Project goal

With this project I am exploring ways to optimize wing profiles without resorting to lengthy or complex mathematical calculations. Instead, I use evolutionary algorithms and machine learning methods to gradually approach the theoretically optimal profile shape. The aim is to develop an efficient, data-driven approach that improves the aerodynamic performance of a profile with little computational effort.

Procedure

The first step is to research a suitable simulation method that is flexible enough to realistically represent different flow scenarios, while at the same time designed to provide precise and efficiently analyzable data for the optimization process. This simulation forms the basis for generating new profile variants using ML and evolutionary principles and automatically evaluating their performance.