Computational Methods for Multi-Physics Simulation of Melting in Steelmaking

Research output: NRELPoster

Abstract

Iron and steel production accounts for approximately 8% of global carbon dioxide (CO) emissions. Pathways to decarbonize include replacing fossil fuels in iron ore reduction and electrifying other steelmaking processes. Iron pellets produced by hydrogen, called Hydrogen Direct Reduced Iron (HDRI), have property differences from those produced using conventional DRI processes. These differences may impact melting in electric arc furnaces (EAF) and other downstream processes. The physical properties of iron pellets vary significantly with temperature during heating, complicating predictions of their behavior. In this project, we seek to develop an integrated simulation, including the fluid flow and convective thermal transport around the pellet particle. We also examine conduction and phase changes within the particle as they impact the melting process. We use adaptive mesh refinement (AMR) to resolve both the changing size of the particle and the complex physics of the interaction between the pellet and the surrounding fluid. We base our simulations on the AMReX-incflo module, which allows large-scale Navier-Stokes simulation while resolving the changing particle size during melting. As we advance our numerical tools, we anticipate an improved understanding of the dynamics of HDRI melting, which will, in turn, accelerate the adoption of low-carbon technologies in the steelmaking industry.
Original languageAmerican English
PublisherNational Renewable Energy Laboratory (NREL)
StatePublished - 2024

Publication series

NamePresented at the 2024 SIAM Annual Meeting, 8-12 July 2024, Spokane, Washington

NREL Publication Number

  • NREL/PO-2C00-90320

Keywords

  • AMReX
  • EAF
  • fluid dynamics
  • steelmelting
  • thermal transport

Fingerprint

Dive into the research topics of 'Computational Methods for Multi-Physics Simulation of Melting in Steelmaking'. Together they form a unique fingerprint.

Cite this