Numerical Prediction of Turbulent Flow over 3D Sinusoidal Hill Using Non-orthogonal Hexahedron Grid and AMG Method

Abstract

Numerical predictions of the wind energy distribution over terrains are essential for the appropriate selection of a suitable site for a wind power plant. This paper presents three-dimension numerical simulations of turbulence flow over sinusoidal hills. AMG in HYPRE (High Performance Preconditioner Library) is introduced in this work and applied on solving the FVM-based discrete systems of the governing equations for turbulent flow. The three-dimensional non-orthogonal collocated grid systems are generated and the AMG-based algorithm is established. Turbulence is modelled with the standard model and flow parameters of Governing equations are solving by AMG method. The numerical results were obtained from the simulations of flow over three types of three-dimensional sinusoidal hills. Comparisons have been made by virtue of experimental data with the computational data. The numerical results agree with the experimental results, the numerical methods are effective.