Process model based on neural networks for online quality forecasts in direct reduced iron making
.
Inhomogeneous properties in iron ore can impair the quality, productivity and energy efficiency of direct reduction plants – especially if meaningful process information in real time is not available. This will not only affect the direct reduction process itself but also further downstream operations i.e. steel meltshop.
Providing reliable insight into real time measurements of process parameters, SIMETAL SIMPAX permits highly accurate prediction of quality parameters of metallization and carbon content during the ongoing process. This allows continuously improved understanding of the process, immediate respond and adaption in case of changes – resulting in more consistent quality, lower energy consumption, increased availability and reduced maintenance costs.