Effective Prediction of Coefficients and Performance of Airfoil Using ANN

The foremost step while designing an aerodyne vehicle is the selection of the proper airfoil as per the flight condition. To determine the correct shape of the vehicle, measuring aerodynamic coefficients is essential. According to the accuracy level, numerous methods exist to calculate these coefficients. The information about different NACA airfoil sections is mainly in demand, although the aircraft engineers using NASA's supercritical airfoil sections as the NACA airfoils are still extensively used to design propellers, aircraft wings, and many other aerodynamic applications. In this paper, computer-based software is used to collect data on these coefficients. An attempt to predict the best set of coefficients is tried by applying Artificial Neural Networks (ANNs) for different sets of Reynolds numbers. The code implements a simple feed-forward neural network for a regression problem using MATLAB.