Mathematical modeling of fluid flow phenomena during tundish filling and subsequent initial casting operation in steel continuous casting process

The purpose of this study is to develop a mathematical model to analyze the fluid flow phenomena of molten steel in the tundish during its filling sta ge and subsequent initial casting operation in the continuous casting proce ss of steel. The ultimate goal is to assure smooth initial casting operatio n without nozzle clogging by avoiding high deposition rate of inclusion on any of the tundish outlets during tundish filling and subsequent initial ca sting operation. The mathematical model is developed based on a computational fluid dynamics technique, named SOLA-MAC, and the k-epsilon two-equation turbulence model . SOLA-MAC technique has the ability to handle the flow problem encountered in tundish filling, which is a transient flow problem with highly distorte d free surfaces and the locations of the free surfaces are to be determined by theory. A fluid particle method is also employed in this study to analy ze the distribution of inclusions in the molten steel and the extent of inc lusion contamination in various strands of the continuous casting tundish. A water model that is one-fourth the scale of an actual billet continuous e aster is also constructed in this study. Water model experiments are conduc ted to verify the accuracy and reliability of the mathematical model. The developed model is first tested on the water model to calculate the flo w pattern of water in the tundish during the very early stage of filling op eration. The simulated filling patterns are compared to the water model exp eriments. Good consistency is observed. The model is then tested on an actu al billet continuous easter with four strands to simulate the fluid flow ph enomena of molten steel in the tundish during the filling and subsequent in itial casting operations. Inclusion distribution and the extents of inclusi on contamination among the outlets of the various strands in the tundish ar e also analyzed. The simulated results show that for the left half of tundi sh, inclusion contamination in #2 strand is significantly more severe then that in #1 strand. This is confirmed by the actual experience on the shop f loor of that particular billet easter that #2 strand experiences more diffi culty in clogging problem during the initial casting operation than #1 stra nd does.