Russian Steelmakers Digitize Furnace Heating With AI Control
EVRAZ’s digital transformation project replaces manual furnace control with intelligent algorithms for heating steel billets.
At the Nizhny Tagil Iron and Steel Works, EVRAZ is rolling out an intelligent metal heating system that digitizes how billets are heated in furnaces. The digital control solution for billet heating in the wide-flange rolling mill was developed and launched in partnership with CyberPhysics.
Engineers created and validated a hybrid mathematical model, a digital twin of the heating process. The model combines fundamental heat transfer physics with neural network algorithms trained on tens of thousands of historical records of production parameters.
Analysis, Calculation, Integration
The digital twin analyzes a wide range of parameters in real time, including steel grade, billet cross-section, rolling mill downtime schedules, and current and planned production targets. Based on this data, it calculates optimal temperature settings for each furnace zone at every stage of heating. The solution has been integrated into day-to-day shop operations: furnace operators receive recommendations on temperature settings, billet spacing, and projected output timing.
The system not only reduces fuel consumption, but also limits overheating of the metal, which has a positive effect on product quality and equipment lifespan. Going forward, the solution could be extended to other production processes, including the ball-rolling mill.
