Aerodynamic Shape Design by Evolutionary Optimization and Support Vector Machines

Abstract

This paper proposes a computational methodology for the aerodynamic shape design of aeronautical configurations, aiming a broad and efficient exploration of the design space. A novel adaptive sampling technique focused on the global optimization problem, the Intelligent Estimation Search with Sequential Learning (IES-SL), is presented. This approach is based on the use of Support Vector Machines (SVMs) as the surrogate model for estimating the objective function, in combination with an evolutionary algorithm (EA) to enable the discovery of global optima. The proposed methodology is applied to improve the aerodynamic performance of a two-dimensional airfoil and a three-dimensional wing and results on surrogate model validation and optimization-focused sampling criteria are discussed.