Lidar Applications For Wind Energy Studies

The overarching goal of the Wind Forecasting Improvement Project (WFIP) is to develop more immediate improvements in short-term forecasting accuracy. This is being achieved by using additional observations from sophisticated instrument platforms that are being assimilated by a suite of numerical weather prediction models with the ultimate objective of reducing power production uncertainties. From the resource assessment and energy production perspective, however, the very presence of wind turbines leaves the air downstream (i.e., the wake) with reduced speed and static pressure as well as higher turbulence. This phenomenon is the source of significant energy production losses in wind power plants. Typically wind farms experience wake losses of 2 to 12% of their annual gross power production. Losses can be higher depending on the wind turbine layout, turbine characteristics, and the local climate, i.e. wind, turbulence and atmospheric stability conditions. Thus, developers, operators, and utilities are presented with a variety of issues related to the siting, operation, and the very placement of wind turbines--problems that suffer from a distinct lack of onsite field data necessary for accurate forecasts and power production estimates.

To explore the efficacy of targeted forecasting and the structure of observed wake effects, we have deployed additional instrumentation at an WFIP participating wind farm, specifically two NRG/Leosphere LiDARs (LiDAR, Light Detection And Ranging). One LiDAR is located upwind (the Windcube8 500m system) and one LiDAR (Windcube7 200m system), downstream. Before deployment of these LiDARs, few observations from operational wind farms existed for validating wake parameterizations; thus, these schemes remain relatively untested for non-idealized simulations.

More specifically, results from three approaches are presented:

1) the wind farm parameterization in WRF-ARW will be tested and verified against the upstream and downstream LiDARs for select case studies. This parameterization represents the effects of wind turbines on the atmosphere by imposing an elevated momentum sink and turbulent kinetic energy (TKE) source on the mean flow;

2) the Deep-Array Wake Model (DAWM) recently developed by AWS Truepower, and

3) AWS Truepower's large-eddy simulation model based on the Advanced Regional Prediction System (ARPS).