SSAB, in close collaboration with SWERIM, is making strong progress on the furnace modeling activities within the TWINGHY project. A key outcome is the development of a thermal FEM‑based furnace model capable of delivering fast, three‑dimensional, time‑resolved predictions of slab temperatures during reheating.

At the SSAB site in Oxelösund, large steel slabs are reheated prior to hot rolling. Precise control of slab temperature is essential to ensure uniform product quality and stable downstream processing. Even small improvements in temperature prediction accuracy can be directly leveraged to improve material quality while reducing overall energy consumption. As the choice of fuel strongly affects heat transfer mechanisms and oxide scale formation, it is crucial that furnace models can adapt to different combustion atmospheres and fuel compositions.

The furnace and slab model has been calibrated and validated using full‑scale experimental data from an industrial reheating furnace in Oxelösund, Sweden. The calibration was first carried out using a controlled heating experiment that measured several points in a slab throughout the heating process. The model was then verified against an independent measurement campaign, showing excellent agreement between predicted and measured temperatures throughout the slab thickness.

Following validation, the model was applied using furnace operating data alone to reconstruct the time‑resolved temperature history of slabs heated during a separate production period. This post‑analysis demonstrated high accuracy and has already contributed to improved process understanding and enhanced control strategies at the site.

In addition, the detailed, time‑resolved temperature profiles generated by the model have enabled more accurate estimation of high‑temperature diffusion processes in steel, such as impurity redistribution during reheating. This further strengthens the value of the model not only as a predictive tool, but also as a foundation for advanced materials and process analysis within the TWINGHY digital twin framework.