Global Optimization Framework for Airfoil Utilizing Geometrical Flexibility with UTCart (2D) and Metaheuristics

Abstract

In this presentation, we introduce a robust and efficient airfoil design method that combines a two-dimensional version of UTCart based on the the Cartesian mesh and global optimization methods based on a heuristic. Since a wide variety of shapes can be generated during the optimization process, the Cartesian mesh method was expected better flow solver for global design optimizations than the body-fitted mesh methods because it uses a hierarchical orthogonal mesh method and the boundaries are handled by embedded boundaries. It was also expected to be easier method to adapt to zero-thickness geometries. In this study, we developed a framework for design optimization flamework using the Cartesian mesh based flow solver and the global optimization based on the metaheuristics such as evolutionary algorithms and Bayesian optimization.