Automatic Aerostructural Analysis and Aerodynamic Optimization for Collaborative Aircraft Design Process

Analysis of scenario

Overall aircraft design is multidisciplinary by nature, it means that results coming from multidisciplinary analysis can greatly improve the final design and reduce the time cost of the design cycle. In addition, a significant improvement can be reached if high-fidelity methods are included in this kind of analysis, especially when unconventional configuration need to be considered. However, several limitations do not allow the possibility to get the benefits of high-fidelity applications during early design phases. The main thesis motivation is to illustrate how it is possible to obtain detailed design analysis in reasonable time and with appropriate accuracy, taking into consideration the coupling between aerodynamic and structural properties. The thesis is the result of a collaboration between the DAF group of the University of Naples “Federico II” and the research group of DLR institute of Hamburg in which the candidate has completed an internship within the AGILE 4.0 project.

Statement of the problem

The work concerns computational fluid dynamic and structural coupled analysis performed through automated Python-based workflows. The tools employed are open-source or provided by partners. Therefore, an appropriate connection and data transfer among modules must be provided, complying with each tool or partner requirements. The implemented workflows provide aerodynamic analysis and optimization, aero-structural analysis at different flow conditions and centralized data format geometry update.

Adopted methodology

The automated aero-structural process developed is based on the DLR centralized data format CPACS. It defines the aircraft geometry and configuration and allows information sharing during the process. Pointwise and SU2 are employed respectively for mesh generation and computational aerodynamic analysis. LA-GRANGE tool provides linear structural analyses. All modules are Python-based and connected to each other through RCE. SLSQP optimizer, adjoint, moving last square and FFD methods are employed.

Main results

Flexible and robust automatic workflows are implemented in a distributed design environment (RCE), provided by DLR. The input and output of the process are provided through CPACS format file. The workflows allow gradient-based shape optimization of a generic aircraft component and aero-structural analysis performed at different flow conditions. The output shape geometry is supplied through CPACS format file. An application on a UAV configuration is performed:mesh sensitivity analysis, aerodynamic constrained wing shape optimization in transonic flow condition, aero-elastic shape deformation in cruise condition and flexible polar analysis are carried out to demonstrate the workflow capabilities.