The burden distribution in the blast furnace affects the efficiency of the upper part of the process and is, therefore, associated with smoothness of operation, coke rate and productivity. A burden distribution model is applied in this work to evolve charging programs fulfilling a goal given by the designer by evolving (mainly) angles for the rotating chute. The charging program is based on a genetic algorithm. It is demonstrated that the method can make efficient progress in the complex search space, finally designing charging programs that produce distributions of ore and coke which closely fit the target patterns. The technique holds promise for application in the ironmaking industry for evolving novel charging programs.
|Tidskrift||Computer Methods in Materials Science|
|Status||Publicerad - 2013|
|MoE-publikationstyp||B1 Artikel i en vetenskaplig tidskrift|