Multiple Aerodynamic Coefficient Prediction of Airfoils Using a Convolutional Neural Network

• Published in: Symmetry (Basel) Vol. 12; no. 4; p. 544
• Main Authors: Chen, Hai, He, Lei, Qian, Weiqi, Wang, Song
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01-04-2020
• Subjects: Accuracy; Aerodynamic characteristics; Aerodynamic coefficients; Aerodynamics; Aircraft; Aircraft design; airfoil; Artificial neural networks; Computer simulation; Convolution; convolutional neural network (CNN); Data processing; Deep learning; Flight characteristics; Flight safety; Geometry; Machine learning; Mathematical models; Methods; Model testing; multiple aerodynamic coefficients; Neural networks; prediction; Prediction models; Pressure distribution; Reynolds number; symmetric
• ISSN: 2073-8994; 2073-8994
• DOI: 10.3390/sym12040544

Both symmetric and asymmetric airfoils are widely used in aircraft design and manufacture, and they have different aerodynamic characteristics. In order to improve flight performance and ensure flight safety, the aerodynamic coefficients of these airfoils must be obtained. Various methods are used to generate aerodynamic coefficients. The prediction model is a promising method that can effectively reduce cost and time. In this paper, a graphical prediction method for multiple aerodynamic coefficients of airfoils based on a convolutional neural network (CNN) is proposed. First, a transformed airfoil image (TAI) was constructed by using the flow-condition convolution with the airfoil image. Next, TAI was combined with the original airfoil image to form a composite airfoil image (CAI) that is used as the input of the CNN prediction model. Then, the structure and parameters of the prediction model were designed according to CAI features. Finally, a sample set that was generated on the basis of the deformation of symmetrical airfoil NACA 0012 was used to train and test the prediction model. Simulation results showed that the proposed method based on CNN could simultaneously predict the pitch-moment, drag, and lift coefficients, and prediction accuracy was high.