The ECMWF Reanalysis v5 (ERA5), the New European Wind Atlas (NEWA), and the 3 km Norwegian Reanalysis (NORA3) are reference datasets that are available for industry and research. The resolution of 3km in both the NORA3 and NEWA datasets sets them apart, while ERA5, with its 31km resolution, continues to serve as a reliable data source that is widely used in the industry. The study offers a thorough analysis of three datasets from three research stations in Sweden, which are accessible through the Integrated Carbon Observation System (ICOS). It has been discovered that all three reference datasets exhibit a strong alignment with the measured data. However, NORA3 and ERA5 outperform NEWA in wind speed and direction estimation. The computation of Annual Energy Production (AEP) using WindPro is performed. We examine the representativeness of the correlation coefficient between the Weibull scale and shape parameters, the agreement of wind rose distributions, and the estimated AEP.

The New European Wind Atlas (NEWA) was developed with an aim to provide high accuracy wind climate data for the region of EU and Turkey. Wind industry always seek for solid performance in wind resource assessment, and it is highly affected by the quality of modelled data. The aim of this study is to validate the newly developed wind atlas for two onshore sites in Sweden. Wind resource assessment is conducted using NEWA mesoscale data as wind condition of the sites. AEP estimation is performed using two different simulation tools, and the results of estimation are compared to the actual SCADA data for the validation of NEWA. During the process of simulation, downscaling is executed for the mesoscale data to reflect micro terrain effects. The result of the current study showed that NEWA mesoscale data represents wind climate very well for the onshore site with simple terrain. On the other hand, NEWA provided overestimated wind speeds for the relatively complex onshore site with forested areas. The overestimation of wind speed led to predict AEP significantly higher than the measurements. The result of downscaling showed only little difference to the original data, which can be explained by the sites’ low orographic complexity. This study contributes to a deeper understanding of NEWA and provides insights into its validity for onshore sites. It is beyond the scope of this study to investigate whole region covered by NEWA. A further study focusing on sites with higher orographic complexity or with cold climate is recommended to achieve further understanding of NEWA.

In recent years, machine learning applications have gained great attention in the wind power industry. Among these, artificial neural networks have been utilized to predict the fatigue loads of wind turbine components such as rotor blades. However, the limited number of contributions and differences in the used databases give rise to several questions which this study has aimed to answer. Therefore, in this study, 5-min SCADA data from the Lillgrund wind farm has been used to train two feed-forward neural networks to predict the fatigue loads at the blade root in flapwise and edgewise directions in the shape of damage equivalent loads.The contribution of different features to the model’s performance is evaluated. In the absence of met mast measurements, mesoscale NEWA data are utilized to present the free flow condition. Also, the effect of wake condition on the model’s accuracy is examined. Besides, the generalization ability of the model trained on data points from one or multiple turbines on other turbines within the farm is investigated. The results show that the best accuracy was achieved for a model with 34 features, 5 hidden layers with 100 neurons in each hidden layer for the flapwise direction. For the edgewise direction, the best model has 54 features, 6 hidden layers, and 125 neurons in each hidden layer.For a model trained and tested on the same turbine, mean absolute percentage errors (MAPE) of 0.78% and 9.31% are achieved for the flapwise and edgewise directions, respectively. The seen difference is argued to be a result of not having enough data points throughout the range of edgewise moments. The use of NEWA data has been shown to improve the model’s accuracy by 10% for MAPE values, relatively. Training the model under different wake conditions did not improve the model showing that the wake effects are captured through the input features to some extent. Generalization of the model trained on data points from one turbine resulted in poor results in the flapwise direction. It was shown that using data points from multiple turbines can improve the model’s accuracy to predict loading on other turbines.