Yield Prediction Based on Bed Temperature and Component in Sintering Process

Abstract

The yield in the iron ore sintering process is managed in daily operations as an important indicator of manufacturing costs and CO₂ emissions. Currently, general yield operation actions are "feedback" based on yield results, and do not prevent yield loss. In order to prevent yield decline by "feed-forward" type action targeting yield changes, yield prediction based on operating conditions is essential. In this study, we assumed that yield is synonymous with the strength of the discharged sintered cake, and attempted to apply the porous material strength estimation formula to the sintered cake. The sintered cake was divided into a cake pore part lost by crushing and a sintered ore part as a matrix portion containing pores that remain after crushing, and the formulation of cake porosity and sintered ore strength was studied. Using the amount of melt (melting ratio) generated in liquid-phase sintering and its penetration length into the ore as key factors, the estimation formulas for cake porosity and sintered ore strength were built. In addition, experiments using diffusion pairs were carried out to verify the hypothesis regarding the penetration distance. A pot test were carried out to verify the constructed model , and confirmed a good correlation between the calculated yield by bed temperature measured and the blending components, and the actual measured yield.