The Control and Prediction of End-Point Phosphorus Content during BOF Steelmaking Process

Removal of phosphorus is a reaction, which plays an important role in combined converter steelmaking process, and the precise control of end‐point phosphorus content during BOF steelmaking process would greatly improve the quality of liquid steel. Therefore, the relation between dephosphorization ratio and temperature of liquid steel, FeO content of slag, slag basicity is clearly clarified through thermodynamic analysis of dephosphorization process in this paper. Besides, by means of combining the methods of multivariate regression analysis and multi‐level recursive completely, the multi‐level recursive regression model, which is used to complete the prediction of end‐point phosphorus content during BOF steelmaking process, is established based on large amount of production data. The verification of the model with the data taken from three steel plants indicates that the hit rate of the multi‐level recursive regression model is above 84% when predictive errors of the model are within ±0.005%, and it could provide a relatively good reference for real production. This paper presents the research on thermodynamic analysis of dephosphorization process and establishment of the multi‐level recursive regression model for the prediction of end‐point phosphorus content during BOF steelmaking process. The verification with the data taken from three steel plants indicates that the hit rate of the model is above 84% when predictive errors of the model are within ±0.005%.