A COMPARATIVE ASSESSMENT OF TURBULENT FORCED CONVECTION HEAT TRANSFER FROM A SINGLE CYLINDER USING RANS AND LES MODELS


ALTAÇ Z., MAHİR N.

ISI BILIMI VE TEKNIGI DERGISI-JOURNAL OF THERMAL SCIENCE AND TECHNOLOGY, cilt.38, sa.1, ss.11-24, 2018 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 38 Sayı: 1
  • Basım Tarihi: 2018
  • Dergi Adı: ISI BILIMI VE TEKNIGI DERGISI-JOURNAL OF THERMAL SCIENCE AND TECHNOLOGY
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, TR DİZİN (ULAKBİM)
  • Sayfa Sayıları: ss.11-24
  • Eskişehir Osmangazi Üniversitesi Adresli: Evet

Özet

Unsteady 2D and 3D turbulent flow and heat transfer characteristics of a single isothermal horizontal cylinder in crossflow of air (Pr=0.7) is investigated to assess the numerical performance of the common turbulence models currently in use. For 2D simulations, Standard k-epsilon (SKE), Re-Normalization Group k-epsilon (RNG), Realizable k-epsilon (RKE), Standard k-omega (SKW), Shear Stress Transport (SST) k-omega and Reynolds Stress Model (RSM) turbulence models are used in conjunction with the two-layer wall (or Enhanced Wall Treatment, EWT) model. For 3D simulations, RNG, SKW, SST, RSM and Large Eddy Simulation (LES) using Smagorinsky-Lilly with and without dynamic stress models are used. In this study, the performance criterion of the turbulence models is based on the accuracy of the computational predictions of 2D and 3D flow as well as heat transfer (C-D, C-L, C-L,C-rms, St and Nu numbers) characteristics. Numerical simulations are carried out for Reynolds numbers of 1000, 3900 and 10000 using FLUENT 6.3.26 (R) CFD software. The flow characteristics, such as the lift/drag coefficients and Strouhal numbers, are computed and tabulated comparatively with available experimental and numerical data. The 2D RANS models are not consistent in predicting the flow characteristics due to three-dimensionality nature of the fluid flow, but RSM performs slightly better than RANS models. The mean Nusselt number for Re=1000 and 3900 is predicted with reasonable accuracy with 2D-RANS models. While the RNG model consistently over estimates the mean Nusselt number, other 3D-RANS models yield values within the ranges predicted by the Nusselt number correlations. It is shown that although LES models yields reasonable flow and heat transfer characteristics for flow conditions considered here, the performance of LES is also dependent on the inlet condition.