TY - JOUR
T1 - Visual Explainable Convolutional Neural Network for Aerodynamic Coefficient Prediction
AU - Zhao, Yanxuan
AU - Zhong, Chengwen
AU - Wang, Fang
AU - Wang, Yueqing
N1 - Publisher Copyright:
© 2022 Yanxuan Zhao et al.
PY - 2022
Y1 - 2022
N2 - 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.
AB - 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.
UR - http://www.scopus.com/inward/record.url?scp=85144050508&partnerID=8YFLogxK
U2 - 10.1155/2022/9873112
DO - 10.1155/2022/9873112
M3 - 文章
AN - SCOPUS:85144050508
SN - 1687-5966
VL - 2022
JO - International Journal of Aerospace Engineering
JF - International Journal of Aerospace Engineering
M1 - 9873112
ER -