Using pod to examine the unsteady wake of an aeroelastic wind energy converter

Both the steady and unsteady wakes of a torsional Aeroelastic Wind Energy Converter are examined. The converter consists of an H-section prism with a pendulum to increase the mass and lower the natural frequency of the system. To provide an understanding of the wake characteristics of the device, a Particle Image Velocimetry (PIV) system was employed to measure the mean and fluctuation velocity field from which various single point statistics were calculated. For the steady case, wake profiles were obtained at angles of attack from -30 ° to +30 ° (at 5 ° intervals) with the H-section fixed at each angle. In the unsteady case, the H-section was free to vibrate due to the flow structure-interaction and a torsional vibration was produced with angles of attack from -30 ° to +30 °. A comparison between the turbulent wake profiles from the steady and unsteady cases was performed to provide insight on the differences between the various single point turbulent velocity statistics. It can be seen that overall the dynamic wake deficit is less than the static wake deficit. On average the dynamic wake deficit is 20% of Uoo and is dispersed across the entire area; whereas, the static wake is concentrated around the trailing edge side of the converter. Through this methodology, the analysis for flow-structural interaction problems can be better understood, and this data can be used as a turbulence calibration model. By applying POD to the spatial data of the aeroelastic wind energy covnverter's wake, it can be seen that only six modes are required to reconstruct an original snapshot of the flow.