Today, Vortex Generators (VGs) are becoming an integral part of a Wind Turbine blade design. However, the challenges that are involved in the computation of the flow around VGs are yet to be dealt with in a satisfactory manner. A large number of VG models for flow solvers have been proposed and among them, the BAY model is one of the most popular for its ease of use and relatively low requirements for user input.
In the present paper, a thorough investigation on the performance and application of the BAY model for aerodynamic Vortex Generator flows is presented. A Fully Resolved Reynolds Averaged Navier Stokes simulation is validated against experiments and then used as the benchmark for the BAY model simulations. The Benchmark case is the flow past a wind turbine airfoil at Reynolds number 0.87e6. When the grid related errors are excluded, it is found that in the model simulations, the generated vortices are weaker than in the fully resolved computation. The latter finding is linked to an inherent deficiency of the model, which is explained in detail. As the vortex generation mechanism is different between the fully resolved and the BAY model simulation, so is the vortex evolution and interaction, even on the same numerical mesh. With regards to grid dependence, the integral BAY force depends on both grid density and grid architecture.