Master thesis and internship[BR]- Master's thesis : Towards simulations of supersonic oxygen blowing in steelmaking with OpenFOAM®[BR]- Internship

Abstract

Seeking to reduce Europe's emissions, all the industrial processes are optimized. As steel is omnipresent in our lifestyles, making steel greener has become a key target. Steelmaking is a complex process subdivided into multiple stages. Each must be fully controlled to optimize the quality of the steel, the service life of the equipment, and reduce the required energy. In this study, the conversion process of molten iron into steel inside the Basic Oxygen Furnace is investigated. This process is mainly driven by the blowing of supersonic oxygen jets onto molten iron. Understanding and being able to predict the behavior of the supersonic jets in this rough environment, and the way it penetrates the molten iron is mandatory for the process optimization. A great interest is thus paid to the modeling of these components. Simulations have already been conducted on licensed software. The objective of this paper is to develop this expertise with open source tools i.e., Gmsh and OpenFOAM®.

Hence a simplified model is considered to validate the tools. Firstly, a single supersonic jet in an atmosphere at room temperature is simulated. Then, a hot atmosphere at 1002K is introduced. The optimal simulation parameters are selected. The results are confronted to experimental, analytical and numerical references.

It is highlighted that structured meshes provide significantly more reliable results than unstructured ones. Moreover, the simulations with the k − ε turbulence model estimate the most accurately the flow. In order to further improve the prediction, the standard Cµ value must however be adjusted for both cold and hot configurations. Due to the structure of OpenFOAM®, implementing a deep change of the model to take into account the temperature fluctuations is highly complex. Therefore, despite being less elegant, hard-coding the adjusted Cµ is selected. Additionally the tools to characterize and analyze the flow are developed. Particularly a new methodology of method validation is proposed to verify the model. Besides looking at the centerline quantities, the profiles at different axial distances are compared to observe self-similarity. Finally a first insight at the influence of the lance height and of the temperature on the flow, and thus on the oxidation, is highlighted.

The adequate parameters to model the flow are thus identified and further, more complex, simulations can be initiated. Particularly, a 3-dimensional simulation to consider the interactions between the multiple jets of the lance can be explored.