Two types of porcelain tiles with steel slag as the main raw material (steel slag ceramics) were synthesized based on the CaO-Al₂O₃-SiO₂ and CaO-MgO-SiO₂ systems, and their bending strengths up to 53.47 MPa and 99.84 MPa, respectively, were obtained. The presence of anorthite, α-quartz, magnetite, and pyroxene crystals (augite and diopside) in the steel slag ceramics were very different from the composition of traditional ceramics. X-ray diffraction (XRD) and electron probe X-ray microanalysis (EPMA) results illustrated that the addition of steel slag reduced the temperature of extensive liquid generation and further decreased the firing temperature. The considerable contents of glass-modifying oxide liquids with rather low viscosities at high temperature in the steel slag ceramic adobes promoted element diffusion and crystallization. The results of this study demonstrated a new approach for extensive and effective recycling of steel slag.

The endpoint carbon content in the converter is critical for the quality of steel products, and accurately predicting this parameter is an effective way to reduce alloy consumption and improve smelting efficiency. However, most scholars currently focus on modifying methods to enhance model accuracy, while overlooking the extent to which input parameters influence accuracy. To address this issue, in this study, a prediction model for the endpoint carbon content in the converter was developed using factor analysis (FA) and support vector machine (SVM) optimized by improved particle swarm optimization (IPSO). Analysis of the factors influencing the endpoint carbon content during the converter smelting process led to the identification of 21 input parameters. Subsequently, FA was used to reduce the dimensionality of the data and applied to the prediction model. The results demonstrate that the performance of the FA–IPSO–SVM model surpasses several existing methods, such as twin support vector regression and support vector machine. The model achieves hit rates of 89.59%, 96.21%, and 98.74% within error ranges of ±0.01%, ±0.015%, and ±0.02%, respectively. Finally, based on the prediction results obtained by sequentially removing input parameters, the parameters were classified into high influence (5%–7%), medium influence (2%–5%), and low influence (0–2%) categories according to their varying degrees of impact on prediction accuracy. This classification provides a reference for selecting input parameters in future prediction models for endpoint carbon content.

With the high efficiency and automation of converter smelting, it is becoming increasingly important to predict and control the endpoint temperature of the converter. Based on the heat balance, a model for predicting the molten pool temperature in a converter was established. Moreover, the statistical method of multiple linear regression was used to calculate the converter heat loss coefficient, greatly improving the prediction accuracy of the mechanistic model. Using the model, the oxidation process for each element in the molten pool, the melting processes of scrap, and the flux were also calculated. The model could better approximate the actual smelting process. Data from a 130 t converter were collected to validate the model. When the error ranges were limited to ±20 and ±15 °C, the model hit rates were 96 and 86.7%, respectively.