Three Dimensional Mathematical Modeling Of Glass Melting Process

This paper presents an integral mathematical model to simulate glass melting process. The heart of the model is two and three dimensional flow model capable of calculating flows and temperature distribution in the melt, two dimensional flow model also considers air bubbling and electric boosting. In order to predict the glass quality, the whole model combines some additional sub-models which describe the glass melt tracking, melting kinetics, homogenizing, refining and corrosion of refractory. The validity of the present computational method is tested by comparison with the previous experimental results in a physical model.In order to improve model experiment of glass furnaces, the laminar natural/mixed convection of high viscosity fluid has been numerically studied in detail. It is found that both natural convection and buoyancy dominated mixed convection in high viscosity fluid can be divided into three flow regimes according to Ra: creeping regime, transition regime and boundary layer regime. For creeping motion regime, the characteristic velocity of natural convection is proportional to Gr and independent Pr, the quantity describing therelative magnitude of natural to force convection becomes Gr/Re, rather than Gr/Re². For boundary layer regime, the characteristic velocity of natural convection in a rectangular enclosure is of order of (Gr/Pr)^1/2, the quantity representing the relative magnitude ofnatural to force convection can be Gr/(Re²Pr), instead of Gr/Re² for given Pr. For buoyancy dominated mixed convection in an electronic glass furnace, the two similarity criteria Ra and Pr can approximately realize the similarity between the prototype and the model.According to 3D numerical simulation for the glass melting process under various working status, the major results are summarized as follows: (1) there isn't any backward flow within the throat, which is different from 2D simulation. (2) the different pulls at two foreharths haven't any effect on the melting process. (3) the width of throat has influence on both the backward flow in the throat and refining process, there exists an optimum width about ten percent of the furnace width. (4) heat loss from bottom wall has more effect on the glass melting process than those from side wall and bridge wall. (5) unsymmetrical transversal surface temperature distribution will change the transversal flow pattern and reduce the glass melting quality.Mathematical modeling has already been applied in the reconstruction of glass furnace and troubleshooting with success.