Degree Type


Date of Award


Degree Name

Doctor of Philosophy


Aerospace Engineering


Aerospace Engineering

First Advisor

Hui Hu


The wake characteristics behind a Horizontal Axis Wind Turbines (HAWT) sited in Atmospheric Boundary Layer (ABL) winds are investigated by using analytical, experimental and computational techniques. Studies of wake effects are crucial to understand the power generation and wind farm siting. Analytical wake models are investigated and incorporated into an algorithm which accurately predicts the thrust coefficient of wind turbine and flow characteristics in the far wake regions (X/D >1.0). Instantaneous flow field measurements obtained by using Particle Image Velocimetry (PIV) technique were analyzed using a suit of principal component analysis techniques, such as Proper Orthogonal Decomposition (POD) and Dynamic Mode Decomposition (DMD) to identify the dominant flow structures in the wind turbine wakes. The analysis was focused on the instabilities of the vortex structures, break-up phenomenon, formation of the shear layer and the momentum transport mechanisms. The analysis shows that, in addition to the existence of the well-known tip vortex structures, the vortices generated at the blade mid-span were found to have the twice the circulation of the tip vortex filament. The wake vortices were found to break-up in the downstream region of X/D = 0.6, forming the shear layer. Analysis of momentum transport also shows how the presence of discreet vortex filaments in the near wake hinders the momentum transport by 40%. The effects of this instability involving mid-span and tip vortex filaments are simulated using an unsteady free-wake method, where the filaments are shown to have a mutual attraction which results in a short wave instability. An experimental study was also conducted to compare the aeromechanic performances (i.e., dynamic wind loads and power generation characteristics) of scaled HAWTs in upwind and downwind configurations. A novel approach to measuring torque was introduced to alleviate previous under-predictions that stem from electric power measurements.



Copyright Owner

Pavithra Dhanuka Premaratne



File Format


File Size

128 pages