Growing the proportion of recycled materials (scrap) in steel production is a key step towards reducing its environmental footprint. To maintain high product quality standards for steels with higher scrap content, it is needed to drastically improve the predictive capability and calculation speed of steel processing models, both at microstructural and at macroscopic level. This challenge is pursued by combining powerful physics-based modelling with machine learning techniques, creating new and efficient hybrid models for process design and control.
These PhD positions are part of a large national research project about "Data Enhanced Physical models to reduce Materials use"