2008 年 48 巻 7 号 p. 891-900
A coupled numerical model based on computational fluid dynamics (CFD), termed a CFD-based nucleation-growth-removal model, has been proposed to investigate the nucleation, growth and removal of inclusions in a bottom-blown gas-stirred ladle during molten steel deoxidation.
In order to account for the turbulent circulating flow induced by gas bubbling through the bottom, the κ–ε turbulence model was employed to evaluate time-averaged Reynolds stress while the fluctuation of the free surface of the melt was traced by means of the VOF technique. This model enables to couple transient flow fields, and temperature fields with concentration fields of elements and inclusions in melt so that it can meet the requirements of evolution of time- and space-dependent particle size distributions (PSD) without any given initial PSD of inclusions but directly from chemical reaction.
The model was applied to the simulation of the molten steel deoxidation process in Fe–Al–O system in an argon gas-stirred ladle. The model successfully predicted important phenomena including turbulent recirculation flow patterns, streamline structure of steel melt, turbulence energy variation and its dissipation rate, transient temperature fields and particle size distributions of alumina inclusion. Comparison with some industrial data showed that the predicted PSDs were reasonable in time-dependent variation of inclusion size distribution.
