Modelagem do processo de desfosforação de ferro-gusa líquido

Abstract

Due to the exploitation of iron ore mines the quality of this raw material is considerably reduced, resulting in a significant increase in phosphorus content in iron ore over the years. The consumer steel market is in the opposite direction, with increasingly stringent requirements, especially in relation to residual levels of undesirable elements such as phosphorus and where steel manufacturing and treatment processes are limited and/or costly. With the need to obtain a pig iron with lower final phosphorus levels and with the lowest cost of the dephosphorization process, pre-treatment of hot metal becomes an important ally of steelmakers. Aiming for greater efficiency in the treatment of dephosphorus studied, a mathematical model was developed capable of predicting the phosphorus content at the end of the dephosphorization treatment in a hot metal pretreatment station. The model took into account the amount of liquid hot metal to be treated, the initial phosphorus content, the initial temperature, the initial silicon contentand the specific consumption of the dephosphorizing agent. These data were collected during the monitoring of the current process of the dephosphorization station. For the development of the model, experimental heats were carried out at Siderúrgica Alterosa S/A, which has a hot metal dephosphorization plant using a micropowdereddephosphorizing agent produced at Tecnosulfur S/A. The model was developed based on the concept of multiple linear regression, and proved to be efficient in predicting the final phosphorus content in pig iron after the treatment of dephosphorization in a ladle station