Optimized production in additive manufacturing (DED process)

Objectives
New technology and digital innovations from (non-space) start-ups and SMEs shall be brought to European space business and projects, increasing efficiency and optimising existing processes at MT Aerospace in cooperation with FLPP and AZO.

Additive Manufacturing based on Directed Energy Deposition (DED) is a technology that has tremendous potential for optimisation of product design, supply chain management, flexibility and cost savings of large-scale aerospace parts. However, process development, quality optimisation and setting up a stable series production requires big efforts in both cost and time.

Solution approach
Systematic data acquisition and processing of all data that can be determined in additive manufacturing (e.g. process parameters, machine data, monitoring data, up-stream simulations) will result in a complex database. Within that database, all kind of information dependent on the various influencing factors of the process will be available.

The classical approach of causal analyses for process optimisation, development of new materials and stabilisation of series production will be both cost and time consuming.

To overcome these challenges, pattern recognition can replace causal analysis for improved processes. Furthermore, to develop that advanced and efficient way of data handling, various sub-areas of AI research (especially machine and deep learning) shall be used.

Benefits
The proposed pilot case will enable new players for space activities and actively contribute to the growing Space Transportation Ecosystem in Bavaria. With this pilot case, direct implementation of digital innovations (spin-in) from start-ups and SMEs for European space projects can be demonstrated.

Use of various sub-areas of AI research (especially machine and deep learning) for the efficient use of all kinds of process related data will provide the following benefits:

• Increase of process stability in additive manufacturing
• Reduction of downstream testing costs for additively manufactured components
• Shortening of the qualification process of new machines, materials and components in additive manufacturing
• Reduction of downtimes and maintenance costs in factory operation through predictive maintenance