Pressure Distribution Prediction of Supercritical Airfoils at Multiple Flight Conditions Using Deep Learning Approach

Abstract

Supercritical airfoils are commonly found in modern civil airplanes. Effective access to pressure distribution around an airfoil under various flight situations is vital for enhancing the quality of supercritical airfoils. With the rapid development of deep learning, the rise of neural networks has provided new powerful tools for obtaining pressure distribution quickly. This paper proposed a deep learning based model to predict the surface pressure distribution around a supercritical airfoil under multiple flight conditions. The airfoil geometry parameters and various flight conditions are taken as input, and the corresponding surface pressure distributions are taken as output. The statistical results show that the proposed method is accurate and generalized in predicting pressure distributions around supercritical airfoils. Our method, in particular, achieves accurate prediction results over the double shock or strong shock area, demonstrating its superiority in handling complex flows.