PR-503-026 - 2019 STEELSIM Conference Proceedings

Dynamic EAF Process Model Thermochemistry and Further Development

Hay, Reimann, Echterhof, Pfeifer

Download PDF

View Sample Page

process parameters. Mathematical models have been used successfully to predict values for these parameters and better understand the process. Generally, statistical (empirical) and analytical (deterministic) models can be used to model complex phenomena such as the EAF process. Statistical models are usually easier to develop and give more accurate predictions when compared to analytical models but are limited by the available data as they can only predict measurable parameters and extrapolation outside of the scope of the input data is not possible. Analytical models can offer deeper understanding of the process and predict values for parameters not measured and provided as input data. Furthermore, extrapolation is possible and a model based on physical principals is valid for different plants while a statistical model is always specific to a single operation that data was provided for.[1] Several models, both analytical and statistical in nature, have been developed for online EAF control as well as for off-line optimization and analysis either of the complete system or of limited aspects such as burners, carbon injection or the electric arc. A comprehensive analytical EAF model can offer better process understanding as well as the potential for process optimization and, if the simulation can be executed fast enough, online process control.

Keywords: Electric Arc Furnace, Process Model, Thermochemistry, Slag