Project description
New toolpath eliminates trial and errors
The 3D printing of aluminium alloys has advanced in recent years. However, it remains a complex process that leads to many defects and a high scrap rate. It’s a cost barrier to scale the technology for use in the automotive and aerospace sectors. In this context, the EU-funded AMA - AM meets AI project will develop a toolpath based on machine learning to achieve a defect-free and first-time-right component. By reducing the cost barrier and using certain cost, quality and environmental impact standards, AMA - AM meets AI will enable competitive sustainable solutions.
Objective
Additive manufacturing holds important promises for sustainable engineering products and applications. Combining the capabilities of 3D printing in terms of bionic lightweight design with the lightweight nature of aluminium and its outstanding properties could lead to a revolution in the field of sustainable mobility. Despite the recent advances, 3D printing of aluminium alloys is a complex process that leads to many defects and a high scrap rate. This becomes a cost barrier to scale the technology especially in the sectors automotive and aerospace. AMA proposes to develop a machine learning based toolpath with process parameters that are optimal to achieve a component defect free and first-time right. In the context of the project, we will provide science-based life cycle analysis of the indirect impact of our technology in enabling competitive sustainable solutions to scale by reducing the cost barrier and also compare the parts before and after using our AI technology in cost, quality and environmental impact criteria. The technical objective will be achieved by designing a testbed demonstrator. For this, a specific Aluminium alloy (ALF357) and a representative geometry from the automotive industry will be selected. The AI process simulation algorithm will be trained with increasing geometry complexity. Once demonstrated the accuracy, the model will be used to achieve first-time right printing and avoid the traditional costly trial and errors. The success of the use case will provide 1000Kelvin with a unique demonstrator, showing the capability of our AI based APIs to solve some of the complex process challenges of the AM industry. This will help position our business as a leader technology provider and enhance our growth perspective to promote the role of women in leading tech start-ups and open the way for a more inclusive 3D printing ecosystem.
Fields of science
- engineering and technologymechanical engineeringvehicle engineeringautomotive engineering
- natural scienceschemical sciencesinorganic chemistrypost-transition metals
- natural sciencesmathematicspure mathematicsgeometry
- engineering and technologymechanical engineeringmanufacturing engineeringadditive manufacturing
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
Programme(s)
- HORIZON.3.2 - European innovation ecosystems Main Programme
Funding Scheme
HORIZON-AG-LS - HORIZON Lump Sum GrantCoordinator
10625 BERLIN
Germany
The organization defined itself as SME (small and medium-sized enterprise) at the time the Grant Agreement was signed.