Full‐Scale Large Eddy Simulation of a 185 t Ladle Optimizing a Hybrid Eulerian Grid and a Discrete Bubble Mapping Approach. Issue 6 (28th December 2021)

• Record Type: Journal Article
• Authors: Haas, Tim; Schubert, Christian; Eickhoff, Moritz; Pfeifer, Herbert
• Is Part Of: Steel research international. Volume 93:Issue 6(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12-28
• Subjects: computational fluid dynamics -- Euler–Lagrange mappings -- ladle furnaces -- large eddy simulations -- upscaling; Steel -- Periodicals; Steel -- Metallurgy -- Periodicals; 669.142
• DOI: 10.1002/srin.202100653
• Languages: English
• ISSNs: 1611-3683

Abstract

Abstract : Numerical simulation is a vital tool for the continuous improvement of all steelmaking processes. In previous works, the authors have dedicated different approaches to optimize modeling of fluid flow in the ladle. This includes a sensitivity analysis of validation techniques, assembling of a validation database, and a model optimization on the water model scale. Herein, the final part of this project is presented at which the optimized model is scaled to the real process. Associated problems are identified and solutions are proposed. To achieve a tradeoff between large eddy simulation (LES) grid requirements and current computational capacities, a hybrid meshing strategy is optimized. To overcome the intrinsic drawbacks of the Euler–Lagrange model, a mapping approach for phase decoupling is implemented and made publicly available on GitHub. The upscaling of semi‐empiric submodels is discussed. Good agreement with experimental mixing time measurements is found, indicating the success of upscaling. A first time‐averaged flow field of a full‐scale 185 t ladle with the LES model is presented. Due to the extreme computing requirements, these results may currently be used as a benchmark, but the trend toward increasing computing capacities may make LES models state of the art in the future. Abstract : A numerical model is scaled to the industrial process. A hybrid grid is optimized to fulfill the requirements of a large eddy simulation (LES) simulation. Intrinsic drawbacks of the Euler–Lagrange model are overcome by implementing a mapping strategy that makes the Lagrangian phase independent of the Eulerian grid. This results in the first time‐averaged full‐scale LES of a ladle.