|About this Abstract
||2018 TMS Annual Meeting & Exhibition
||9th International Symposium on High Temperature Metallurgical Processing
||Research on Factors Affecting and Prediction Model of Silicon Content in Hot Metal of Corex
||Bingjie Wen, ShengLi Wu, Heng Zhou, Jiacong Zhang, Kai Gu
|On-Site Speaker (Planned)
In practical production process, the average of silicon content in hot metal (HM) of COREX process (the average is 1.58%) is obviously higher than that in blast furnace (the value is below 0.6%), which leads to an increase in the cost of steelmaking. In this work, the factors affecting and impact mechanism of silicon content in HM are investigated by statistical analysis using actual operating data. The results show that the fuel rate, binary basicity of slag and the temperature of HM are positively correlated with the silicon content, while the sulfur content of HM and binary basicity of burden are negatively correlated with the silicon content. On this basis, a neural network model based on back propagation is developed to predict and control the silicon content in HM. All the findings of this work are useful for the guiding and optimizing of the COREX operation.