COMPARISON OF LINEAR AND NONLINEAR RNG-BASED k-epsilon MODELS FOR INCOMPRESSIBLE TURBULENT FLOWS

Linear and nonlinear renormalization group (RNG) k-epsilon models are compared for the prediction of incompressible turbulent flows. The multidimensional finite-volume code KIVA-3 \[1] is used to explore the alternative models versus the standard k-epsilon model. Test cases include the classic backward-facing step and the confined co-flow jet flows. Our results suggest that the linear RNG k-epsilon model can yield significant improvements over the standard k-epsilon model for recirculatory flows, because of its less dissipative nature. While the nonlinear RNG k-epsilon model can also improve predictions compared to the standard k-epsilon model, its greatly increased cost compared to the linear RNG model renders it less attractive. However, for the case of shear flows, such as for confined co-flow jets, the RNG-based k-epsilon models are in less favorable agreement with experiments compared to the standard k-epsilon model. Overall, it is concluded that combining the claimed universality of the RNG-based k-epsilon model constants with the anisotropies introduced by the nonlinear k-epsilon model cannot enhance predictions of both recirculating and shear incompressible flows.