Novel, yet similar: A similarity-assisted product family design approach for structural aero-engine components

Licentiate thesis, 2023

The aviation industry is in a state of transformation. The climate crisis has amplified the need to innovate, and consequently manufacturers in the aviation industry need to investigate new and more sustainable design concepts. This is challenging, because there is no obvious replacement for kerosene-fueled aero-engines, though there are multiple technologies in development that may potentially take its place. Examples of such technologies include electric or hybrid-electric propulsion, or combustion engines fueled by hydrogen or synthetic sustainable aviation fuels. This increases the challenge for manufacturers, who must deal with high technological uncertainty. At the same time, manufacturers need to assert that the cost of realization is feasible for new aero-engine component designs, while also fulfilling the requirements for safety and performance. There is therefore a need for methods and tools that will assist designers in making fast and efficient design evaluations, to enable the exploration of large design spaces at reduced costs and lead-times.

To make design space exploration more efficient, a similarity-assisted design space exploration method is proposed. This method provides increased trustworthiness in design space exploration results, while also highlighting opportunities for reuse of knowledge and other assets from legacy designs. Additionally, a software tool for automatically generating aero-engine structural components has been developed. This software enriches all generated geometries with information used to facilitate automated manufacturability analysis, as well as evaluation of structural performance. By utilizing the automated geometry generation tool in conjunction with the proposed design space exploration method, designers can quickly and efficiently evaluate the manufacturability and structural performance of novel concepts.