The performance and quality of steel products are significantly impacted by the alloying element control. The efficiency of alloy utilization in the steelmaking process was directly related to element yield. This study analyses the factors that influence the yield of elements in the steelmaking process using correlation analysis. A yield prediction model was developed using a t-distributed stochastic neighbor embedding (t-SNE) algorithm, a whale optimization algorithm (WOA), and a long short-term memory (LSTM) neural network. The t-SNE algorithm was used to reduce the dimensionality of the original data, while the WOA optimization algorithm was employed to optimize the hyperparameters of the LSTM neural network. The t-SNE-WOA-LSTM model accurately predicted the yield of Mn and Si elements with hit rates of 71.67%, 96.67%, and 99.17% and 57.50%, 89.17%, and 97.50%, respectively, falling within the error range of ±1%, ±2%, and ±3% for Mn and ±1%, ±3%, and ±5% for Si. The results demonstrate that the t-SNE-WOA-LSTM model outperforms the backpropagation (BP), LSTM, and WOA-LSTM models in terms of prediction accuracy. The model was applied to actual production in a Chinese plant. The actual performance of the industrial application is within a ±3% error range, with an accuracy of 100%. Furthermore, the elemental yield predicted by the model and then added the ferroalloys resulted in a reduction in the elemental content of the product by 0.017%. The model enables accurate prediction of alloying element yields and was effectively applied in industrial production.

Carbon deoxidation is a clean steelmaking route, but rapid CO generation can increase gas holdup, raise the melt level, and cause splashing. However, a quantitative link between internal deoxidation kinetics and melt-level rise has not yet been established. Hot-state experiments at 1873 K in a 2 kg crucible under argon with 0.20–1.00 wt.% C were coupled with a dual-pathway kinetic model developed by decoupling free-surface mass transfer from internal CO generation through heterogeneous nucleation at active refractory pores and incorporating an effective reaction depth constrained by hydrostatic pressure. The model reproduced measured oxygen evolution and revealed three deoxidation stages. Internal gas generation initially dominated, contributing 96.6% of the overall rate. As carbon–oxygen supersaturation fell below a critical threshold, the nucleation zone receded upward and free-surface mass transfer became dominant. Within the investigated conditions, the internal apparent deoxidation rate constant during the rapid stage showed an empirical linear relationship with the measured maximum melt-level rise. The critical splash-prevention rate was 4.25 min⁻¹ for the crucible and was estimated to be 2.43 min⁻¹ for a selected 85 t ladle by combining this relationship with the average gas holdup relation and ladle geometry. The present results provide a kinetic basis for assessing splash risk during atmospheric carbon deoxidation.