Numerical studies of the turbulent planar mixing layer using nonlinear k-ε model

Keh Chin Chang, Uzu Kuei Hsu, Yi Chang Kuo

Research output: Contribution to journalArticlepeer-review

Abstract

The two phase mixing layer flow that forms between two fluid streams moving with different velocities. In order to simulate the downstream flowfield from the developing to self-preserved regions of the turbulent planar mixing layer, it cost lot of computational resource using DNS (Direct Numerical Simulation) or LES (Large-Eddy Simulation) method to solve. In generally, the two-equation turbulent model is usually, but the flowfiled remains the non-isotropic eddy-viscosity problem. There is no an art-of-state numerical method accuracy to simulate in recently. Hence to solve this problem using RANS method we rely on turbulence modeling techniques, in which the nonlinear eddy-viscosity formulation perform on the predictions of the Reynolds stresses to replace the model associated with the Boussinesq hypothesis. Two cases with inflow-speed-radio r=0.62 and 0.39 were testing. From RANS turbulence models, including the standard k-ε model which is under the Boussinesq hypothesis, the explicit algebraic stress model of GS (1993), the nonlinear eddy-viscosity model of Abe (2003), and the low Reynolds number explicit algebraic stress model of Rahman (2006), are tested and their predictions of flow properties are compared with the measurement data. It shows that nonlinear eddy-viscosity model can achieve accurate quantitative simulation as well as a reasonable Reynolds stresses of the mixing layer.

Original languageEnglish
Pages (from-to)43-54
Number of pages12
JournalHangkong Taikong ji Minhang Xuekan/Journal of Aeronautics, Astronautics and Aviation, Series B
Volume43
Issue number1
Publication statusPublished - 2011 Apr

All Science Journal Classification (ASJC) codes

  • Aerospace Engineering
  • Space and Planetary Science

Fingerprint

Dive into the research topics of 'Numerical studies of the turbulent planar mixing layer using nonlinear k-ε model'. Together they form a unique fingerprint.

Cite this