Browse > Article
http://dx.doi.org/10.5407/jksv.2022.20.3.094

Prediction of aerodynamic force coefficients and flow fields of airfoils using CNN and Encoder-Decoder models  

Janghoon, Seo (Department of Naval Architecture and Ocean Engineering, Pusan National University)
Hyun Sik, Yoon (Department of Naval Architecture and Ocean Engineering, Pusan National University)
Min Il, Kim (Department of Naval Architecture and Ocean Engineering, Pusan National University)
Publication Information
Journal of the Korean Society of Visualization / v.20, no.3, 2022 , pp. 94-101 More about this Journal
Abstract
The evaluation of the drag and lift as the aerodynamic performance of airfoils is essential. In addition, the analysis of the velocity and pressure fields is needed to support the physical mechanism of the force coefficients of the airfoil. Thus, the present study aims at establishing two different deep learning models to predict force coefficients and flow fields of the airfoil. One is the convolutional neural network (CNN) model to predict drag and lift coefficients of airfoil. Another is the Encoder-Decoder (ED) model to predict pressure distribution and velocity vector field. The images of airfoil section are applied as the input data of both models. Thus, the computational fluid dynamics (CFD) is adopted to form the dataset to training and test of both CNN models. The models are established by the convergence performance for the various hyperparameters. The prediction capability of the established CNN model and ED model is evaluated for the various NACA sections by comparing the true results obtained by the CFD, resulting in the high accurate prediction. It is noted that the predicted results near the leading edge, where the velocity has sharp gradient, reveal relatively lower accuracies. Therefore, the more and high resolved dataset are required to improve the highly nonlinear flow fields.
Keywords
Convolutional Neural Network; Encoder-Decoder; Deep Learning; Airfoil; Computational Fluid Dynamics;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Swingler, K., 1996, "Applying neural networks: a practical guide," Morgan Kaufmann.
2 Yilmaz, E. and German, B., 2017, "A convolutional neural network approach to training predictors for airfoil performance," In 18th AIAA/ISSMO multidisciplinary analysis and optimization conference, 3660.
3 Zhang, Y., Sung, W. J. and Mavris, D. N., 2018, "Application of convolutional neural network to predict airfoil lift coefficient," In 2018 AIAA/ASCE/AHS/ASC structures, structural dynamics, and materials conference, 1903.
4 Chen, H., Weiqi, Q. and Song, W., 2020, "Multiple Aerodynamic Coefficient Prediction of Airfoils Using a Convolutional Neural Network," Symmetry, Vol.12(4), pp.544.   DOI
5 Duru, C., Alemdar, H. and Baran, O. U. 2021, "CNNFOIL: Convolutional encoder decoder modeling for pressure fields around airfoils," Neural Computing and Applications, Vol.33(12), pp.6835-6849.   DOI
6 Siemens, 2018. STAR CCM+ User Guide.
7 Kurtulus, D.F., 2015, "On the unsteady behaviour of the flow around NACA 0012 airfoil with steady external conditions at Re=1000," Int. J. Micro Air Veh., Vol.7(3), pp.301-326.   DOI