The U.S. Department of Energy’s (DOE’s) Atmosphere to Electrons (A2e) initiative has identified the evolution of wakes in turbulent inflow as a key physical process affecting power production and turbine loads in wind farms. Wind energy researchers know that wakes shed from upstream wind turbines lead to reduced power production, increased loading, and higher maintenance costs on downstream turbines. The result is a higher cost of energy.
Although the impacts of wakes on wind energy economics is well documented from operating wind farms, potential solutions remain highly debated in the industry. Wind turbines are now large enough that researchers must rely on models using powerful computing resources to simulate potential solutions. High-quality field data are needed to determine the adequacy of current models and to invest in the development of future models. DOE’s A2e initiative is embarking on a tightly-coupled experiment and modeling-simulation campaign to rigorously address this challenge.
DOE and Sandia National Laboratories’ (SNL’s) Scaled Wind Farm Technology (SWiFT) facility in Lubbock, Texas, is ideally suited for a large portion of these experiments. The site contains a carefully spaced array of three highly instrumented, research-scale wind turbines along with meteorological towers to measure and record mesoscale weather around the turbine array as well as inflow directly into each turbine.
To support specific physical objectives and ambitious measurement objectives of this integrated test campaign, SNL is developing a new subscale rotor, the National Rotor Testbed (NRT), for the SWiFT turbines. Features of the new subscale rotor (27-meter diameter) are determined by scaling relevant aerodynamic performance parameters and design drivers from a representative megawatt-scale rotor (70-meter diameter and larger). The subscale rotor will be well-suited to support turbine-to-turbine interaction research at the SWiFT facility and will also represent full-sized turbines in the United States. This physical relevance is especially important for the coupled experimental and modeling-simulation campaign because it ensures that the credibility of numerical models is demonstrated within physical regimes that are directly relevant to full-scale industry applications.
The NRT team has performed preliminary numerical simulations to define the scope and configuration of future tests. Numerical simulations show how rotor design may affect wind turbine wake behavior in terms of wake recovery. The team is using medium-fidelity vortex and high-fidelity computational fluid dynamics methods to understand the potential tradeoffs of different rotor designs in a wind plant scenario. The codes are also used to inform the future test campaign at the SWiFT facility, which will in turn be used to determine credibility of the codes.
Preliminary numerical simulations have shown the potential of certain rotor designs to offer improved wake recovery at wind farms by deviating from typical, “high-efficiency” wind turbine rotor design practices. As an example, the SNL research team has demonstrated two conceptual rotor designs. The figures below show the wake velocity deficit for the two designs at various diameters downstream of the rotor; the wake deficits were predicted by a relaxed-wake vortex method using steady, uniform inflow.
The first approach, referred to as maximum coefficient of power (CP) design, produces a maximum CP via a theoretical optimum distribution of axial induction. Axial induction is how much the incoming wind slows as it meets the wind turbine’s rotor and energy is extracted. Too much or too little slowing causes the wind to pass around or through the rotor disc without extracting optimal energy. The second approach—lower tip loading design—exhibits a modified axial induction distribution while still operating at the same rotor thrust coefficient as the first blade and only a slightly lower rotor performance coefficient. This design shows potential for faster wake recovery, which enables closer turbine spacing and potentially higher wind energy capture from a given amount of land area.