Effect of Computational Grid on Prediction of a Vertical axis Wind turbine Rotor Using Delayed Detached-Eddy Simulations

Abstract

The paper presented the grid influence on the prediction of aerodynamic performances of the vertical axis wind turbine DARRIEUS type H. Two different blade mesh topologies, structured and unstructured meshes and different grid resolutions are tested.The simulations are carried out using Delayed detached-eddy simulation (DDES) with two eddy viscosity turbulence models, Spalart-Allmaras (SA) and Menter Shear stress transport (SST) k-o). A high order spatial discretization is used in these computations. For each proposed rotor configuration, flow field characteristics are investigated at same values of tip speed ratio with an incoming wind velocity of $8\mathrm {m}/\mathrm {s}$, allowing a quantification of the influence of grid on flow features and dynamic quantities, such as rotor torque and power. The structured mesh topology is observed to give the best results employing the SST $\mathrm {k}-(j)$ turbulence model, but the computational cost is more expensive as the grid contains a wake block that increases the number of cells.