Modelagem física e matemática da mistura de aços em um distribuidor de veio único de lingotamento contínuo de placas durante a troca de panelas

Abstract

In the continuous casting process, the tundish performance is important in producing high-quality steel. During ladle exchanges, significant variations in the flow pattern inside the tundish can occur, affecting the temperature distribution and chemical composition of the steel, which can jeopardize the quality of the produced slab. In some cases, the ladle exchange is accompanied by variations in the chemical composition of the steel being cast. This thesis investigated the mixing phenomenon of different steel grades inside a single-strand slab continuous casting tundish during ladle exchanges. The effects of thermal phenomena and minimum bath level on the mixing pattern are evaluated through reduced-scale physical modeling, using water as the working fluid due to its similarity to liquid steel in terms of kinematic viscosity. The dimensionless Froude number was adopted as similarity criteria. Four experimental scenarios were considered, varying thermal conditions (isothermal and non-isothermal) and the minimum bath level of the tundish (high and low). Mathematical modeling and numerical solution was conducted using Ansys Fluent 22R1 software to simulate the multiphase flow in the vessel. The mathematical methodology considers the Volume of Fluid (VOF) multiphase approach. Different software parameters, including turbulence models (SST k-ω and realizable k-ε) and pressure-velocity coupling schemes (SIMPLE - Semi-Implicit Method for Pressure-Linked Equations, PISO - Pressure-Implicit with Splitting of Operators, and COUPLED) were tested to validate the numerical calculations with the physical model. Experimental results revealed that thermal effects and the minimum bath level significantly influence the tracer concentration curve at the tundish outlet. In non-isothermal scenarios, temperature gradients generate variations in the flow pattern, which affect the mixing between different types of steel and the location of dead volumes inside the tundish. Reducing the minimum bath level proved to be beneficial in minimizing steel contamination during ladle exchanges. Numerical simulations corroborated the experimental results. The comparison between different numerical parameters of the software highlighted the importance of appropriate parameter selection for simulation accuracy and duration. It is concluded that the integration of physical and mathematical modeling approaches provides a comprehensive understanding of the phenomena involved in steel mixing during ladle exchanges. The results of this study contribute significantly to the optimization of the continuous casting process, improving the efficiency and quality of the produced steel.