Multi-Fidelity Aerodynamic Data Set Generation for Early Aircraft Design Phases

Abstract

A widespread industrial adoption of Computational Fluid Dynamics in early design phases would significantly increase the modelling fidelity compared to classical conceptual design tools. However, relatively high computational cost to obtain comprehensive data sets and a significant manual effort needed for the creation of suitable geometries for meshing at early design stages are limiting factors today. In this light, this paper presents a multi-fidelity workflow for aerodynamic data set generation which addresses the two aforementioned bottlenecks. First, the automated generation of geometries coupled with meshing capabilities. Second, the capability to efficiently generate high-fidelity-based aerodynamic data sets. All tools of the workflow use a common parametric data exchange format for the integration into a conceptual design framework and share the same master geometry which is derived to create a tool specific geometrical representation. Computational Fluid Dynamics tools of varying fidelity provide the input for the framework for multi-fidelity surrogate modelling to efficiently enable comprehensive aerodynamic data set generation. Throughout the paper a generic triple-delta combat aircraft is considered as use case.