The effects of different RANS closures on modelling swirling flow in a macro-scale multi-inlet vortex reactor

The swirling flow in a macro-scale multi-inlet vortex reactor (MIVR) is modeled using RANS (Reynolds-averaged Navier-Stokes) simulations.  Typically, a linear eddy viscosity model such as k-ε or k-ω would be used. However, it is challenging to apply conventional RANS models to study the macro-scale MIVR because of the swirling flow, out-of-plane flow, and backflow at the center of the reactor. Furthermore, the high degrees of anisotropy, streamline curvature, rotational effects, and recirculating flow make the flow simulation difficult. Therefore, the Reynolds-Stress transport models such as the LRR [1] model and the SSG [2] model are applied to investigate the possibilities of improving flow prediction.  In this study, CFD results were compared to the experimental data available in the literature [3] using different RANS models such as k-ε, RNG k-ε, Realizable k-ε, k-ω, LRR, and SSG at different Reynolds numbers.  Each velocity component will be presented with respect to the radius for comparison.

References:

[1] B. E. Launder, G. J. Reece, and W. Rodi, “Progress in the development of a Reynolds-stress turbulence closure,” J. Fluid Mech., vol. 68, no. 03, pp. 537–566.

[2] C. G. Speziale, S. Sarkar, and T. B. Gatski, “Modelling the pressure–strain correlation of turbulence: an invariant dynamical systems approach,” J. Fluid Mech., vol. 227, pp. 245–272.

[3] Z. Liu, A. Passalacqua, M. G. Olsen, R. O. Fox, and J. C. Hill, “Dynamic delayed detached eddy simulation of a multi-inlet vortex reactor,” AIChE J., vol. 62, no. 7, pp. 2570–2578.