Research Progress and Intelligent Trend of Slag Foaming Prediction

• • Slag foaming prediction is crucial for EAF efficiency and stability; the review identifies that single models fail to balance real-time capability and accuracy under multi-variable coupling, necessitating hybrid frameworks. • • Five prediction methods (empirical formulas, dimensionless modeling, thermodynamic calculations, CFD, machine learning) are systematically compared; thermodynamic tools like FactSage are used for phase equilibria, while CFD handles gas-liquid two-phase flow, but each has limitations in dynamic industrial conditions. • • The review proposes a hybrid prediction framework integrating mechanism-based and data-driven models, emphasizing physical constraints, multi-scale coupling, and multi-source data fusion to achieve 'controllable and adjustable' foaming. • • The study highlights the importance of green steelmaking, with slag foaming prediction directly impacting energy efficiency and process stability, aligning with national research funding (National Natural Science Foundation of China, No. 52074001).