Aerodynamic modeling and active sampling based on uncertainty prediction

ZHANG Zijun; LI Huailu; ZHAO Tong; WANG Xu; ZHANG Weiwei

doi:10.7638/kqdlxxb-2024.0045

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Aerodynamic modeling and active sampling based on uncertainty prediction

更新时间：2026-05-28

- Aerodynamic modeling and active sampling based on uncertainty prediction
- Vol. 43, Issue 1, Pages: 12-21(2025)
- 作者机构：
  
  1. 沈阳飞机设计研究所,沈阳,110035
  2. 西北工业大学航空学院,西安,710072
- 作者简介：
- 基金信息：
- DOI：10.7638/kqdlxxb-2024.0045
  CLC： V211.5
- Published：2025
- 稿件说明：
移动端阅览
张子军, 李怀璐, 赵彤, 等. 基于不确定性预测的气动力建模与主动采样[J]. 空气动力学学报, 2025, 43(1): 12−21.

ZHANG Zijun, LI Huailu, ZHAO Tong, et al. Aerodynamic modeling and active sampling based on uncertainty prediction[J]. 2025, 43(1): 12-21.
张子军, 李怀璐, 赵彤, 等. 基于不确定性预测的气动力建模与主动采样[J]. 空气动力学学报, 2025, 43(1): 12−21. DOI： 10.7638/kqdlxxb-2024.0045

ZHANG Zijun, LI Huailu, ZHAO Tong, et al. Aerodynamic modeling and active sampling based on uncertainty prediction[J]. 2025, 43(1): 12-21. DOI： 10.7638/kqdlxxb-2024.0045.

摘要

神经网络方法作为一种高效高精度建模方法，在多个领域得到广泛应用，但其自身的“黑箱”特性结合工程问题特有的小样本现象使得模型可靠性不足，预测结果不确定性大，严重制约了神经网络模型的使用。为提高神经网络模型的工程适用性，以飞机纵向非定常气动特性为研究对象，利用时间卷积神经网络实现了纵向大幅振荡风洞试验的时域非定常气动力建模，并使用MC-Dropout技术对预测结果的不确定性进行评估。在此基础上结合不确定性分析结果，开展了风洞试验样本主动采样。结果表明，模型不确定性可作为预测精度的先验评价指标，模型预测误差与不确定性具有强线性关系，主动采样策略较随机采样策略可以最多降低40%的样本需求，验证了该方法在提升黑箱模型可信度与降低建模样本需求量方面的有效性。

Abstract

Neural network methods

as an efficient and accurate modeling approach

have been widely used in various fields. However

the "black-box" feature of neural networks

combined with the engineering problem of few-shot phenomenon

leads to insufficient model reliability and high uncertainty in the prediction results

severely limiting the use of neural network models. In order to enhance the engineering applicability of neural network models

this study focuses on the unsteady aerodynamic characteristic and utilizes time convolutional networks (TCN) to model the temporal unsteady aerodynamic forces in large-amplitude oscillatory wind tunnel tests. The MC-Dropout technique is employed to evaluate the uncertainty of prediction results. Based on the uncertainty analysis results

active sampling of wind tunnel test samples is conducted. The results indicate that model uncertainty can be used as a prior evaluation of the prediction accuracy. There is a strong linear relationship between the model prediction error and the uncertainty. The active sampling strategy can reduce the required samples by up to 40% compared to the random sampling strategy. This validates the effectiveness of the present method in improving the trustworthiness of black-box models and reducing the number of modeling samples required.

关键词

Keywords

references

ZHANG Z J, ZHAO T, SUN Y, et al . Review of the study on high-angle-of-attack flight problems of aircraft[J ] . Advances in Aeronautical Science and Engineering, 2022 , 13 ( 3 ): 74 − 85 (in Chinese).

WANG H F, ZHAN J X, CHEN K, 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).

WANG Q, QIAN W Q, DING D . A review of unsteady aerodynamic modeling of aircrafts at high angles of attack[J ] . Acta Aeronautica et Astronautica Sinica, 2016 , 37 ( 8 ): 2331 − 2347 (in Chinese).

KOU J Q, ZHANG W W . Data-driven modeling for unsteady aerodynamics and aeroelasticity[J ] . Progress in Aerospace Sciences, 2021 , 125 : 100725 .

TANG Z G, ZHU L Y, XIANG X H, et al . Some research progress and prospect of intelligent aerodynamics[J ] . Acta Aerodynamica Sinica, 2023 , 41 ( 7 ): 1 − 35 (in Chinese).

ZHANG W W, KOU J Q, LIU Y L . Prospect of artificial intelligence empowered fluid mechanics[J ] . Acta Aeronautica et Astronautica Sinica, 2021 , 42 ( 4 ): 524689 (in Chinese).

BAGHERZADEH S A . Nonlinear aircraft system identification using artificial neural networks enhanced by empirical mode decomposition[J ] . Aerospace Science and Technology, 2018 , 75 : 155 − 171 .

LI K, KOU J Q, ZHANG W W . Deep neural network for unsteady a erodynamic and aeroelastic modeling across multiple Mach numbers[J ] . Nonlinear Dynamics, 2019 , 96 ( 3 ): 2157 − 2177 .

CHEN X, ZHAN J X, CHEN K, et al . Unsteady aerodynamic modeling research and virtual flight test verification[J ] . Journal of Experiments in Fluid Mechanics, 2022 , 36 ( 3 ): 65 − 72 (in Chinese).

HE K F, QIAN W Q, WANG Q, et al . Application of data fusion technique in aerodynamics studies[J ] . Acta Aerodynamica Sinica, 2014 , 32 ( 6 ): 777 − 782 (in Chinese).

WANG X, KOU J Q, ZHANG W W, et al . Incorporating physical models for dynamic stall prediction based on machine learning[J ] . AIAA Journal, 2022 , 60 ( 7 ): 4428 − 4439 .

GAWLIKOWSKI J, TASSI C R N, ALI M, et al . A survey of uncertainty in deep neural networks[J ] . Artificial Intelligence Review, 2023 , 56 ( 1 ): 1513 − 1589 .

RAGHU M, BLUMER K, SAYRES R, et al. Direct uncertainty prediction for medical second opinions[C]// 36th International Conference on Machine Learning, PMLR 97: 5281−5290, 2019.

JI W Q. Research on data-driven remaining useful life prediction method based on deep learning and uncertainty quantification[D]. Hefei: University of Science and Technology of China, 2020.

HAN Z H . Kriging surrogate model and its application to design optimization: A review of recent progress[J ] . Acta Aeronautica et Astronautica Sinica, 2016 , 37 ( 11 ): 3197 − 3225 (in Chinese).

DAI Y T, YANG C . Methods and advances in the study of aeroelasticity with uncertainties[J ] . Chinese Journal of Aeronautics, 2014 , 27 ( 3 ): 461 − 474 .

SHIMOYAMA K, KAWAI S . A kriging-based dynamic adaptive sampling method for uncertainty quantification[J ] . Transactions of the japan society for aeronautical and space sciences, 2019 , 62 ( 3 ): 137 − 150 .

PENG X H, KOU J Q, ZHANG W W . Multi-fidelity nonlinear unsteady aerodynamic modeling and uncertainty estimation based on Hierarchical Kriging[J ] . Applied Mathematical Modelling, 2023 , 122 : 1 − 21 .

GAL Y, GHAHRAMANI Z . Dropout as a Bayesian approximation: Representing model uncertainty in deep learning[J ] . 33rd International Conference on Machine Learning, 2016 .

TU L H, JIANG J X, YUAN J P . Prediction of spin characteristic for a certain advanced trainer[J ] . Flight Dynamics, 2011 , 29 ( 5 ): 1 − 4 (in Chinese).

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Development and application of aerodynamic research technologies for fighters at high angle of attack

Dynamic test technique with large amplitude in 2.4 m transonic wind tunnel

Active control of supersonic jet based on multiple microjets

Aerodynamic and wind energy harvesting integrative design for UAVs

A multi-fidelity surrogate model with sparse neural kernel

Related Author

WANG Haifeng

ZHAN Jingxia

CHEN Ke

CHEN Xiang

CHEN Zijun

MA Shang

ZHAO Zhongliang

JIANG Minghua

Related Institution

成都飞机设计研究所

High Speed Aerodynamics Institute of China Aerodynamics Research and Development Center

Department of Aerospace Engineering, Indian Institute of Technology, Kanpur, Kanpur

Zhengzhou Tobacco Research Institute of CNTC, Zhengzhou

Hangzhou International Innovation Institute, Beihang University, Hangzhou

AI问答

⁰