the contradiction between the large sample requirements of intelligent models and the small sample characteristics of the aerodynamic field

a transfer learning framework for aerodynamic performance prediction tasks is developed based on model parameter reuse methods and domain adversarial learning

effectively reducing the training sample requirements of the model. In terms of dataset construction

this paper builds a generalized test dataset containing over ten thousand samples

including shape generalization

condition generalization

and dual generalization

based on proprietary simulation performance data of three-dimensional aircraft. The results show that the aerodynamic performance prediction model requires only 3.096 s for predicting the aerodynamic performance of a single three-dimensional aircraft

achieving a computational efficiency improvement of two orders of magnitude compared to computational fluid dynamics (CFD) methods. The non-standard mean relative error of the prediction results is only 5.6%

and the model demonstrates excellent generalization across shape and condition test datasets. Under the transfer learning framework proposed in this paper

the aerodynamic performance prediction model can maintain the same accuracy level (10%) while reducing the training sample requirement by 70%

effectively addressing the challenge of scarce three-dimensional simulation data in aerodynamics. This approach is expected to provide key technical support for aircraft design optimization and rapid development.

关键词

Keywords

references

Antoine N E, Kroo I M. Framework for aircraft conceptual design and environmental performance studies[J]. AIAA Journal, 2005, 43(10): 2100−2109. doi: 10.2514/1.13017

Wang Xu, Ning Chenjia, Wang Wenzheng, et al. Intelligent fusion method of multi-source aerodynamic data for flight tests[J]. Acta Aerodynamica Sinica, 2023, 41(2): 12−20(in Chinese). doi: 10.7638/kqdlxxb-2021.0428

Che Jing, Tang Shuo. The engineering calculation of aerodynamics for quasi-waverider vehicle[J]. Acta Aerodynamica Sinica, 2007, 25(3): 381−385(in Chinese).

Wang Haifeng, Zhan Jingxia, Chen Ke, et al. Development and application of aerodynamic research technologies for fighters at high angle of attack[J]. Acta Aerodynamica Sinica, 2022, 40(1): 1−25(in Chinese). doi: 10.7638/kqdlxxb-2021.0306

Zhang Laiping, Deng Xiaogang, He Lei, et al. The opportunity and grand challenges in computational fluid dynamics by exascale computing[J]. Acta Aerodynamica Sinica, 2016, 34(4): 405−417(in Chinese). doi: 10.7638/kqdlxxb-2014.0118

Wu Zhi, Fan Dewei, Zhou Yu, et al. Advances in control of turbulence by artificial intelligence: systems, algorithms, achievements and data analysis methods[J]. Advances in Mechanics, 2023, 53(2): 273−307(in Chinese). doi: 10.6052/1000-0992-22-045

Zhang Weiwei, Zhu Linyang, Liu Yilang, et al. Progresses in the application of machine learning in turbulence modeling[J]. Acta Aerodynamica Sinica, 2019, 37(3): 444−454(in Chinese). doi: 10.7638/kqdlxxb-2019.0036

Zhang Weiwei, Wang Xu, Kou Jiaqing, et al. Prospects of multi-paradigm fusion methods for fluid mechanics research[J]. Advances in Mechanics, 2023, 53(2): 433−467(in Chinese). doi: 10.6052/1000-0992-22-050

Tang Zhigong, Zhu Linyang, Xiang Xinghao, et al. Some research progress and prospect of Intelligent Aerodynamics[J]. Acta Aerodynamica Sinica, 2023, 41(7): 1−35(in Chinese). doi: 10.7638/kqdlxxb-2023.0128

Cheng Wendi, Zhang Xiao, Pan Zhaohui, et al. Research on key technologies for efficient storage and access of turbulent big data[J]. Big Data Research, 2024, 10(4): 3−20(in Chinese). doi: 10.11959/j.issn.2096-0271.2024046

Tang Zhigong, Qianweiqi, He Lei, et al. Thoughts and prospects on large model research in aerodynamics[J]. Acta Aerodynamica Sinica, 2024, 42(12): 1−11(in Chinese). doi: 10.7638/kqdlxxb-2024.0056

Liu Jiaqi, Chen Rongqian, Lou Jinhua, et al. Aerodynamic shape optimization of high-speed helicopter rotor airfoil based on deep learning[J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(9): 529828(in Chinese). doi: 10.7527/S1000-6893.2023.29828

Kim M I, Yoon H S, Seo J H. Deep learning models for the evaluation of the aerodynamic and thermal performance of three-dimensional symmetric wavy wings[J]. Symmetry, 2024, 16(1): 21. doi: 10.3390/sym16010021

Du Q W, Liu T Y, Yang L K, et al. Airfoil design and surrogate modeling for performance prediction based on deep learning method[J]. Physics of Fluids, 2022, 34: 015111. doi: 10.1063/5.0075784

Lou Jinhua, Chen Rongqian, Liu Jiaqi, et al. Aircraft aerodynamic performance prediction and inverse design based on a gated diffusion model[J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(10): 135−148(in Chinese). doi: 10.7527/S1000-6893.2024.31183

Sabater C, Stürmer P, Bekemeyer P. Fast predictions of aircraft aerodynamics using deep-learning techniques[J]. AIAA Journal, 2022, 60(9): 5249−5261. doi: 10.2514/1.J061234

Chen H, He L, Qian W Q, et al. Multiple aerodynamic coefficient prediction of airfoils using a convolutional neural network[J]. Symmetry, 2020, 12(4): 544. doi: 10.3390/sym12040544

Chen Bingyan, Liu Chuanzhen, Bai Peng, et al. Aerodynamic prediction for waveriders using deep residual learning[J]. Acta Aerodynamica Sinica, 2019, 37(3): 505−509(in Chinese). doi: 10.7638/kqdlxxb-2019.0027

Chen Hai, Qian Weiqi, He Lei, et al. Aerodynamic coefficient prediction of airfoils based on deep learning[J]. Acta Aerodynamica Sinica, 2018, 36(2): 294−299(in Chinese). doi: 10.7638/kqdlxxb-2017.0098

Lyu Zhaoyan, Nie Xueyuan, Zhao Aobo. Prediction of wing aerodynamic coefficient based on CNN[J]. Journal of Beijing University of Aeronautics and Astronautics, 2023, 49(3): 674−680(in Chinese). doi: 10.13700/j.bh.1001-5965.2021.0276

Li R Z, Zhang Y F, Chen H X. Transfer learning from two-dimensional supercritical airfoils to three-dimensional transonic swept wings[J]. Chinese Journal of Aeronautics, 2023, 36(9): 96−110. doi: 10.1016/j.cja.2023.04.008

Yang Y J, Li R Z, Zhang Y F, et al. Rapid aerodynamic prediction of swept wings via physics-embedded transfer learning[J ] . AIAA Journal, 2024, 63(6): 2545−2559. doi: 10.2514/1.J064576

Dai J H, Liu P Q, Qu Q L, et al. Aerodynamic optimization of high-lift devices using a 2D-to-3D optimization method based on deep reinforcement learning and transfer learning[J]. Aerospace Science and Technology, 2022, 121: 107348. doi: 10.1016/j.ast.2022.107348

Charles R Q, Hao S, Mo K C, et al. PointNet: Deep learning on point sets for 3D classification and segmentation[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA. IEEE, 2017: 77-85. doi: 10.1109/CVPR.2017.16

Morton G M. A computer oriented geodetic data base and a new technique in file sequencing[J]. physics of plasmas, 1966. doi: 10.1063/1.4930281

Jiang J G, Ji Y F, Wang X M, et al. Regressive domain adaptation for unsupervised keypoint detection[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA. IEEE, 2021: 6776-6785. doi: 10.1109/cvpr46437.2021.00671

Fan H Q, Su H, Guibas L. A point set generation network for 3D object reconstruction from a single image[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA. IEEE, 2017: 2463-2471. doi: 10.1109/CVPR.2017.264

Wu Z R, Song S R, Khosla A, et al. 3D ShapeNets: A deep representation for volumetric shapes[C]// 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA. IEEE, 2015: 1912-1920. doi: 10.1109/CVPR.2015.7298801

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

基于迁移学习的点云机翼气动性能预测

基于NURBS曲线的任意圆心角度叶型前尾缘构造

挡风墙距离影响高速列车气动性能LES研究

大型运输类飞机设计中雷诺数效应问题研究进展

基于多精度深度神经网络的汽车气动外形优化设计方法