Modeling of steel grade transition in continuous slab casting processes

X. Huang, B. G. Thomas

Research output: Contribution to journalArticlepeer-review

Abstract

Mathematical models have been developed and applied to investigate the composition distributions that arise during steel grade changes in the continuous slab casting processes. Three-dimensional (3-D) turbulent flow and transient mixing phenomena in the mold and the strand were calculated under conditions corresponding to a sudden change in grade. The composition distribution in the final slab was then predicted. Reasonable agreement was obtained between predicted and experimental concentration profiles in the slab centerlines. Intermixing in the center extends many meters below the transition point, while intermixing at the surface extends above. Higher casting speed increases the extent of intermixing. Mold width, ramping of casting speed, and nozzle design have only small effects. Slab thickness, however, significantly influences the intermixing length of the slab. The axial transport of solute due to turbulent eddy motion was found to be many orders of magnitude greater than molecular diffusion and thus dominates the resulting composition distribution. Different elements, therefore, exhibited the same mixing behavior under the same casting conditions, despite having different molecular properties. Numerical diffusion caused by the finite difference schemes was investigated and confirmed to be much less important than turbulent diffusion. In the lower portion of the strand (lower than 3 m below the meniscus), the convection and diffusion can be reasonably approximated as one-dimensional (1-D) axial flow.

Original languageEnglish (US)
Pages (from-to)379-393
Number of pages15
JournalMetallurgical Transactions B
Volume24
Issue number2
DOIs
StatePublished - Apr 1993

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'Modeling of steel grade transition in continuous slab casting processes'. Together they form a unique fingerprint.

Cite this