A typical approach to optimize wind turbine blades separates the airfoil shape design from the blade planform design. This approach is sequential, where the airfoils along the blade span are pre-selected or optimized and then held constant during the blade planform optimization. In contrast, integrated blade design optimizes the airfoils and the blade planform concurrently and thereby has the potential to reduce cost of energy (COE) more than sequential design. Nevertheless, sequential design is commonly performed because of the ease of precomputation, or the ability to compute the airfoil analyses prior to the blade optimization. This research investigates two integrated blade design approaches, the precomputational and free-form methods, that are compared to sequential blade design. The first approach is called the precomputational method because it maintains the ability to precompute, similar to sequential design, and allows for partially flexible airfoil shapes. This method compares three airfoil analysis methods: a panel method (XFOIL), a Reynolds-averaged Navier-Stokes computational fluid dynamics method (RANS CFD), and using wind tunnel data. For each airfoil analysis method, there are two airfoil parameterization methods: the airfoil thickness-to-chord ratio and blended airfoil family factor. The second approach is called the free-form method because it allows for fully flexible airfoil shapes, but no longer has the ease of precomputation as the airfoil analyses are performed during the blade optimization. This method compares XFOIL and RANS CFD using the class-shape-transformation (CST) method to parameterize the airfoil shapes. This study determines if the precomputational method can capture the majority of the benefit from integrated design or if there is a significant additional benefit from the free-form method. Optimizing the NREL 5-MW reference turbine shows that integrated design reduce COE significantly more than sequential design. The precomputational method improved COE more than sequential design by 1.6%, 2.8%, and 0.7% using the airfoil thickness-to-chord ratio, and by 2.2%, 3.3%, and 1.4% using the blended airfoil family factor when using XFOIL, RANS CFD, and wind tunnel data, respectively. The free-form method improved COE more than sequential design by 2.7% and 4.0% using the CST method with XFOIL and RANS CFD, respectively. The additional flexibility in airfoil shape reduced COE primarily through an increase in annual energy production. The precomputational method captures the majority of the benefit of integrated design (about 80%) for minimal additional computational cost and complexity, but the free-form method provides modest additional benefits if the extra effort is made in computational cost and development time.
Identifer | oai:union.ndltd.org:BGMYU2/oai:scholarsarchive.byu.edu:etd-7673 |
Date | 01 January 2018 |
Creators | Barrett, Ryan Timothy |
Publisher | BYU ScholarsArchive |
Source Sets | Brigham Young University |
Detected Language | English |
Type | text |
Format | application/pdf |
Source | Theses and Dissertations |
Rights | http://lib.byu.edu/about/copyright/ |
Page generated in 0.0014 seconds