Modelling and quantification of uncertainty in NACA wing flows

Abstract

In the field of aerospace engineering, the optimization of airfoil design plays a critical role in enhancing aircraft performance. This thesis proposes a novel algorithm for Uncertainty Quantification, which facilitates the solution of Navier - Stokes through an iterative process. UQ is carried out by employing Monte Carlo and Polynomial Chaos methods. This methodology aims to improve computational efficiency while accurately assessing the impact of various parameters, such as Reynolds number and airfoil camber, on aerodynamic characteristics. The study emphasizes the UQ results, revealing how variations in Reynolds number and airfoil camber significantly influence aerodynamic characteristics. The findings highlight critical areas of uncertainty, particularly around the leading and trailing edges of the airfoil and the wake region, where variations in flow conditions introduce significant fluctuations. While the research demonstrates the algorithm's effectiveness, limitations are acknowledged, particularly in handling unsteady flows and complex geometries, as well as the need for improvements in directly computing lift and drag forces. Future work may involve integrating the developed algorithm with machine learning models for design optimization and extending its capabilities to accommodate more intricate flow conditions.