Argon bubble transport and capture in continuous casting with an external magnetic field using GPU-based large eddy simulations

Jin, Kai

Permalink

https://hdl.handle.net/2142/95333 Copy

Description

Title

Argon bubble transport and capture in continuous casting with an external magnetic field using GPU-based large eddy simulations

Author(s)

Jin, Kai

Issue Date

2016-11-23

Director of Research (if dissertation) or Advisor (if thesis)

• Thomas, Brian G.
• Vanka, Surya P.

Doctoral Committee Chair(s)

• Thomas, Brian G.
• Vanka, Surya P.

Committee Member(s)

• Ruzic, David N.
• Brooks, Caleb
• Smith, Kyle C.

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Computational fluid dynamics (CFD)
• Multi-GPU Computing
• Continuous Casting
• Two-phase flow
• Electromagnetic Braking

Abstract

Continuous casting produces over 95% of steel in the world today, hence even small improvements to this important industrial process can have large economic impact. In the continuous casting of steel process, argon gas is usually injected at the slide gate or stopper rod to prevent clogging, but entrapped bubbles may cause defects in the final product. Many defects in this process are related to the transient fluid flow in the mold region of the caster. Electromagnetic braking (EMBr) device is often used at high casting speed to modify the mold flow, reduce the surface velocity and fluctuation. This work studies the physics in continuous casting process including effects of EMBr on the motion of fluid flow in the mold region, and transport and capture of bubbles in the solidification processes. A computational effective Reynolds-averaged Navier-Stokes (RANS) model and a high fidelity Large Eddy Simulation (LES) model are used to understand the motion of the molten steel flow. A general purpose multi-GPU Navier-Stokes solver, CUFLOW, is developed. A Coherent-Structure Smagorinsky LES model is implemented to model the turbulent flow. A two-way coupled Lagrangian particle tracking model is added to track the motion of argon bubbles. A particle/bubble capture model based on force balance at dendrite tips is validated and used to study the capture of argon bubbles by the solidifying steel shell. To investigate the effects of EMBr on the turbulent molten steel flow and bubble transport, an electrical potential method is implemented to solve the magnetohydrodynamics equations. Volume of Fluid (VOF) simulations are carried out to understand the additional resistance force on moving argon bubbles caused by adding transverse magnetic field. A modified drag coefficient is extrapolated from the results and used in the two-way coupled Eulerian-Lagrangian model to predict the argon bubble transport in a caster with EMBr. A hook capture model is developed to understand the effects of hooks on argon bubble capture.

Graduation Semester

2016-12

Type of Resource

text

Permalink

http://hdl.handle.net/2142/95333

Copyright and License Information

Copyright 2016 Kai Jin

Owning Collections