Virtual Development of Composite Materials
This showcase deals with the virtual development of composite materials in a machine learning framework. Such materials are of key importance, for instance, for the automotive market. However, to facilitate accessibility and broad adoption of advanced machine learning tools by companies developing materials, user-friendly, integrated workflows need to be developed. The goal of this showcase is to achieve seamless and user-friendly workflows that help material developers to discover new materials while minimising the number of necessary experiments, thus reducing time-to-market and development costs.
To this end, interoperability of relevant databases and a platform including machine learning tools specifically developed for materials engineering will be investigated. Users of the platform will be able to upload, store and re-use their data from previous projects, and interoperability in the DOME 4.0 framework will facilitate access to other relevant material databases and tools. Data security and anonymisation will need to be ensured. The machine learning tools will allow including uncertainty characterization in the analysis and will also be highly scalable, thus allowing the inclusion of a large number of parameters in the design space exploration for new materials.