LBM-LES Simulation of the Transient Asymmetric Flow and Free Surface Fluctuations under Steady Operating Conditions of S
- PDF / 5,048,908 Bytes
- 15 Pages / 593.972 x 792 pts Page_size
- 89 Downloads / 162 Views
teel in the mold presents transient characteristics and is complicated by the fluctuations of the top free surface of the mold. Accordingly, various flow structures are resulted and affect the quality of the final product, significantly.[1] Plant observations suggest that the asymmetrical distribution of defects in the slab, slip crack, pinholes, ‘‘pencil pipe’’ slivers, as well as blisters are, to a large extent, associated with the slag inclusions and gas bubbles entrapment in the mold.[2] In essence, the motion behaviors of either inclusion droplets or gas bubbles
PENG ZHAO, Ph.D. Candidate, QIANG LI, Associate Professor, and ZONGSHU ZOU, Professor, are with the School of Metallurgy, Northeastern University, Heping District, Shenyang 110819, P.R. China, and also with the Key Laboratory of Ecological Utilization of Multi-metallic Mineral of Education Ministry, Northeastern University, Heping District, Shenyang 110819, P.R. China. Contact e-mail: [email protected] S.B. KUANG, Research Fellow, is with the Department of Chemical Engineering, Monash University, Clayton, VIC 3800, Australia. Manuscript submitted January 28, 2016. METALLURGICAL AND MATERIALS TRANSACTIONS B
are influenced by the transient flow behaviors in regard with the asymmetrical jets from the Submerged Entry Nozzle (SEN), the multiple vortices inside the mold, and the free surface fluctuations at the top of the mold. Therefore, a better understanding of the flow behaviors of liquid steel in the mold, especially their turbulence and fluctuation characteristics, can be very useful to develop various measures and strategies to control the quality of casting steel product. In the past years, in line with the development of computational technology, computational fluid dynamics (CFD) studies, combined with physical experiments, have been increasingly carried out in this regard, as brief reviewed below. To simulate the turbulent flow of molten steel in the continuous casting process, Reynolds-Averaged Navier-Stokes (RANS) is widely used in various turbulence flows such as k-e, k-x, and Reynolds stress models.[3–5] The RANS models provide time-averaged variables of the turbulent flows. Therefore, they cannot capture the instantaneous turbulence characteristics, which can play important roles for the complicated transfer behaviors of the inclusion droplets and gas bubbles of various types. This problem may be overcome by Direct Numerical Simulation (DNS), which
solves all scales of turbulence. However, DNS is computationally extremely demanding if not impossible when applied to high Reynolds numbers, e.g., above 105 in the continuous casting process. A compromised way is to employ Large Eddy Simulation (LES) between RANS and DNS computations. In the LES approach, large-scale eddies are solved directly with the NavierStokes (N-S) filtered equations, while the small-scale eddies are modeled as subgrid-scale (SGS) grids. Thus, it can, to a large extent, provide the details of the turbulent flow in the mold with the current computer capacity. In the past few years, variou
Data Loading...
