A Surrogate Model Based Shape Optimization Framework for Compressible Flows

In this study, a shape optimization framework is designed and developed in order to use surrogate model-based shape optimization methods in the context of compressible flows. While designing the shape optimization framework, various open-source libraries were implemented and made to communicate with each other through Python language. In addition to using Kriging method for surrogate model-based shape optimizations, Gradient Enhanced Kriging (GEK) and Gradient Enhanced Kriging with Partial Least Square (GEKPLS) methods are also investigated. They are investigated from different perspectives and tested on a two-dimensional airfoil as a benchmark case. In addition to optimizations with single-objective functions, as currently modeled using the framework, the capabilities of the framework will be extended to multi-objective and multipoint studies.