Search

In order to research the effect of suction coefficients and directions on suction control, and the relationship between effectiveness and cost of suction, a local suction distribution on the upper surface of the NACA0012 airfoil is designed to control the flow separation. The effectiveness of the designed suction distribution is investigated numerically using Roe scheme and LU-SGS implicit scheme with dual-time-stepping technique. The mechanism of suction control is analyzed by comparing the velocity profiles in the boundary layer. The penalty for utilizing suction control is also estimated. The computational results show that the designed suction distribution is capable of suppressing the flow separations and improving the aerodynamic performances. With the increase of the suction coefficients, the lift-to-drag ratio increases fast firstly and then decrease slowly, while the maximum increase is about 130%. The analysis of energy consumption shows that the cost for using suction to suppress separation flow is relatively small with an appropriate suction coefficient. The effect of suction angles on suction control is obvious. As the suction angle becomes larger, not only the lift-to-drag ratio increases, but also the cost of suction decreases. These results will further help to provide new ideas and methods for flow control design. [ABSTRACT FROM AUTHOR]

In order to solve k-ω type turbulence model equations and Reynolds-averaged Navier-Stokes equations (RANS), cloud reconstruction technology, which can obtain well results from high aspect radio anisotropic cloud of points, is adopted in moving least squares meshless method. With the cloud reconstruction technique and the meshless method, it is realistic to simulate turbulence flows. The turbulence prediction abilities of the k-ω SST (Shear Stress Transport) and the k-ω TNT (Turbulent Non Turbulent) modes are demonstrated on a number of cases including NACA0012 airfoil, RAE2822 airfoil and backward-facing step. It can be found that; for weak advanced pressure gradient flows, the two models have the same ability for prediction; while in shock introduced separation turbulence flows, the SST model show better predicted ability of shock capture than TNT model; for large scale separation flows, there are some differences between the results of the two models and the experimental data, while the overall trends are in good agreement with the experimental data. [ABSTRACT FROM AUTHOR]