The SWiFT Benchmarks: Wake Evolution and Dynamics
This heavily instrumented site in the U.S. Great Plains provides the opportunity to validate simulations of wind-turbine wakes without the influence of complex terrain and weather.
Participate!
The benchmark is open to all participants of the International Energy Agency (IEA) Wind Task 31: “WakeBench” (benchmarks of wake models). The SWiFT benchmarks are part of Phase 3 of WakeBench, running between 2018 and 2019. To sign up, fill out this form. The benchmark kick-off will take place via a webinar on October 30, 2018. Contact the organizers to participate.
What is WakeBench?
The WakeBench task seeks to quantitatively assess the performance of wind-turbine wake models. This is done by quantitatively comparing the simulation results obtained with various codes to field measurements. The methodology is defined a priori and well documented, and all code used to process the data is publicly released so that results can be replicated by any participants. The same processing methodology is applied consistently to all models, and the end result of the exercise will be a journal publication in which the various codes are validated against measurements. This an iterative process whereby models will be improved as the comparisons proceed. All participants will have access to wake measurements before they are publicly available.
What is SWiFT?
The Scaled Wind Farm Technology (SWiFT) facility is a ground for research and technology innovation in the field of wind-turbine and wind-plant aerodynamics. The facility is funded by the United States (U.S.) Department of Energy (DoE), and located at Texas Tech University’s National Wind Institute Research Center in Lubbock, Texas. All data collected at SWiFT through the wake-steering experiment will be freely available through the DoE Data Archive Portal (DAP).
Atmospheric Conditions at SWiFT
SWiFT is located in the U.S. Great Plains and is therefore exempt from complex, terrain-induced flow patterns. In the absence of weather phenomena (e.g., fronts, storms) the atmospheric conditions at SWiFT approximate canonical diurnal cycles. The relative simplicity of atmospheric conditions at the site make it a valuable resource for research in complex turbulent flows such as wind-turbine wakes. In other words, the mean and dynamics characteristics of wakes and their effect on downstream turbines can be considered without the influence of complex terrain and weather. For more detail on the atmospheric conditions at this site, see Kelley and Ennis (2016).
Wind Turbines at SWiFT
The facility hosts 3 variable-speed, variable-pitch, modified Vestas V27 wind turbine generators (WTGs). These are smaller than utility-scale WTGs (i.e., rotor diameter is 27 m and hub height is 32.1 m), facilitating the deployment of instrumentation and reducing the cost of conducting experiments and testing new technology.
Instrumentation at SWiFT
The largest wind-energy experiment conducted at SWiFT to date is the “wake-steering experiment”. In 2016 and 2017, the upstream WTG was intentionally yawed into and out of the wind in order to steer the propagation direction of its wake. This deliberate control of the wake has a direct effect on the power and loads of both WTGs. The data used to define this benchmark were collected during this experiment. A meteorological tower collected freestream atmospheric measurements and a rear-facing, nacelle-mounted lidar mesured the wake. In addition, power and loads measurements were collected at the wake-producing WTG and the wake-bearing WTG.
The Benchmarks
Researchers at the National Renewable Energy Laboratory (NREL) and Sandia National Laboratories (SNL) have worked together to define three benchmarks for validation of wind-turbine wake simulations:
- Wake Evolution in a Nearly Neutral Atmosphere
- Wake Evolution in a Stable Atmosphere
- Wake Dynamics in an Unstable Atmosphere
Participants are encouraged to engage in all benchmarks in order to investigate the performance of their simulation tool under different stratification scenarios. Detailed information about each benchmark was released on October 30, 2018 and slides are available here.
Status
Each benchmark has 3 phases. Each phase will be carried out concurrently for the three benchmarks.
Phase 1, Code Calibration: The objective of this phase is to provide an opportunity for model calibration before the blind comparison.
Phase 2, Blind Comparison: This phase has two parts. During the first part, a code-to-code comparison is made (without considering measurements) to identify potential errors in model configuration. During the second part, codes are compared to measurements to provide a first assessment of the skill of the simulation tool.
Phase 3, Iteration: The objective of this phase is to interrogate results to understand where models work well and where improvements are required.
Points-of-Contact
Paula Doubrawa, Patrick Moriarty
References
Christopher Lee Kelley and Brandon Lee Ennis, “SWiFT Site Atmospheric Characterization” (Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States), 2016), https://doi.org/10.2172/1237403.
