Project Team
Students
Joshua Forrest
Aerospace Engineering
Penn State University Park
Faculty Mentors
Dr. Mark Miller
Penn State University Park
Aerospace Engineering
Project
Project Video
Project Abstract
Testing Horizontal Axis Wind Turbine (HAWT) designs in wind tunnels is timely and cost-effective for experimentally improving and verifying aerodynamic designs before implementation. Field experiments are complicated and time-consuming, begging a need for controlled testing of rotor performance and structural response, particularly for yawed flow conditions. However, matching all three experimental aerodynamic non-dimensional parameters: Reynolds number, tip speed ratio, and Mach number, for the model scale matching the geometry of a full-scale wind turbine, has proved challenging in standard wind tunnels. Previously, researchers have matched Mach number and tip speed ratio but sacrificed Reynolds number parity in wind tunnel experiments. Mach number and tip speed ratio are essential for comparable rotor performance. Reynolds number effects have significant impacts locally on the performance of the blades, particularly for the thicker airfoils used for wind turbines. Increasing the air pressure of the wind tunnel to 500 PSI allows for more suitable Reynolds number matching. Experiments at the Princeton University High Reynolds number Test Facility (HRTF) have demonstrated Reynolds number insensitivity for rotor diameter Reynolds numbers of 2×106 to 16×106, although the overall poor performance of the model rotor. Reaching the threshold of Reynolds number insensitivity would increase the value of wind tunnel data. This research set out to design a stiff model rotor for high-pressure wind tunnel testing that reaches Reynolds number insensitivity and also has comparable rotor performance to the NREL 5MW Reference Wind Turbine using 2D airfoil inputs from XFOIL in XTURB. Stall delay, boundary layer tripping, chord adjustment, and twist adjustment for matching non-dimensional local blade airfoil circulation were explored. The stall delay feature in XTURB simulations and setting tripped boundary layers in XFOIL did not lead to the desired boundary layer effects expected. Matching non-dimensional local circulation along the blade did lead to improved thrust and power coefficient agreement between the model and full-scale simulations. The NREL 5MW rotor uses thick inboard airfoils created by the Delft University of Technology that have limited or no available 2D airfoil data. Experimental validation is necessary for XFOIL airfoil polars used for XTURB simulations. Future work will be to create a model rotor using scaling techniques and findings derived from this paper using airfoils in possession of Penn State to use experimentally obtained airfoil polars. The development of an aerodynamically scaled wind turbine model would allow for wind tunnel experiments with results directly applicable to full-scale HAWT development and design.
Evaluate this Project
Use this form link to provide feedback to the presenters, and add your project evaluation for award(s) consideration.