Unstructured, Reynolds Averaged Navier-Stokes Model of the NREL Phase VI Wind Turbine

A Thesis Presented for the Master of Science in Computational Engineering Degree, The University of Tennessee at Chattanooga

Justin Whitt, April 2009

As early as the 9th century A.D., humans began extracting mechanical energy from the wind. Today–faced with a growing need for clean, sustainable forms of energy–modern societies are again looking to the wind. Recently, the Department of Energy’s study, “20% Wind Energy by 2030,” outlined the milestones, benefits, and challenges to the use of wind for large-scale electricity production. The report identifies impediments to wind technology advancement such as: the high cost of aerodynamic testing and a lack of adequately sized experimental test facilities coupled with a lack of accurate, 3-D, computational design tools.

The development of robust, high fidelity simulations of wind turbines is a cost effective approach to shortening the design cycle and improving the overall production capacity of the U. S. wind fleet. To this end, a steady-state computational model of a National Renewable Energies Lab (NREL) prototype wind turbine has been developed and validated using an unstructured, Reynolds Averaged flow solver developed by the University of Tennessee at Chattanooga SimCenter: National Center for Computational Engineering. The model predictions for blade surface pressures are compared to those measured experimentally at NASA Ames. These predictions alongside flow field visualizations are presented as a tool to gain insight into the associated flow phenomena.

Justin Whitt Masters Thesis