High Fidelity Numerical Investigations of Tailored Magnetic Fields for Defect Reduction in Continuous Casting of Steel

该补助金为研究通过应用定制磁场来改变结晶器区域的流体流动以减少缺陷的形成来提高连铸钢质量的策略提供资金。磁场对钢流施加额外的力，从而改变影响夹杂物运输的速度和在当前工艺中产生大多数缺陷的液位波动。为了实现这一目标，将根据湍流及其与磁场相互作用的精确建模开发高保真计算机模型。基于图形处理单元的高级并行计算机将用于减少计算时间并提高网格分辨率。开发的计算机模型将被应用于研究模型和工业几何形状的流场，以表征不同磁场配置的影响。最终目标是获得性能和实际可行性最理想的配置。如果成功，该项目将减少缺陷，并减少钢铁连铸过程中的能源，材料和资源浪费，该过程用于生产美国和全球90%以上的钢铁。主要研究人员通过伊利诺伊大学的连铸联盟与几家钢铁生产商进行了密切的互动，该联盟提供了行业视角以及将结果付诸实践的手段。该研究有望有助于对复杂区域中不同磁场配置下湍流流动的基本理解。使用图形处理单元进行流体流动模拟最近在速度和准确性方面取得了显着优势，这项工作也将扩展到多个图形处理单元的集群。

This grant provides funding for the investigation of strategies to improve the quality of continuous-cast steel by applying tailored magnetic fields to alter fluid flow in the mold region to reduce the formation of defects. Magnetic fields apply additional forces to the steel flow which change the velocities that affect the transport of inclusions and level fluctuations that produce most of the defects in the current process. To accomplish this, high-fidelity computer models will be developed based on accurate modeling of the turbulence and its interactions with the magnetic fields. Advanced parallel computers based on graphics processing units will be used to reduce the computational times and increase the mesh resolution. The computer models developed will be applied to study flow fields in model and industrial geometries to characterize the effects of different magnetic field configurations. The ultimate objective is to obtain the most desirable configuration for performance and practical viability.If successful, the project will reduce defects and reduce the waste of energy, material and resources during the continuous-casting of steel, a process which is used to produce over 90 percent of the steel in the United States and globally. The principal investigators have close interactions with several steel producers through the Continuous Casting Consortium at the University of Illinois which provides an industry perspective as well as the means to implement results into practice. The research is expected to contribute to the fundamental understanding of turbulent flows in complex domains subjected to varied magnetic field configurations. The use of graphics processing units for fluid flow simulations has recently enabled significant advantages in speed and accuracy and this work will also be extended to clusters of multiple graphics processing units.