The flow around the blade of a horizontal axis wind turbine wind tunnel model, operating at its optimal tip speed ratio in axial flow, has been investigated by means of stereoscopic particle image velocimetry (SPIV). The aim was to assess the possibility of measuring the loads impinged on the blade, using the acquired velocity field and its spatial derivatives. Thus, the three-dimensional (3D) velocity field and the pressure distribution surrounding the blade, as well as the aerodynamic loads, responsible for thrust and torque, were obtained with a non-intrusive method. The SPIV equipment was mounted on a traverse system and provided with phase-locked velocity planes perpendicular to the blade axis, scanning the blade from the root to the tip, at a fixed azimuthal position. The obtained velocity fields were used to estimate the pressure distribution and the aerodynamic loads on each plane, using a 3D formulation. Finally, these results were compared and discussed with similar results obtained computationally with a panel method model, showing good consistency. In the future, the proposed methodology could be used to study relevant topics such as active load control techniques, rotational augmentation, dynamic stall phenomena and the aerodynamics of small wind turbines operating at low Re numbers, among others. The processed and averaged flow fields from the experimental SPIV data are made available online to the reader. See appendix for description of the files.Peer ReviewedPostprint (published version

Stereoscopic particle image velocimetry (SPIV) has been used to characterize the flow around the blade of a yawed horizontal axis wind turbine model. The goal was to assess the possibility of obtaining the 3D velocity field around the blade, the pressure distribution, and the aerodynamic loads being exerted on the blade under unsteady flow conditions. The SPIV equipment was mounted on a traverse system and provided with phase‐locked velocity planes perpendicular to the blade axis, scanning the blade from the root to the tip, at three different azimuthal positions, so as to have information of the time variation of the flow. The pressure distribution and the aerodynamic loads on each plane were obtained via 3D formulation. Main differences encountered when measuring loads in yawed flow compared with axial flow have been discussed. Finally, the consistency of the results with similar results obtained computationally with a panel model was assessed. The proposed methodology presents one step further in the application of SPIV to measure forces on a horizontal axis wind turbine, assessing the possibility of estimating the blade loads when the turbine is operating under non‐axial flow conditions, with the goal of better simulating real working operating regimes in a wind farm, where the flow is typically not uniform. The proposed methodology could be developed and used to better understand relevant wind energy issues such as dynamic loading and active load control efficiency, in the future. The processed and averaged flow fields from the experimental SPIV data are made available online to the reader. See appendix for description of the files. Copyright © 2014 John Wiley & Sons, Ltd.

Lifting line vortex models have been widely used to predict flow fields around wind turbine rotors. Such models are known to be deficient in modelling flow fields close to the blades due to the assumption that blade vorticity is concentrated on a line and consequently the influences of blade geometry are not well captured. The present study thoroughly assessed the errors arising from this approximation by prescribing the bound circulation as a boundary condition on the flow using a lifting line free-wake vortex approach. The bound circulation prescribed to free-wake vortex model was calculated from two independent sources using (1) experimental results from SPIV and (2) data generated from a 3D panel free-wake vortex approach, where the blade geometry is fully modelled. The axial and tangential flow fields around the blades from the lifting line vortex model were then compared with those directly produced by SPIV and the 3D panel model. The comparison was carried out for different radial locations across the blade span. The study revealed the cumulative probability error distributions in lifting line model estimations for the local aerofoil flow field under both 3D rotating and 2D non-rotating conditions. It was found that the errors in a 3D rotating environment are considerably larger than those for a wing of infinite span in 2D flow. Finally, a method based on the Cassini ovals theory is presented for defining regions around rotating blades for which the lifting line model is unreliable for estimating the flow fields.

No abstract

In this study, low Reynolds number wind turbine aerodynamics was considered. The overall goal was to characterize the flow in order to optimize the power output of the system. First, BEMT theory (Blade Element Momentum Theory) was formulated for this flow where Prandtl’s tip- and hub-loss corrections were included, as well as Glauert’s thrust coefficient correction. The theory was validated with experimental data from National Renewable Energy Laboratory (NREL) for larger scale wind turbines. Also, a physical model of a low Reynolds number horizontal-axis wind turbine (HAWT) was built. Particle Image Velocimetry (PIV) was used to calculate the velocity field around the HAWT. This allowed for planar measurements of the velocity field at different location in the wake of the rotor. The measurements were performed in a water channel allowing for better control of PIV seeding and improved flow visualization. PIV results allowed observation of the velocity field and vorticity field in the wake of the rotor. This data is currently being compared to BEMT theory suggesting good agreement.