Modeling the effect of natural convection on macrosegregation of steel ingot by using a novel method

A program is developed to simultaneously solve macroscopic mass, momentum, energy, and species conservation equations during solidification based on a volume-averaged model. The program is applied to a binary Fe-C steel ingot to examine the effect of natural convection on the heat, melt flow and solute transfer and further to predict the macrosegregation quantitatively. Unidirectional heat transfer boundary conditions are assumed. The numerical results show that convection dramatically alters the shape of isotherms and the distribution of solute concentration. The linear equations are solved using Bi-conjugate Gradient Stabilized (BI-CGSTAB) and multi-grid (MGM) iterative methods, which greatly improve the computing efficiency. The predicted results using novel algorithm are compared with those using conventional method and they are more accurate. These robust algorithms are available to simulate real castings with high accuracy and rapid speed.