No abstract

Remote wind sensing technologies, such as lidar, are becoming more and more mature and the wind energy industry is rapidly adopting them for a variety of purposes. One of these use cases is utilizing lidar measurements from a nacelle mounted device in order to improve the accuracy of load simulations by creating more realistic synthetic wind inputs. In this work we present an open source numerical framework, called ViConDAR for "Virtual Constrained turbulence and liDAR measurements", used for simulating lidar measurements and applying them as constraints in synthetic wind field generation. A realistic lidar simulator is used to obtain the virtual lidar measurements by scanning a synthetic wind field. These measurements are fed to open source constrained turbulence generation codes (TurbSim and PyConTurb), coupled to ViConDAR. The resulting constrained wind fields are compared to the original ones in order to quantify the level of convergence and can be used directly as inputs to aeroelastic simulations. Finally, two indicative applications of this framework are shown. First, a sensitivity analysis of the lidar parameters versus varying atmospheric conditions is carried out to investigate the potential of the lidar measurements to capture the wind field properties. Secondly, a sensitivity analysis is presented on the influence of different lidar parameters on the convergence of the full wind fields comparing both turbulence generation codes under varying atmospheric conditions.

Abstract. This study proposes two methodologies for improving the accuracy of wind turbine load assessment under wake conditions by combining nacelle-mounted lidar measurements with wake wind field reconstruction techniques. The first approach consists of incorporating wind measurements of the wake flow field, obtained from nacelle lidars, into random, homogeneous Gaussian turbulence fields generated using the Mann spectral tensor model. The second approach imposes wake deficit time series, which are derived by fitting a bivariate Gaussian shape function to lidar observations of the wake field, on the Mann turbulence fields. The two approaches are numerically evaluated using a virtual lidar simulator, which scans the wake flow fields generated with the dynamic wake meandering (DWM) model, i.e., the target fields. The lidar-reconstructed wake fields are then input into aeroelastic simulations of the DTU 10 MW wind turbine for carrying out the load validation analysis. The power and load time series, predicted with lidar-reconstructed fields, exhibit a high correlation with the corresponding target simulations, thus reducing the statistical uncertainty (realization-to-realization) inherent to engineering wake models such as the DWM model. We quantify a reduction in power and loads' statistical uncertainty by a factor of between 1.2 and 5, depending on the wind turbine component, when using lidar-reconstructed fields compared to the DWM model results. Finally, we show that the number of lidar-scanned points in the inflow and the size of the lidar probe volume are critical aspects for the accuracy of the reconstructed wake fields, power, and load predictions.

No abstract

In recent years wind turbine down-regulation has been used or investigated for a variety of applications such as wind farm power optimisation, energy production curtailment and lifetime management. This study presents results from measurement data of tower loads and power obtained from two turbines located in the German offshore wind farm alpha ventus. The free streaming turbine, located closely to a fully equipped meteorological mast, was down-regulated to 50% for a period of 8 months, while the downwind turbine was operating normally. The results are compared to periods where both turbines were operated in normal conditions. Changes in loads and power are analysed according to incoming wind direction and magnitude. Results show a high reduction in the loads of the down regulated turbine, up to a level of 40%. For the turbine in wake the effects in loads are more prominent, showing a maximum reduction of 30%, compared to the effects in power and are seen in a wider sector of about 20° for loads and 10° for power.