Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

MM: Fachverband Metall- und Materialphysik

MM 56: Topical Session (Symposium MM): Big Data in Materials Science - Managing and exploiting the raw material of the 21st century

MM 56.1: Topical Talk

Donnerstag, 15. März 2018, 11:45–12:15, H 0107

On the need for a digital representation of materials data along scientific and industrial processes — •Christoph Schweizer, Eva Augenstein, Heiner Oesterlin, and Adham Hashibon — Fraunhofer Institute for Mechanics of Materials IWM, Freiburg, Germany

In the field mechanics of materials and material science and engineering in general, an ongoing challenge is to identify process-structure-property relationships. Due to the vast amount of manufacturing processes, mechanical properties as well as characterization and simulation methods on all scales, the resulting research data is extremely heterogeneous in nature. In this work, first steps are undertaken to set up an ontological knowledge base, which can be used to manage and analyze a large amount of material histories and provides the flexibility, which is needed to work on a scientific level. The ontology provides the structure and logical relations, which are needed to set up and later analyze the knowledge base. A small domain ontology is created based on the upper level Basic Formal Ontology (BFO) and the European Materials Modelling Ontology (EMMO), which are currently under development. The knowledge base which can be represented as a network graph is created and filled with research data from available public funded projects dealing with metallic high temperature materials, the underlying manufacturing processes, characterization methods, the microstructure and the mechanical properties. It is demonstrated, that the resulting network graph can be analyzed to identify process-structure-property relationships.