Smart Ladle: AI-Based Tool for Optimizing Caster Temperature

Abstract

From the BOF/EAF to the caster, the ability to quantify and respond to the variables that affect steel casting temperature is crucial for achieving consistent casting quality and maximizing productivity. Deviations from the optimum steel casting temperature can require adjustment to casting speed, which impacts productivity and can also harm product quality. This work will use a deep-learning network to develop quantifiable relationships between the casting temperature and various factors during the ladle refining process to enable predictions of casting temperature and precise adjustments to steel temperature prior to the ladle reaching the casting stage of the production process.