Numerical study of particle filtration in an induction crucible furnace

The present paper deals with a numerical investigation of the turbulent melt flow driven by the electromagnetic force in an induction furnace. The main scope of the paper is to present a new principle to remove non-metallic particles from steel melt in an induction furnace by immersing a porous filter in the melt. The magnetic field acting on the melt is calculated by using the open source software MaxFEM®, while the turbulent flow is simulated by means of the open source computational fluid dynamics library OpenFOAM®. The validation of the numerical model is accomplished by using experimental results for the flow without the immersed filter. Here it is shown that the time-averaged flow, obtained numerically is in a good quantitive agreement with the experimental data. Then, the validated numerical model is employed to simulate the melt flow with the immersed filter in the induction furnace of a new type of real steel casting simulator investigated at Technische Universität Bergakademie Freiberg. The considerable effect of the filter on the flow pattern is indicated in the present work. Moreover, it is shown that the filter permeability and its position have a significant influence on the melt flow in the induction furnace. Additionally, particles are injected in the flow domain and tracked by using Lagrangian framework. In this case, the efficiency of the used filter is determined in the present investigation depending on its permeability, its position and the particles diameter.