Search

In recent years, the potential of remote sensing-based minute-scale forecasts to improve the integration of wind power into our energy system has been shown. In lidar-based forecasts, the wind speed is extrapolated from the measuring to the forecast height, i.e. the wind turbines' hub height, by assuming a stability-corrected logarithmic wind profile. The objective of this paper is the significant reduction of large forecasting errors associated with the height extrapolation. Hence, we introduce two new approaches and characterise their skill under different atmospheric conditions. The first one is based on an empirical set of parameters derived from lidar data and operational wind turbine data. The second approach derives the wind speed tendency of two consecutive forecasts at the measuring height and applies this to operational wind speed data at hub height. We identified the uncertainty in stability estimates and measurement height as the main cause for large extrapolation errors of the existing lidar-based forecast. Monte Carlo simulations revealed the new approaches' low sensitivity to uncertainty in lidar data processing, propagation and height extrapolation. Forecasting errors of a 5‑minute-ahead wind speed forecast of free-flow turbines at an offshore wind farm were significantly reduced for the two newly developed methods as compared to the existing forecast during stable atmospheric conditions. Persistence could be outperformed during unstable and neutral atmospheric conditions and for situations with higher turbulence intensity. Overall, we found lidar-based forecasts to be less sensitive to atmospheric conditions than persistence. We discuss the importance of accurate vertical wind speed profile estimation, the advantages and shortcomings of the two newly introduced methods and their skill compared to persistence. In conclusion, the additional use of wind turbine operational data can significantly improve minute-scale lidar-based forecasts. We further conclude that the characterisation of forecast skill dependent on atmospheric conditions can be valuable for decision-making processes.

Energy system modeling and analysis can provide comprehensive guidelines to integrate renewable energy sources into the energy system. Modeling renewable energy potential, such as wind energy, typically involves the use of wind speed time series in the modeling process. One of the most widely utilized datasets in this regard is ERA5, which provides global meteorological information. Despite its broad coverage, the coarse spatial resolution of ERA5 data presents challenges in examining local-scale effects on energy systems, such as battery storage for small-scale wind farms or community energy systems. In this study, we introduce a robust statistical downscaling approach that utilizes a machine learning approach to improve the resolution of ERA5 wind speed data from around 31 km × 31 km to 1 km × 1 km. To ensure optimal results, a comprehensive preprocessing step is performed to classify regions into three classes based on the quality of ERA5 wind speed estimates. Subsequently, a regression method is applied to each class to downscale the ERA5 wind speed time series by considering the relationship between ERA5 data, observations from weather stations, and topographic metrics. Our results indicate that this approach significantly improves the performance of ERA5 wind speed data in complex terrain. To ensure the effectiveness and robustness of our approach, we also perform thorough evaluations by comparing our results with the reference dataset COSMO-REA6 and validating with independent datasets.

Mesoscale wind speed fluctuations influence the characteristics of offshore wind energy. These recurring wind speed changes on time scales between tens of minutes and six hours lead to power output fluctuations. In order to investigate the meteorological conditions associated with mesoscale wind speed fluctuations, a measure is needed to detect these situations in wind speed time series. Previous studies used the empirical Hilbert-Huang Transform to determine the energy in the mesoscale frequency range or calculated the standard deviation of a band-pass filtered wind speed time series. The aim of this paper is to introduce newly developed empirical mesoscale fluctuation measures and to compare them with existing measures in regard to their sensitivity to recurring wind speed changes. One of the methods is based on the Hilbert-Huang Transform, two on the Fast Fourier Transform and one on wind speed increments. It is found that despite various complexity of the methods, all methods can identify days with highly variable mesoscale wind speeds equally well.

In a highly renewable future pan-European power system, wave power might complement the renewable generation mix in a beneficial way. The potential of wave energy is estimated to be highest at Western European coastlines. Thus, the Iberian Peninsula is characterized by high wind, photovoltaic and wave resources. Five years of data on generation and load were used to identify the optimal share of wave power in a fully renewable power system on the Iberian Peninsula. This optimal share is defined by the minimization of needed backup energy from dispatchable sources in the system. First, the properties of the mix are investigated for the case of an isolated Iberian power system. Second, the mix is investigated when the Iberian Peninsula is connected to a fully renewable pan-European power system. The optimal share of wave power on the isolated Iberian Peninsula with respect to the need for additional backup is found to be 25% (wind 52%, photovoltaics 23%). This optimum does not change significantly, if hydro power is added to the generation mix. If compared to a system without wave power, the benefit from wave power equals an reduction of 6–8% of the backup energy need. For a fully connected European power system, the optimal mix on the Iberian Peninsula is determined to be 21% wave, 4% PV and 75% wind.

Unlike deterministic forecasts, probabilistic predictions provide estimates of uncertainty, which is an additional value for decision-making. Previous studies have proposed the analog ensemble (AnEn), which is a technique to generate uncertainty information from a purely deterministic forecast. The objective of this study is to improve the AnEn performance for wind power forecasts by developing static and dynamic weighting strategies, which optimize the predictor combination with a brute-force continuous ranked probability score (CRPS) minimization and a principal component analysis (PCA) of the predictors. Predictors are taken from the high-resolution deterministic forecasts of the European Centre for Medium-Range Weather Forecasts (ECMWF), including forecasts of wind at several heights, geopotential height, pressure, and temperature, among others. The weighting strategies are compared at five wind farms in Europe and the U.S. situated in regions with different terrain complexity, both on and offshore, and significantly improve the deterministic and probabilistic AnEn forecast performance compared to the AnEn with 10‑m wind speed and direction as predictors and compared to PCA-based approaches. The AnEn methodology also provides reliable estimation of the forecast uncertainty. The optimized predictor combinations are strongly dependent on terrain complexity, local wind regimes, and atmospheric stratification. Since the proposed predictor-weighting strategies can accomplish both the selection of relevant predictors as well as finding their optimal weights, the AnEn performance is improved by up to 20 % at on and offshore sites.

With increasing wind power capacity, accurate uncertainty forecasts get more and more important for grid integration. The uncertainty of forecasts can be quantified by ensemble forecasts. We use ensemble forecasts from the COSMO-DE EPS to generate short-term ensemble forecasts of regionally aggregated wind power. The wind power forecasts are generated by an optimised regional power curve model that is based on minimum score estimation and leads to wind power forecasts with small deterministic errors. Remaining bias and dispersion errors in the wind power forecasts are removed by statistical post-processing (also called calibration) with ensemble model output statistics and the temporal rank correlation of the raw ensemble is maintained by ensemble copula coupling. The verification of raw and calibrated ensembles shows both strong improvements by calibration and the benefit of ensuring time consistency with ensemble copula coupling. The improvements are indicated by the multivariate energy score as well as in a proposed univariate verification approach that is based on integrated wind power forecast and measurement trajectories. Slight deficits in time consistency of the forecasts remain because the theoretical assumptions of ensemble copula coupling are not always fulfilled as the COSMO-DE EPS is based on distinguishable ensemble members. The more training days are used for calibration against measurements of regionally aggregated wind power, the lower is the improvement by calibration which contradicts former results for different variables like wind speed.

Very short-term forecasts of wind power provide electricity market participants with extremely valuable information, especially in power systems with high penetration of wind energy. In very short-term horizons, statistical methods based on historical data are frequently used. This paper explores the use of dual-Doppler radar observations of wind speed and direction to derive five-minute ahead deterministic and probabilistic forecasts of wind power. An advection-based technique is introduced, which estimates the predictive densities of wind speed at the target wind turbine. In a case study, the proposed methodology is used to forecast the power generated by seven turbines in the North Sea with a temporal resolution of one minute. The radar-based forecast outperforms the persistence and climatology benchmarks in terms of overall forecasting skill. Results indicate that when a large spatial coverage of the inflow of the wind turbine is available, the proposed methodology is also able to generate reliable density forecasts. Future perspectives on the application of Doppler radar observations for very short-term wind power forecasting are discussed in this paper.

Shares of renewables continue to grow in the European power system. A fully renewable European power system will primarily depend on the renewable power sources of wind and photovoltaics (PV), which are not dispatchable but intermittent and therefore pose a challenge to the balancing of the power system. To overcome this issue, several solutions have been proposed and investigated in the past, including storage, backup power, reinforcement of the transmission grid, and demand side management (DSM). In this paper, we investigate the potential of DSM to balance a simplified, fully renewable European power system. For this purpose, we use ten years of weather and historical load data, a power-flow model and the implementation of demand side management as a storage equivalent, to investigate the impact of DSM on the need for backup energy. We show that DSM has the potential to reduce the need for backup energy in Europe by up to one third and can cover the need for backup up to a renewable share of 67%. Finally, it is demonstrated that the optimal mix of wind and PV is shifted by the utilisation of DSM towards a higher share of PV, from 19% to 36%.

The capacity factor of a power plant is the ratio of generation over its potential generation. It is an important measure to describe wind and solar resources. However, the fluctuating nature of renewable power generation makes it difficult to integrate all generation at times. Whenever generation exceeds the load, curtailment or storage of energy is required. With increasing renewable shares in the power system, the level of curtailment will further increase. In this work, the influence of the curtailment on the capacity factors for a highly renewable German power system is studied. An effective capacity factor is introduced, and the implications for the distribution of renewable power plants are discussed. Three years of highly-resolved weather data were used to model wind and solar power generation. Together with historical load data and a transmission model, a possible future German power system was simulated. It is shown that effective capacity factors for unlimited transmission are strongly reduced by up to 60% (wind) and 70% (photovoltaics) and therefore of limited value in a highly renewable power system. Furthermore, the results demonstrate that wind power benefits more strongly from a reinforced transmission grid than photovoltaics (PV) does.