BMFTR - DiReSiC
The BMFTR-funded research project DiReSiC aims at digitalising the manufacturing process of advanced ceramics such as silicon-infiltrated silicon carbide (SiSiC) and diamond-reinforced silicon carbide (DiaSiC) using recycled materials. It achieves this through the development of a data ontology for the material manufacturing lifecycle and subsequently with machine learning models which will enable the reproducibility of high-quality material properties.
The data ontology will be built upon the PMD Core Ontology, resulting in a domain-specific ontology that captures and links all stages of the material life cycle. The ontology will ensure consistent structuring, uniform data integration, and interoperability of heterogeneous and experimental material data from various industrial and academic sources within a centralized database server. This serves as the basis for the training of machine learning models, which will aim to identify holistic relationships and correlations along all steps of the digitalised material life cycle. Data-aggregating multi-task deep neural networks show promising potential for learning the process-microstructure-property correlations and possibly further far-reaching correlations in the material life cycle.
Project Team:
Prof. Dr. Christopher Künneth
Principle Investigator
Rayan Hamid Mohiuddin (M.Sc.)