Smooth delivery of molten metal to a chilled substrate is a key prerequisite for the development of a successful technique for thin-strip casting steel. In this work, a numerical model of a thin-strip, liquid steel delivery system was developed, validated, and used to make design predictions. The proposed slot-type planar-flow delivery system could contain a porous flow modifier, and features an extended-pool metal delivery system to the chilled substrate.;The steady-state, computational-fluid-dynamic (C. F. D.) model developed for this research is two-dimensional, and uses the finite-difference, control-volume formulation. The implicit solutions for the fluid flow and energy fields are fully coupled and include treatment of the solidification phenomena using the enthalpy-porosity approach. The model includes porous flow modifying regions within the delivery zone. These are treated as regions of complex media and make use of the "Brinkman-Forchheimer extended-Darcy" model. Turbulence was modelled using an ad hoc approach for both effective viscosity and effective thermal conductivity.;Numerical predictions of fluid flow were compared with experimental flow measurements and flow visualization using a water model of the proposed delivery system. The experiments confirmed that the flow modifiers had a very beneficial smoothing effect on fluid delivery to the substrate in the feeding system. The numerical predictions were in good agreement with the experimental results. As well, numerical predictions of shell thickness were compared with several different simple semi-analytical test cases.;The model was used for several parametric studies. The effect of flow modification, in conjunction with varying pool lengths, was studied. The model predicted that the presence of the flow modifiers would smooth the fluid flow to the substrate, and promote even extraction of heat, despite extension of the reservoir's length. Extending the pool length delayed the onset of solidification. Another study made predictions for cases with differing inlet and substrate boundary conditions, as well as for different exit gap sizes. This part of the work demonstrated the type of predictions possible with this model for use in the design of a prototype of the proposed delivery system.
展开▼
