CFD Computation of the H-Darrieus Wind Turbine—The Impact of the Rotating Shaft on the Rotor Performance

Abstract

Aerodynamics of the Darrieus wind turbine is an extremely complex issue requiring the use of very advanced numerical methods. Additional structural components of this device, such as, for example, a rotating shaft disturb the flow through the rotor significantly impairing its aerodynamic characteristics. The main purpose of the presented research is to validate the commonly-used unsteady Reynolds averaged Navier–Stokes (URANS) approach with the shear stress transport (SST) k-ω turbulence model based on the particle image velocimetry (PIV) studies of a two-bladed rotor operating at the moderate tip speed ratio of 4.5. In the present numerical studies, a two-dimensional turbine rotor with a diameter of 1 meter was considered. The following parameters were evaluated: instantaneous velocity fields; velocity profiles in the rotor shadow and aerodynamic blade loads. The obtained numerical results are comparable with the reference experimental results taken from the literature. The second purpose of this work was to examine the influence of the rotating rotor shaft/tower on the wind turbine performance. It has been proven that the cylindrical shaft reduces the power of the device by 2.5% in comparison with the non-shaft configuration.

Performance analysis of a Darrieus-type wind turbine for a series of 4-digit NACA airfoils

“…The minimum values of this torque coefficient are about 0.3 after 12 full rotations of the rotor. A similar conclusion was made by Rogowski (2019), who studied the effect of initial conditions by simulating 50 full rotor revolutions. His simulation showed that 15 full revolutions of the rotor was sufficient to obtain repeatable results of both aerodynamic blade loads and velocity profiles behind the rotor.…”

“…Release 17.1), the length of the time step size should be small enough for the required level of convergence of the solving equations to be obtained in the defined maximum number of iterations per time step. In addition, Rogowski (2019) showed that even if the assumed time step satisfies the desired convergence criterion, it may not be sufficient to capture all aerodynamic phenomena, such as, for example, the effect of the tower's aerodynamic shadow on wind turbine blade loads. Rogowski (2019) and Rezaeiha et al (2019a) agree that for a "clean rotor" (a rotor consisting only of blades) the length of the time step can be assumed equivalent to an azimuth increase, Δθ, of 0.1 deg.…”

“…In addition, Rogowski (2019) showed that even if the assumed time step satisfies the desired convergence criterion, it may not be sufficient to capture all aerodynamic phenomena, such as, for example, the effect of the tower's aerodynamic shadow on wind turbine blade loads. Rogowski (2019) and Rezaeiha et al (2019a) agree that for a "clean rotor" (a rotor consisting only of blades) the length of the time step can be assumed equivalent to an azimuth increase, Δθ, of 0.1 deg. with the maximum iterations of 20 per time step, therefore, exactly such time step settings were adopted in these simulations.…”

“…This work is not dedicated to investigating the impact of the turbulence model on solving the problem of transient flow around a working rotor. Current simulations use a model widely used in this kind of flows (Rogowski, 2019), the Shear-Stress Transport (SST) k-ω model.…”

“…5 consists of 234 840 elements and 136 059 nodes. The computational grid with global settings described above was used also by Rogowski (2019). This author shoved that these mesh settings provided sufficiently accurate results of the velocity fields and aerodynamic blade loads.…”