Wind farm production frequently falls short of expectations. Poor forecasts of low-altitude winds, suboptimal wind plant design and operation, and higher-than-expected downtimes and maintenance costs all undermine project profitability. Each of these issues results from industry reliance on insufficient computational approaches for simulating wind farm inflow and plant behavior under realistic operating conditions. These tools largely neglect the influence of changing weather conditions and other environmental impacts, such as terrain characteristics, on simulated flows at turbine heights.

To address these shortcomings, the U.S. Department of Energy’s Lawrence Livermore National Laboratory (LLNL) is creating a multiscale atmosphere/wind farm simulation toolkit that incorporates the full spectrum of weather, environmental, and turbine impacts to better predict wind farm flows. This toolkit will address industry needs, including resource characterization, inflow prediction, siting, design, and operations, and will apply to challenging environments with complex flows and terrain.

The toolkit is being implemented within the Weather Research and Forecasting (WRF) model, a widely used, publicly available atmospheric modeling framework that supports simulations ranging from global weather to large eddies of turbulent flows. Development of the toolkit within the popular WRF model enables easy collaboration among industry, academia, and Energy Department laboratories. It also provides a direct patch for incorporating future capabilities and improvements contributed by the worldwide community of WRF users and developers.

Toolkit Features

Image

Accurate multiscale simulation requires proper physics parameterizations, numerical methods, and model configurations. The LLNL simulation toolkit addresses these requirements by providing:

  • Actuator models for wind turbine and farm simulation
  • Advanced turbulence subgrid models for better large-eddy simulations
  • Algorithms to improve downscaling, both from mesoscale to large-eddy simulations and within large-eddy simulations
  • Ensemble-based forecasting to provide uncertainty bounds
  • Vertical mesh refinement for multiscale simulation
  • Immersed boundary methods for improved simulations over complex terrain.

Simulating Weather-Turbine Interactions

The toolkit’s turbine models are embedded in a large-eddy simulation, which resolves turbine/turbulence interactions. Next, the simulation is nested within larger-scale simulations that model weather effects, therefore capturing atmospheric and environmental characteristics, such as weather features, that impact both the plant inflow and turbine response.

The multiscale simulation below shows a weather front just before (left) and after (right) passing through an array of actuator turbine models using LLNL’s WRF-based wind farm simulation toolkit. Color contours show wind speed, with the darkest colors indicating the lowest values. Turbine wakes appear as pairs of dark blue streaks extending downstream from the actuator models. Arrows show wind speed and direction. Note the change in wind direction and impacts on turbine wakes during the frontal passage.

Image

Obtaining accurate simulations of flow over complex terrain is a continuing challenge for the wind energy industry. The immersed boundary method eliminates numerical errors caused by grid distortion in simulations over complex terrain. In combination with vertical grid nesting, the immersed boundary method extends LLNL’s advanced multiscale simulation capability to flows over arbitrarily complex terrain, providing more accurate wind simulations for the challenging environments in which many wind farms operate.

These efforts support projects within the Energy Department A2e initiative. More information is available at the LLNL wind energy website.