RANS解析における水車ランナを模擬した翼列に発生する剥離予測精度の改善検討

Abstract

The investigation of turbulence model in CFD using RANS model was carried out to improve the stall prediction accuracy. Decreasing the coefficient of eddy viscosity in k-ω SST turbulence model improved the prediction accuracy of the separating flow on a NACA631-012 blade, but it made the prediction accuracy of the non-separating flow worse. Thus, the modified turbulence model that decrease the eddy viscosity based on the adverse pressure gradient and GEKO turbulence model were applied to analysis of a NACA631-012 blade. As a result, both of these turbulence model improved the predection accuracy of the lift characteristic in a wide range of angles of attack. In addition, it was clarified that the separating flow on the blade simulating the hydro turbine runner can be predicted with high accuracy by using these turbulence model. It is expected that the modified turbulence model based on the pressure gradient and GEKO model will improve the performance prediction accuracy at off design operationg points in the analysis of hydro turbine runner.