The Research Of Endpoint Composition Control In EAF-LF Steelmaking Process

EAF steelmaking is an important technique to promote the green manufacturing of iron and steel industry. The composition control of molten steel is the key link in the EAF metallurgy process. At present, without considering the relationships and influences among each procedure, the composition control in EAF steelmaking is analyzed isolatedly. On account of the carbon content of the EAF raw materials being increased dramatically by the hot metal charging technology, it is essential to enhance the oxygen supply intensity and as a result, it’s more difficult to control the endpoint carbon contentprecisely in EAF steelmaking. Meanwhile, it brings forth high requirements for controlling the oxygen content and the alloying composition in the refining process.With the development of the EAF steelmaking technology and the intelligent control technology in BOF steelmaking, the model control technology is applied in EAF steelmaking gradually. In this paper, the control model for the endpoint composition of molten steel in EAF-LF steelmaking process is built to realize the precise composition control of the molten steel.Firstly, this paper revise the model of calculating the material for electric arc furnace and ladle furnace as a whole EAF-LF steelmaking process, to get material model in EAF-LF steelmaking process under different proportion of molten iron.Through the analysis of composition changes in EAF-LF steelmaking process, the direction of the composition control of optimization is obtained.For EAF endpoint carbon control, decarburization reaction is studied based on oxygen operation and analysis of furnace gas. At the end of smelting, this study adopts the integrated BP network technology to predict endpoint carbon content. The BP neural network prediction reached89%in the range of0.025%, which can meet the requirements of the endpoint carbon control.For LF refining alloy composition control, through the establishment of alloy element yield dynamic library, alloy charging optimization model is established by using the two-stage simplex method. Through the operation of steelmaking is experimental running online, the optimization model of the alloy obtain the alloy yield of different steel grade, and improve the accuracy of dosage of the alloy. By optimizing alloy charging, the cost of alloy charging for each steel grade is reduced by6.76%～11.40%, and the average cost is reduced by9.74%.By the research of optimization for the EAF composition endpoint carbon control and LF alloy composition control, accurate control methods in EAF-LF steelmaking process is obtained. Aiming at optimal cost, the paper optimized deoxidization process and alloy charging process, to realize reasonable control method for calculating oxygen composition before LF refining.Through the above research, the method for composition precise control in EAF-LF steelmaking process is obtained. On this basis, combined with the optimal cost as the goal, the control model for the endpoint composition of molten steel in EAF-LF steelmaking process is built.Through experimental run online, the model stabilizes molten steel endpoint composition and alloy material cost, and the application results show:the standard deviation of EAF endpoint C is reduced by0.010%～0.050%; for LF refining endpoint composition, C standard deviation is reduced by0.001%～0.005%, Si standard deviation is reduced by0.001%～0.022%, Mn standard deviation is reduced by0.002%～0.007%, Cr standard deviation is reduced by0.002%～0.005%, Mo standard deviation is reduced by0.001%～0.002%; the cost standard deviation of the total alloy material per ton steel is reduced by0.753～21.225yuan.