Parameterized Geometry Formulation for Inverse Design and Optimization

Abstract

This contribution focuses on the importance of preprocessing tools for successful design and optimization in practice of turbomachinery engineering. The development of problem-oriented computational geometry generation software is illustrated for the example of aerodynamic inverse design of transonic flow elements which define the compatible boundary conditions (surfaces) in detail. Resulting from learned sensitivity of high speed flows to small changes in airplane wing or turbomachinery blade geometry, preprocessing software is provided to create parametric shapes to be varied for optimization cycles or numerical simulation of mechanical adaptation processes. Supporting the need to design from a multidisciplinary viewpoint, parameterized geometry components for aerodynamic, as well as for thermal and structural considerations are defined. Examples for turbomachinery blade design and optimization are given.