Visual Explainable Convolutional Neural Network for Aerodynamic Coefficient Prediction

Recently, aerodynamic performance analysis has been widely studied due to its importance in aircraft design. Most works adopted computational fluid dynamics (CFD) simulation to compute the aerodynamic forces, which is time consuming. To reduce the simulation time, several works proposed to use deep...

Full description

Saved in:
Bibliographic Details
Main Authors: Yanxuan Zhao, Chengwen Zhong, Fang Wang, Yueqing Wang
Format: Article
Language:English
Published: Wiley 2022-01-01
Series:International Journal of Aerospace Engineering
Online Access:http://dx.doi.org/10.1155/2022/9873112
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832561611317968896
author Yanxuan Zhao
Chengwen Zhong
Fang Wang
Yueqing Wang
author_facet Yanxuan Zhao
Chengwen Zhong
Fang Wang
Yueqing Wang
author_sort Yanxuan Zhao
collection DOAJ
description Recently, aerodynamic performance analysis has been widely studied due to its importance in aircraft design. Most works adopted computational fluid dynamics (CFD) simulation to compute the aerodynamic forces, which is time consuming. To reduce the simulation time, several works proposed to use deep learning model as the surrogate model of CFD simulation. However, the explainability of deep learning models is poor and has been widely criticized, which limits the further development of deep learning in aerodynamic performance analysis. In this paper, a novel neural network is proposed to predict the aerodynamic forces of airfoils. To improve the explainability, the circular padding is proposed to replace traditional zero padding in the convolutional layers. Moreover, the saliency map of the predicted aerodynamic force on the input airfoil is shown in a more intuitive way. In this manner, the influence of different parts of airfoil on the final aerodynamic force can be easily analyzed. Extensive experiments on different data sets show that our work is efficient and effective. Most importantly, these results explain the potential relationship between the airfoil and the aerodynamic force.
format Article
id doaj-art-07140f73e7ac48899404105536f48a5c
institution Kabale University
issn 1687-5974
language English
publishDate 2022-01-01
publisher Wiley
record_format Article
series International Journal of Aerospace Engineering
spelling doaj-art-07140f73e7ac48899404105536f48a5c2025-02-03T01:24:38ZengWileyInternational Journal of Aerospace Engineering1687-59742022-01-01202210.1155/2022/9873112Visual Explainable Convolutional Neural Network for Aerodynamic Coefficient PredictionYanxuan Zhao0Chengwen Zhong1Fang Wang2Yueqing Wang3School of AeronauticsSchool of AeronauticsChina Aerodynamics Research and Development CenterChina Aerodynamics Research and Development CenterRecently, aerodynamic performance analysis has been widely studied due to its importance in aircraft design. Most works adopted computational fluid dynamics (CFD) simulation to compute the aerodynamic forces, which is time consuming. To reduce the simulation time, several works proposed to use deep learning model as the surrogate model of CFD simulation. However, the explainability of deep learning models is poor and has been widely criticized, which limits the further development of deep learning in aerodynamic performance analysis. In this paper, a novel neural network is proposed to predict the aerodynamic forces of airfoils. To improve the explainability, the circular padding is proposed to replace traditional zero padding in the convolutional layers. Moreover, the saliency map of the predicted aerodynamic force on the input airfoil is shown in a more intuitive way. In this manner, the influence of different parts of airfoil on the final aerodynamic force can be easily analyzed. Extensive experiments on different data sets show that our work is efficient and effective. Most importantly, these results explain the potential relationship between the airfoil and the aerodynamic force.http://dx.doi.org/10.1155/2022/9873112
spellingShingle Yanxuan Zhao
Chengwen Zhong
Fang Wang
Yueqing Wang
Visual Explainable Convolutional Neural Network for Aerodynamic Coefficient Prediction
International Journal of Aerospace Engineering
title Visual Explainable Convolutional Neural Network for Aerodynamic Coefficient Prediction
title_full Visual Explainable Convolutional Neural Network for Aerodynamic Coefficient Prediction
title_fullStr Visual Explainable Convolutional Neural Network for Aerodynamic Coefficient Prediction
title_full_unstemmed Visual Explainable Convolutional Neural Network for Aerodynamic Coefficient Prediction
title_short Visual Explainable Convolutional Neural Network for Aerodynamic Coefficient Prediction
title_sort visual explainable convolutional neural network for aerodynamic coefficient prediction
url http://dx.doi.org/10.1155/2022/9873112
work_keys_str_mv AT yanxuanzhao visualexplainableconvolutionalneuralnetworkforaerodynamiccoefficientprediction
AT chengwenzhong visualexplainableconvolutionalneuralnetworkforaerodynamiccoefficientprediction
AT fangwang visualexplainableconvolutionalneuralnetworkforaerodynamiccoefficientprediction
AT yueqingwang visualexplainableconvolutionalneuralnetworkforaerodynamiccoefficientprediction