Mathematical modelling of oxygen steelmaking

Oxygen steelmaking is complex because of the presence of multiple phases, many components, and the non-steady state/non-homogenous conditions within the process. The severe operating conditions make it difficult to take measurements and directly observe the process. Although some mathematical models have been developed to describe the kinetics of oxygen steelmaking, improve understanding of the system and optimize process control, existing theories have not been successfully applied under dynamic process conditions. For example, there is evidence that the period when the bloated droplets are suspended in the emulsion phase enhances the reaction areas, and the decarburization rates. In this study a computer based model has been developed that incorporates bloated droplet theory under dynamic conditions to evaluate its influence on the overall kinetics of the process. The process variables influencing the decarburization reaction kinetics considered in the model were hot metal, scrap and flux charges, hot metal, scrap and slag compositions, oxygen blowing conditions, temperature of the bath, flux dissolution, scrap melting, the droplet generation rate, droplet size, residence time of droplets in the emulsion, and decarburization rates in the emulsion and impact zones. These process variables were modelled individually. The equations involved in this model were solved numerically on the basis of parameters encountered in the operation of oxygen steelmaking furnaces. All the developed models were translated into computational code and linked to one another in this study. The global model was tested with actual data for a 200 t top-blown furnace under the full scale operating conditions. The model predictions proved that the carbon content of liquid iron agreed with the actual process data. The model suggests that 45% of total carbon was removed via emulsified metal droplets and the remainder was removed from the impact zone during the entire blow. It was found that the residence time of droplets and the decarburization reaction rate via emulsified droplets, was a strong function of the bloating behavior of the droplets. The estimated residence times of the metal droplets in the emulsion were between 0.4 and 45 seconds throughout the blow. The global model enabled a comparison of the decarburization rates in different reaction zones and provided a better understanding of the process variables affecting in each reaction zone. On the basis of this model, the decarburization rates in the emulsion phase reached 60% of the overall decarburization rate during the main blow. These findings will provide a further theoretical understanding of the oxygen steelmaking process and provide a predictive tool for industrial applications. This development represents an original contribution to our understanding of steelmaking.