• Energy Research

Abstract. Offshore wind farms are more commonly installed in wind farm clusters, where wind farm interaction can lead to energy losses; hence, there is a need for numerical models that can properly simulate wind farm interaction. This work proposes a Reynolds-averaged Navier–Stokes (RANS) method to efficiently simulate the effect of neighboring wind farms on wind farm power and annual energy production. First, a novel steady-state atmospheric inflow is proposed and tested for the application of RANS simulations of large wind farms. Second, a RANS-based wind farm parameterization is introduced, the actuator wind farm (AWF) model, which represents the wind farm as a forest canopy and allows to use of coarser grids compared to modeling all wind turbines as actuator disks (ADs). When the horizontal resolution of the RANS-AWF model is increased, the model results approach the results of the RANS-AD model. A double wind farm case is simulated with RANS to show that replacing an upstream wind farm with an AWF model only causes a deviation of less than 1 % in terms of the wind farm power of the downstream wind farm. Most importantly, a reduction in CPU hours of 75.1 % is achieved, provided that the AWF inputs are known, namely, wind farm thrust and power coefficients. The reduction in CPU hours is further reduced when all wind farms are represented by AWF models, namely, 92.3 % and 99.9 % for the double wind farm case and for a wind farm cluster case consisting of three wind farms, respectively. If the wind farm thrust and power coefficient inputs are derived from RANS-AD simulations, then the CPU time reduction is still 82.7 % for the wind farm cluster case. For the double wind farm case, the RANS models predict different wind speed flow fields compared to output from simulations performed with the mesoscale Weather Research and Forecasting model, but the models are in agreement with the inflow wind speed of the downstream wind farm. The RANS-AD-AWF model is also validated with measurements in terms of wind farm wake shape; the model captures the trend of the measurements for a wide range of wind directions, although the measurements indicate more pronounced wind farm wake shapes for certain wind directions.

Abstract In wind resource assessments, which are critical to the pre-construction of wind farms, measurements by LiDARs or masts are a source of high-fidelity data, but are expensive and scarce in space and time, particularly for offshore sites. On the other hand, numerical simulations, using for example the Weather Research and Forecasting (WRF) model, generate temporally and spatially continuous data with relatively low-fidelity. A hybrid approach is proposed here to combine the merit of measurements and simulations for the assessment of offshore wind. Firstly a temporal data fusion using deep Multi Fidelity Gaussian Process Regression (MF-GPR) is performed to combine the intermittent measurement and the continuous simulation data at an onshore location. Then a spatial data fusion using a neural network with Non-linear Autoregression (NAR) and Non-linear Autoregression with external input (NARX) are conducted to project the wind from onshore to offshore. The numerical and measured wind speeds along the west coast of Denmark were used to evaluate the method. We show that the proposed data fusion technique using a gappy onshore measurement results in accurate offshore wind resource assessment within a 2% margin error.

High-quality tall mast and wind lidar measurements over the North and Baltic Seas are used to validate the wind climatology produced from winds simulated by the Weather, Research and Forecasting (WRF) model in analysis mode. Biases in annual mean wind speed between model and observations at heights around 100 m are smaller than 3.2% at offshore sites, except for those that are affected by the wake of a wind farm or the coastline. These biases are smaller than those obtained by using winds directly from the reanalysis. We study the sensitivity of the WRF-simulated wind climatology to various model setup parameters. The results of the year-long sensitivity simulations show that the long-term mean wind speed simulated by the WRF model offshore in the region studied is quite insensitive to the global reanalysis, the number of vertical levels, and the horizontal resolution of the sea surface temperature used as lower boundary conditions. Also, the strength and form (grid vs spectral) of the nudging is quite irrelevant for the mean wind speed at 100 m. Large sensitivity is found to the choice of boundary layer parametrization, and to the length of the period that is discarded as spin-up to produce a wind climatology. It is found that the spin-up period for the boundary layer winds is likely larger than 12 h over land and could affect the wind climatology for points offshore for quite a distance downstream from the coast.

We present how a nacelle mounted scanning lidar, the so-called SpinnerLidar, can be used to derive all six terms of the Reynolds stress tensor in front of a wind turbine, essentially without bias. 

AbstractThe atmospheric flow in the coastal zone is investigated using lidar and mast measurements and model simulations. Novel dual‐Doppler scanning lidars were used to investigate the flow over a 7 km transect across the coast, and vertically profiling lidars were used to study the vertical wind profile at offshore and onshore positions. The Weather, Research and Forecasting model is set up in 12 different configurations using 2 planetary boundary layer schemes, 3 horizontal grid spacings and varied sources of land use, and initial and lower boundary conditions. All model simulations describe the observed mean wind profile well at different onshore and offshore locations from the surface up to 500 m. The simulated mean horizontal wind speed gradient across the shoreline is close to that observed, although all simulations show wind speeds that are slightly higher than those observed. Inland at the lowest observed height, the model has the largest deviations compared to the observations. Taylor diagrams show that using ERA‐Interim data as boundary conditions improves the model skill scores. Simulations with 0.5 and 1 km horizontal grid spacing show poorer model performance compared to those with a 2 km spacing, partially because smaller resolved wave lengths degrade standard error metrics. Modeled and observed velocity spectra were compared and showed that simulations with the finest horizontal grid spacing resolved more high‐frequency atmospheric motion.

Accurately quantifying wind turbine wakes is a key aspect of wind farm economics in large wind farms. This paper introduces a new simulation post-processing method to address the wind direction uncertainty present in the measurements of the Horns Rev offshore wind farm. This new technique replaces the traditional simulations performed with the 10 min average wind direction by a weighted average of several simulations covering a wide span of directions. The weights are based on a normal distribution to account for the uncertainty from the yaw misalignment of the reference turbine, the spatial variability of the wind direction inside the wind farm and the variability of the wind direction within the averaging period. The results show that the technique corrects the predictions of the models when the simulations and data are averaged over narrow wind direction sectors. In addition, the agreement of the shape of the power deficit in a single wake situation is improved. The robustness of the method is verified using the Jensen model, the Larsen model and Fuga, which are three different engineering wake models. The results indicate that the discrepancies between the traditional numerical simulations and power production data for narrow wind direction sectors are not caused by an inherent inaccuracy of the current wake models, but rather by the large wind direction uncertainty included in the dataset. The technique can potentially improve wind farm control algorithms and layout optimization because both applications require accurate wake predictions for narrow wind direction sectors. © 2013 The Authors. Wind Energy published by John Wiley & Sons, Ltd.

Abstract. The design of wind turbines and wind farms can be improved by increasing the accuracy of the inflow models representing the atmospheric boundary layer. In this work we employ one-dimensional Reynolds-averaged Navier–Stokes (RANS) simulations of the idealized atmospheric boundary layer (ABL), using turbulence closures with a length-scale limiter. These models can represent the mean effects of surface roughness, Coriolis force, limited ABL depth, and neutral and stable atmospheric conditions using four input parameters: the roughness length, the Coriolis parameter, a maximum turbulence length, and the geostrophic wind speed. We find a new model-based Rossby similarity, which reduces the four input parameters to two Rossby numbers with different length scales. In addition, we extend the limited-length-scale turbulence models to treat the mean effect of unstable stratification in steady-state simulations. The original and extended turbulence models are compared with historical measurements of meteorological quantities and profiles of the atmospheric boundary layer for different atmospheric stabilities.

Abstract. Blockage effects due to the interaction of five wind turbines in a row are investigated through both Reynolds-averaged Navier-Stokes simulations and site measurements. Since power performance tests are often carried out at sites consisting of several turbines in a row, the objective of this study is to evaluate whether the power performance of the five turbines differs from that of an isolated turbine. A number of simulations are performed, in which we vary the turbine inter-spacing (1.8, 2 and 3 rotor diameters) and the inflow angle between the incoming wind and the orthogonal line to the row (from 0° to 45°). Different values of the free-stream velocity are considered to cover a broad wind speed range of the power curve. Numerical results show consistent power deviations for all the five turbines when compared to the isolated case. The amplitude of these deviations depends on the location of the turbine within the row, the inflow angle, the inter-spacing and the power curve region of operation. We show that the power variations do not cancel out when averaging over a large inflow sector (from −45° to +45°) and find an increase in the power output of up to +1 % when compared to the isolated case. We simulate power performance ‘measurements’ with both a virtual mast and nacelle-mounted lidar and find a combination of power output increase and upstream velocity reduction, which causes an increase of +4 % of the power coefficient. We also use measurements from a real site consisting of a row of five wind turbines to validate the numerical results. From the analysis of the measurements, we also show that the power performance is impacted by the neighboring turbines. Compared to when the inflow is perpendicular to the row, the power output varies of +1.8 % and −1.8 % when the turbine is the most downwind and upwind of the line, respectively.