From aerospace heat exchangers to medical implants and vibration-isolation devices for spacecraft, Metal Additive Manufacturing opens up new possibilities for designing and producing advanced, customised and more efficient components.
In this context, CATEC is leading the METALIA project (Enabling Technologies for the Implementation of Artificial Intelligence in the Value Chain of Additive Manufacturing of New Metal Alloys), which aims to design, optimise and manufacture various metal devices using additive manufacturing (3D printing), incorporating innovative Artificial Intelligence (AI) tools to enable more automated customisation and to predict their behaviour under different operating conditions, integrating them throughout the additive manufacturing value chain.
Metal Additive Manufacturing (AM) technologies offer significant competitive advantages for the development of specialised or customised products, notably the ability to produce complex geometries, reduce material usage, make components lighter, and move towards more sustainable production processes by reducing the need for specific tools, moulds and jigs.
Therefore, the METALIA project aims to address some of the key challenges associated with the additive manufacturing of metal components, such as design optimisation, control of the production process for each alloy, improvement of mechanical properties, and prediction of the final performance of the parts.
The initiative is coordinated by our technology centre as leader of the AEI consortium, which also includes the research centres LEITAT, IMDEA Materiales and the Polytechnic University of Madrid, alongside the CDTI consortium of companies comprising EGILE, Bitmetrics, Grupo Sevilla Control, Cubicoff and DLYTE. This collaboration is fostering an environment conducive to generating scientific results that reach advanced levels of technological maturity, thanks to effective transfer between scientific research and technological development.
The project forms part of the ‘Science and Innovation Missions’ programme, funded by the TransMisiones 2023 initiative of the Centre for Technological Development and Innovation (CDTI), and receives funding from the Ministry of Science, Innovation and Universities, the State Research Agency and the CDTI.
Main areas of work and progress
With METALIA, the combination of FAM and AI will enable progress towards a new generation of optimised metal components that are lighter, customised and offer improved performance for strategic sectors such as aerospace, biomedicine, energy and industry.
Since its launch two years ago, the project consortium has developed AI tools to analyse the properties – including defectology, surface roughness and porosity – of novel alloys with enhanced functionality, designed specifically for 3D printing, for each of the components to be developed within the project. Notable among the selected alloys are the copper alloys GRCop-42 and CuCrZr, the aluminium alloy, the A6061-RAM2 aluminium alloy, and Nitinol.
Another area of research that has been explored in depth is the prediction of mechanical properties using AI. This approach is based on pre-selected lattice designs and textured surfaces that represent a wide range of mechanical properties, to which systematic design variations have been applied. Furthermore, to provide the industry with tools capable of generating complex designs and geometries that help to realise the potential of metal additive manufacturing, work is underway on the development of cellular automaton (CA)-based models for generative design. Thus, 3D cellular automaton models have been developed to generate both metamaterials and metasurfaces using CA, integrating principles of computational modelling, biomimicry and additive manufacturing.
Furthermore, to improve manufacturing, an analysis has been carried out on the applicability of AI tools to the monitoring of additive manufacturing processes, both in powder bed fusion (PBF) and direct energy deposition (DED). In this case, initial results have demonstrated the viability of using segmentation models for the automated monitoring of additive manufacturing processes, establishing a solid foundation for the research and development of a monitoring model aimed at both the automatic detection of fusion defects and the early identification of geometric deviations.
Furthermore, the post-FA processes – primarily machining, electropolishing and inspection – are being optimised, and data acquisition systems have been integrated during the machining of components; these will enable analysis of machining quality based on the datasets generated. Thanks to an in-depth study of the parameters influencing dry electropolishing (electrolyte type, mobility, times, particle size), it is possible to predict the surface quality of parts before and after electropolishing using AI tools. Similarly, within the METALIA project, X-ray tomography inspection is being optimised using AI, which will reduce the time required for each tomography scan and the number of scans needed to determine the part’s properties in detail.
Finally, CATEC has begun researching methodologies, algorithms and machine learning architectures for computer vision based on deep learning, enabling the automatic quantification and analysis of the phases and microstructure of alloys. The resulting tool would not only allow for a more accurate characterisation of the morphology and distribution of microstructural features, but would also enable a more reliable prediction of the mechanical properties of 3D-printed materials.
