KiKomp | AI-based tool for certification-compliant part and component design taking into account multiple virtual verification tests and individual user requirements

The aim of the KiKomp research project is to develop an AI-based software tool that significantly accelerates the design of aircraft components. Instead of time-consuming iterative finite element simulations, a trained AI should predict mechanical properties of components in real time and thereby enable rapid, certification-compliant design decisions. Aircraft floor panels in sandwich construction serve as a demonstrator.

As consortium leader, SCALE develops the AI-based analysis method as well as the software-supported design tool. This includes the development of the data structure, the integration of optimization methods, the prototypical implementation of a web-based user interface, and the integration of all components into an interactive software solution. The basis is formed by the virtual validation tests developed and validated by TU Dresden.

The project combines virtual validation tests, artificial intelligence, and optimization methods into a novel development approach. As a result, development times can be significantly reduced, physical testing efforts decreased, and robust component designs that account for manufacturing tolerances and certification requirements can be enabled. In the long term, the software solution is intended to be used in the aviation industry after the project concludes.

The project is funded under the grant number (FKZ) 20Q2422 by the Federal Ministry for Economic Affairs and Energy (BMWE) as part of the aviation research program LuFo VII-1.