Fast Predictions of Aircraft Aerodynamics using Deep Learning Techniques

Abstract

The numerical analysis of aerodynamic components based on the Reynolds Average Navier Stokes equation has become critical for the design of transport aircraft but still entails large computational cost. Simulating a multitude of different flow conditions with high-fidelity methods as required for loads analysis or aerodynamic shape optimization is still prohibitive. Within the past few years the application of machine learning methods has being proposed as a potential way to overcome these shortcomings. This is leading towards a new data-driven paradigm for the modelling of physical problems. The objective of this paper is the development of a deep learning methodology for the prediction of aircraft surface pressure distributions and the rigorous comparison with existing state-of-the-art non-intrusive reduced order models. Three data driven-methods, Gaussian Processes, proper orthogonal decomposition combined with an interpolation technique and deep learning are proposed. Results are compared for a 2D airfoil case and the NASA Common Research Model transport aircraft as a relevant 3D case. Results show that all methods are able to properly predict the surface pressure distribution at subsonic conditions. In transonic flow, when shock waves and separation lead to non-linearities, deep learning methods outperform the others by also capturing the shock wave location and strength accurately