OSA2.3

A method to combine a highly resolved All sky imager (ASI) network forecast with a satellite based forecast has been developed. The ASI network forecast input is based on the data from the DLR’s Eye2Sky network. This network is installed in North West Germany and includes 29 ASIs, ten Rotating Shadowband Irradiometers (RSIs) and two reference meteorological stations (based on thermal irradiometers) in an extent of 100 km². This network forecast was developed by our colleges from DLR-SF (publication in preparation). It has a forecast horizon of 30 minutes and a step of 1 min with an update of 30 seconds on a domain of 40 km². The second input is based on our operational satellite forecast at DLR-VE and has a horizon of 6 hours with a step and update of 15 minutes. The satellite domain is reduced to the same 40 km² area.

The method consists of three blocks, forecasts homogenization, regression and prediction. In the homogenization block the satellite forecast is interpolated in space and time to the resolutions of the ASI network forecast. We applied linear interpolation for both resolutions as first test case. In the second block, a linear regression is applied to find the optimal weights of the linear combination of the forecast inputs, including a bias term. The regression is based on timeseries extracted from the historical forecasts (features) where the reference is taken from the historical timeseries of ground measurements (samples). Historical data is used in order to indirectly characterize the mean actual local weather conditions on the domain. It is important to note that the regression is performed independently for every lead time. In the third block, we use the optimized weights and biases along with the present (not historical) forecasts to produce the hybrid forecasts. The hybrid forecast resolutions are the same as the ASI based forecast. The output product can be given as maps or timeseries.

For the test case, we are limited from the ASI network side to a dataset of two full months of forecasts (July and August 2020). The highly resolved hybrid forecast was validated against the individual input sources and satellite persistence. We found that this newly developed forecast outperforms the RMSE of persistence and the individual input forecasts for all calculated lead times. It shows an improvement on RMSE of 5.1% to 14.0% with respect to satellite forecasts and 7.6% to 15.1% with respect to the ASI network forecast on lead times going from 5 to 30 minutes. It also shows a lower RMSE under high variability conditions.

How to cite: Lezaca, J., Hammer, A., and Lünsdorf, O.: High resolution hybrid forecast based on the combination of satellite and an All Sky Imager (ASI) network forecasts, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-486, https://doi.org/10.5194/ems2022-486, 2022.

How to cite: Good, G.: Pareto optical flow solutions for ensemble, satellite-based forecasts of irradiance and PV, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-175, https://doi.org/10.5194/ems2022-175, 2022.

Cloud forecasting is an enormous challenge in numerical weather prediction (NWP) models because of the complex physical processes, high variability in spatial and temporal scales, and lack of observations for model evaluation. The capability to represent cloud fields in the NWP models directly impacts the accuracy in predicting surface solar irradiance. Verifying cloud forecasts from NWP models is essential to investigate the source of uncertainty and error stemming from predicting various types of clouds. However, it requires high-quality observations containing various information of clouds and in-depth analysis over a wide range of regions to assess the cloud fields. In this study, we produce day-ahead cloud forecast over the contiguous United States (CONUS) for 2018 using the Weather Research and Forecasting-Solar Ensemble Prediction System (WRF-Solar EPS) which is a state-of-the-art ensemble NWP model specialized for solar applications. The strengths and limitations of WRF-Solar EPS in reproducing cloud fields is diagnosed using satellite observations from the National Solar Radiation Database (NSRDB). The frequency of clouds and various cloud detection metrics including the probability of detection (POD), the false alarm rate (FAR), the hit rate (HR), Kuiper’s skill scores (KSS), and mismatched cloud frequency (MCF) are calculated to assess the performance of WRF-Solar EPS. In the first part of this study, we focus on monthly analysis using the detection metrics to account for seasonal performance of WRF-Solar EPS. In the second part, the MCF classified by cloud top height and cloud optical depth is analyzed to investigate the model’s capability to predict nine different types of clouds. The study exhibits that the WRF-Solar EPS has difficulty predicting optically thin clouds; overall MCFs show 46% (cumulus), 34% (stratocumulus), 19% (stratus), 33% (altocumulus), 23% (altostratus), 16% (nimbostratus), 27% (cirrus), 13% (cirrostratus), and 8% (deep convective) for the nine cloud types.

How to cite: Sengupta, M., Yang, J., Xie, Y., Jimenez, P., and Kim, J.-H.: Using Satellite Information to Evaluate Cloud Forecast from WRF-Solar EPS, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-609, https://doi.org/10.5194/ems2022-609, 2022.

Photovoltaic production is highly variable with a limited predictability due to stochastic variation of cloud attenuation of surface solar irradiance. Forecasting this irradiance helps to manage the safe and stable operation of the grid and thus enables to increase solar electricity penetration for a lower carbonated energy mix. In particular, intraday horizon forecasts (typically up to 6 h) are more and more required for the grid power reserve management, the increase of intra-day auctions in the electricity market and the current worldwide development of micro-grids.

Forecast irradiance using images from geostationary meteorological satellite is particularly appropriate for intraday horizon. It gives better performance than NWP models, does not require instrumentation, or costly computing resource. However, the accuracy of such methods is very sensitive to the cloud cover state and its short-term evolution. For instance, methods based on the temporal extrapolation of cloud motion present better results for passing cloud events than for sudden cloud appearance or disappearance. In several studies, reliability predictors have been identified for satellite-based irradiance forecast. They showed clear signals when uncertainty is computed as a function of season, solar zenith angle, cloud albedo and more recently synoptic weather regimes. This anticipation of error range would help grid managers to prepare the sizing of storage capacity or ancillary electricity resource.

A pertinent predictor must have a significant influence on forecast error. It also must be easy to obtain in operational forecast conditions. In this work, we propose to exploit the only observation source required in satellite-based forecast: the satellite image itself. Using 5 years (2017-2021) of image-derived forecasts at 15 min time step over Palaiseau (France), we computed the forecast uncertainties as a function of multiple parameters derived from the HRV channel of Meteosat Second Generation satellite. We highlighted the influence of cloud albedo spatial variance and cloud motion vector field spatial-temporal features in the forecast uncertainties. In addition to the error range provided to users, this work can help forecasters to better characterize their sources of error and to select new predictors for machine-learning approaches.

How to cite: Cros, S., Badosa, J., Szantaï, A., and Haeffelin, M.: On hand available predictors for operational satellite-based forecast, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-572, https://doi.org/10.5194/ems2022-572, 2022.

With the increase of solar power use, the need for solar irradiance forecasts is also increasing. Solar power is a naturally fluctuating energy source, which makes solar irradiance forecasts necessary for e.g. grid integration and management. Both the use of solar energy and the need for forecasts are present also in the Nordic countries, where the sub-Arctic latitude brings its own challenges.

Various methods exist for forecasting solar irradiance. Numerical Weather Prediction (NWP) models have been found superior for forecasting for the following days, while satellite-based models are found suitable for the first few hours of the forecast. The satellite-based model Solis-Heliosat has been found to perform well also in the high latitudes. The current operational NWP models in Finland, however, have not been yet extensively validated for this purpose.

To determine the suitability of the operational NWP models for forecasting solar irradiance in the Nordic countries, we have comparatively validated the MetCoOp Ensemble Model (MEPS), and the MetCoOp Nowcasting Model (MNWC) against in situ irradiance measurements at several stations in Finland and Sweden. We have also included the Solis-Heliosat model in the study, to improve our understanding on the differences and the relative accuracy between the NWP and satellite-based models. As a benchmark, two persistence models are included. The comparison is made for one summer, including all model runs and all MEPS ensemble members, with both hourly and 15 minute output depending on the model.

The results show all models to somewhat under predict irradiance. MEPS shows very good performance in the full length of the forecast, while Solis-Heliosat is better in the first 2-3 hours of the forecast. Solis-Heliosat has some difficulty with the forecasts starting in the morning, whereas MNWC slightly struggles in the afternoon.

Overall we find the NWP models very suitable for forecasting solar irradiance in Finland and Sweden, particularly with the full forecast horizon of MEPS, and the 15-minute time step of MNWC. Nevertheless, Solis-Heliosat brings further value to the beginning of the forecast.

Kallio‐Myers, V., Riihelä, A., Schoenach, D., Gregow, E., Carlund, T., & Lindfors, A. V. (2022). Comparison of irradiance forecasts from operational NWP model and satellite‐based estimates over Fennoscandia. Meteorological Applications, 29(2), e2051.

How to cite: Kallio-Myers, V., Riihelä, A., Schoenach, D., Gregow, E., Carlund, T., and Lindfors, A.: Comparison of irradiance forecasts from operational NWP model and satellite-based estimates over Fennoscandia, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-203, https://doi.org/10.5194/ems2022-203, 2022.

As Europe strives for a decarbonization of electricity supply, many countries proclaim ambitious targets for capacity additions of wind and photovoltaic (PV) power. By increasing renewable capacities, countries seek to reduce their dependency on fossil fired electricity generation. For instance, Germany recently announced new capacity targets that mean a triplication of wind and PV power capacities while at the same time aiming for a full coal exit until 2030. However, the potential of renewable energies is not equally distributed over the European continent. Due to higher irradiance levels with decreasing latitude, PV power achieves higher full load hours and thus higher generation per installed capacity in southern Europe.  Wind power reaches higher capacity factors close to shorelines, where near-surface wind speeds are higher.

In light of the unequal distribution of PV and wind power generation potentials, national capacity targets may not lead to optimal utilization of capacities, i.e. maximum renewable generation. We analyse the effects of such national capacity targets for the case of PV generation. To do so, we compare two scenarios: A scenario with PV generation capacity distributed over Europe according to national political targets and a scenario with a distribution of newly installed capacities chosen such that it optimizes the renewable power potentials. Next to PV generation potentials, our assessment considers the European power grid, because line capacities limit the transmission and hence the trading between different countries. Line limits can have a negative impact on the overall power generation since renewable generation has to be curtailed if power can neither be locally used nor transmitted to another region.  We employ a linear optimization-based electricity market model with inter-European transmission constraints and hourly resolution weather data for 35 historic weather years (1982-2016). We analyse the scenarios if terms of differences in power generation and curtailment, greenhouse gas emissions, and system operational cost. We pay special attention to extreme weather situations and the effects of a sub-optimal PV power allocation on security of supply.

How to cite: Czock, B., Keutz, J., and Fiedler, S.: Follow the sun? The effects of national solar capacity targets on renewable generation and security of supply, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-198, https://doi.org/10.5194/ems2022-198, 2022.

Modelled irradiance data based on satellite products is frequently used in solar energy applications and atmospheric sciences. This kind of data is offered by many different institutional or commercial providers, and currently it is not possible for users to independently identify the best provider for their specific application and location. This work presents a benchmark of satellite-derived global horizontal irradiance (GHI) as well as direct normal irradiance (DNI) at 129 ground-based radiation measurement stations distributed globally. High temporal resolution data (1 min) from these stations from between 2015 and 2020 has been quality controlled by a team of experts using a comprehensive set of best practices and newly implemented quality control procedures. The ground stations provide measurements of GHI and DNI and/or diffuse horizontal irradiance (DIF) from 25 different providers or networks. The 129 stations are spread out worldwide with 31 stations in Africa, 31 in Asia, 27 in North America, 20 in Europe, 13 in Australia, 5 in South America and 2 in Antarctica. GHI and DNI data from eleven different commercial or open-access radiation models is compared against these stations’ high-quality ground data. Additionally, one common measurement year is used to perform a site adaptation of the model datasets. The site adaptation is based on the empirical quantile mapping method and is the same for all test data sets. The comparison of the raw model data as well as the site adapted data is conducted at both 60-min and 15-min temporal resolutions. The performance of the raw and site-adapted data is analysed with respect to different regions and climate zones. Users can rely on the results of this work to make an informed decision about which surface radiation model(s) and data providers are most suited for their application.

How to cite: Forstinger, A., Wilbert, S., Jensen, A. R., Kraas, B., Fernández Peruchena, C., Gueymard, C. A., Ronzio, D., Yang, D., Collino, E., Polo Martinez, J., Ruiz-Arias, J. A., Hanrieder, N., Blanc, P., and Saint-Drenan, Y.-M.: Worldwide solar radiation benchmark of modelled surface irradiance, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-547, https://doi.org/10.5194/ems2022-547, 2022.

The estimation of solar surface irradiance at high spatio-temporal resolution from geo-stationary satellite images is a well-established technique, for example by using the Heliosat method. The method has widely reduced the need for expensive ground measurements, especially in remote regions. However, the location of cloud shadows at the ground is difficult to determine and thus a significant source of errors when either the distance from the sub-satellite point or the cloud top height (CTH) increases. Although several methods have been proposed in the literature to reduce these errors, it is still an issue. We present a novel approach to correct the cloud shadow location based on the satellite-cloud-sun geometry using the CTH maps from the EUMETSAT data archive. It uses satellite viewing angles and solar position angles to determine the correct cloud shadow location for each cloudy pixel. The method is tested on cloud index (CI) maps for the months of July, August and September 2018 derived by applying the Heliosat method on the 0.6 um visible channel images from Meteosat-8 located at 41.5°E. Convective clouds with large CTHs are frequently observed over the Indian subcontinent in these three months due to the Indian summer monsoon. The global horizontal solar irradiance (GHI) obtained from the corrected CI image is validated at two BSRN stations. The normalized root mean square error (nRMSE) is reduced from 23.2% to 20.9% for the Gurgaon station and from 15.4% to 13.9% at Tiruvallur. In general, correcting the cloud shadow location on CI map improved the accuracy of the estimated GHI. Nonetheless, the method is sensitive to the accuracy of the CTH dataset and individual cases were found for which the correction reduced the accuracy.

How to cite: Roy, A., Hammer, A., Heinemann, D., and Lünsdorf, O.: Cloud Shadows in Satellite-Based Solar irradiance Estimation: Improved Correction using EUMETSAT’s Cloud Top Height Data, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-504, https://doi.org/10.5194/ems2022-504, 2022.

Accurate intraday forecasts of surface solar radiation (SSR) are essential for utility companies and electricity grid operators. However, satellite SSR estimates can suffer from surprisingly low accuracies on short observation time scales and from significant spatial and temporal biases when compared to ground-based SSR measurements.

We present a bias assessment of two high-resolution satellite SSR products for intra-day application time scales. The satellite SSRs are retrieved with the Heliosat SARAH-2 and the HelioMont algorithms (Müller et al., 2015; Stöckli, 2013). We investigate intra-day and intra-hour estimates for altitudes from 200 to 3570 m a.s.l. We make use of 133 ground stations of the high-precision monitoring network SwissMetNet for the bias analysis. For solar zenith angle (SZA) lower than 90 degrees, we find that Heliosat SARAH-2 underestimates SSR at high altitudes (over 1000m) with an instantaneous root mean squared deviation (RMSD) of 179 W/m2and a mean bias deviation (MBD) of -68 W/m2 in the winter half year. The bias magnitude is approximately the double w.r.t. low altitude stations, due to difficulties in distinguishing snow-covered surfaces from clouds.

We also present an intra-hour bias correction approach: a deep neural network exploiting time-encoding features to model the bias in the time dimension. Our model achieves up to 30% RMSE reduction in the case of Heliosat SARAH-2, especially in mountainous regions. Moreover, we highlight the importance of the clear-sky index for bias correcting the Heliosat SSR estimates. Including a clear-sky index as a regressor in the bias correction improves the bias correction on average from 15.5% to 21.9% RMSE reduction. We also discuss the relevance of bias correcting satellite-derived SSR maps for short-term forecasting applications of SSR.

References

• Müller, R., U. Pfeifroth, C. Träger-Chatterjee, J. Trentmann, and R. Cremer (2015), Digging the METEOSAT Treasure-3 Decades of Solar Surface Radiation, Remote Sensing, 7(6), 8067-8101, doi:10.3390/rs70608067.
• Stöckli (2013). The HelioMont Surface Solar Radiation Processing. Scientific Report 93, MeteoSwiss, 122 pp.

How to cite: Carpentieri, A., Wild, M., Folini, D., and Meyer, A.: Characterizing and correcting Heliosat Surface Solar Radiation bias on intra-day time scales with deep neural networks, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-366, https://doi.org/10.5194/ems2022-366, 2022.

Dramatically reduced costs, increased electrification and a growing resistance to onshore wind power, makes solar energy, and photovoltaics increasingly relevant in Norway. However, there is a lack of accurate knowledge on the available solar resource. Of key importance is solar irradiance, and the accuracy of this is paramount for the accuracy of energy production estimates. Global horizontal irradiation (GHI) for a location can be found in several different databases, but they are less reliable for high latitudes such as Norway. In our study of the solar resource in Norway we assess surface measurements, satellite data, re-analysis and weather prediction model data. Surface measurements are available for about 100 locations in Norway, but the quality of the data is often not well documented, and quality assurance is necessary. Due to the lack of coverage of geostationary satellite data over the Northernmost part of Norway polar orbiting satellites are preferred, but the passages of the polar satellites are temporally irregular. Model GHI data has regular temporal resolution and availability, but the accuracy is uncertain and needs further exploitation. In this work we present a quality assurance of surface GHI measurements including visual inspection tools, and a comparison of the quality assured surface measurements with satellite and model derived GHI. A discussion of the three data sets is given and a preliminary solar resource map for Norway is also presented.  Our study shows that locations with the highest annual GHI of ca. 1000-1100 kWh/m2 are encountered in high mountain areas with GHI peaking in late spring early simmer when the mountains still are covered with snow. Slightly lower GHI values of about 1000 kWh/m2 are found at southern coastal sites reaching the peak GHI in mid-summer. Lower values of typically 700-800 kWh/m2 are seen at the west coast and 600-700 kWh/m2 are in northern Norway. 

How to cite: Dobler, A., Berge, E., Eastwood, S., Førland, J. R., Hygen, H. O., and Sætra, M. L.: Solar resource mapping in Norway, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-404, https://doi.org/10.5194/ems2022-404, 2022.

West Africa has great potential for the use of solar energy systems, as it has both a high solar radiation rate and a lack of energy production.  West Africa is a very aerosol-rich region, whose effects on photovoltaic (PV) use are due to both atmospheric conditions and existing solar technology.  This study reports the variability of aerosol optical properties in the city of Koforidua, Ghana over the period 2016 to 2020, and their impact on the radiation intensity and efficiency of a PV cell. The study used AERONET ground (Giles et al., 2019) and satellite data produced by CAMS (Gschwind, et al., 2019), which both provide aerosol optical depth (AOD) and metrological parameters used for radiative transfer calculations with libRadtran (Emde, et al., 2016). A spectrally resolved PV model (Herman-Czezuch et al., 2022) is then used to calculate the PV yield of two PV technologies: polycrystalline and amorphous silicon. It is observed that for both data sets, the aerosol is mainly composed of dust and organic matter, with a very increased AOD load during the harmattan period (December-February), also due to the fires observed during this period.

We compared CAMS satellite data with AERONET ground data. A good annual correlation (correlation coefficient, R² ~0.8) was observed between the CAMS and AERONET AOD data. However, CAMS satellites tend to underestimate the high AOD measured on the ground by AERONET photometers; but they also overestimate the low AOD compared to AERONET. Both datasets also show differences in the average assumed optical properties of the aerosol encountered.  A detailed analysis of the year 2020 shows a daily reduction in PV yield for the polycrystalline cell of up to 90% with the AERONET data and of up to 32% with the CAMS data. For the amorphous cell, the daily reduction in PV yield is up to 71% with the AERONET data and 34% with the CAMS data. These strong differences are due to the seasonal dispersion of the measured AODs but also to the variability of water vapor and ozone concentrations provided by CAMS and AERONET over Koforidua.

Acknowledgements: Funding was provided by the German BMBF under contract 03SF0567A-.

References

• Emde et al. (2016). The libRadtran software package for radiative transfer calculations (version 2.0.1). Geosci. Model Dev. 9 (5), 1647-1672.
• Giles et al. (2019). Advancements in the Aerosol Robotic Network (AERONET) Version 3 database. Atmospheric Measurement Technique, 12(1), 169-209.
• Gschwind et al. (2019). Improving the McClear model estimating the downwelling solar radiation at ground level in cloud free conditions – McClear-V3. Meteorol. Z./Contrib. Atm. Sci. 28, 147-163.
• Herman-Czezuch et al. (2022). Impact of Aerosols on Photovoltaic Energy Production Using a Spectrally Resolved Model Chain: Case Study of Southern West Africa. submitted to Solar Energy

How to cite: Meilinger, S. and Zemo Mekeng, A.: Influence of aerosols on photovoltaic power in Ghana: Case study from Koforidua, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-286, https://doi.org/10.5194/ems2022-286, 2022.

Over the last years, climate monitoring and operational weather forecasts have become an important topic for the renewable energy sector. An effective operation of national, and in the case of EU international, power generation aims to find the right balance between the minimization of CO2 emission and reduction of energy costs. In Germany, a considerable part of the electricity generation comes from wind. Therefore, an accurate forecast of low-level wind is essential to predict the generation of electrical power produced by wind parks. This enables timely adjustments of the conventional power plants. Currently, short-term low-level wind forecasts have considerable uncertainties. One of the cost-effective solutions to improve low-level wind forecasts is the assimilation of new observations into numerical weather prediction models. Even though in the last decade, the number of remote-sensing sites has been continuously growing, the coverage is far from being optimal to achieve significant improvement in the short-term wind forecast. However, before building new large networks of ground-based instruments it is important to estimate in advance which instruments to install, what effect to expect, and what spatial density of the distributed instruments should be.

Ground-based instruments that can provide valuable information for low-level wind forecasts are Doppler lidars. In this study, we focus on the estimation of the potential impact of Doppler lidars for short-term low-level wind forecasts essential for sustainable energy applications. The potential impact is analyzed using the ensemble sensitivity analysis (ESA) [1]. ESA is based on the Ensemble Transform Kalman Filter and allows us to investigate how the assimilation of hypothetical Doppler lidars can reduce the wind forecast variance. The impact of a Doppler lidar network was analyzed with respect to surface measurements operationally assimilated by national weather services. We investigated the sensitivity of the obtained results to the number of Doppler lidars in the network, the number of altitude layers observed by Doppler lidars, and forecast lead time. Our analysis is based on a 1000-member ensemble simulation for the urban and highly populated Rhein-Ruhr area and surrounding regions [2]. The simulation uses a full-physics non-hydrostatic regional model (SCALE-RM) and covers a two-week time period in May/June 2016.

This work has been conducted in the framework of the Hans-Ertel-Centre for Weather Research funded by the German Federal Ministry for Transportation and Digital Infrastructure (grant number BMVI/DWD 4818DWDP5A). This online publication is based upon work within the COST Action CA18235 supported by COST (European Cooperation in Science and Technology). We acknowledge RIKEN for providing the SCALE-RM model data.

References 
[1] Ancell, B., and G. J. Hakim, 2007: Comparing adjoint-and ensemble-sensitivity analysis with applications to observation targeting, MWR., doi.org/10.1175/2007MWR1904.1. 

[2] Necker, T., et al, 2020: A convective-scale 1000-member ensemble simulation and potential applications. QJRMS, doi.org/10.1002/qj.3744.

How to cite: Nomokonova, T., Griewank, P., Löhnert, U., Miyoshi, T., Necker, T., and Weissmann, M.: Benefits of Doppler wind lidars to improve short-term low-level wind forecasts, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-216, https://doi.org/10.5194/ems2022-216, 2022.

The increasing amount of electricity generated from wind power leads to a stronger variability in electricity supply to the national electricity grids. This makes wind power forecasts a necessary component in the wind energy market. The forecasts are used to estimate the amount of energy that will be generated and prevent potential imbalances between production and consumption in electricity grids. Energy price in Finland is set every day at the Nordic energy trading stock Nordpool. Unaccounted shortage or surplus of power in the grid needs to be compensated. This may result in costs required to balance the grid.

To account for the uncertainties in wind power calculations, a probabilistic wind power forecasting tool has been developed. It is based on the MetCoOp Ensemble Prediction System (MEPS), in which the HARMONIE Numerical Weather Prediction model is run with 2.5 km horizontal resolution and 65 vertical levels. The system is operational through joint efforts of a group of Nordic countries participating in the MetCoOp cooperation.

The skills of wind power calculations, however, depend not only on the skills of the MEPS system, but also on the information on the wind farms. Detailed information on the locations of wind turbines (WT), their hub heights and technical specifications is important in adequate calculations of power production and power losses. Most essential WT’s parameters are so-called power and thrust coefficient (C_T) curves. Power curve is necessary to convert wind speed to corresponding power output. C_T curve is needed to calculate wind flow retardation by WT, also called wake effect, resulting in power losses downwind in a wind farm. Power and C_T curves, as well as other WT technical parameters, are turbine-specific and typically provided by the manufacturer under commercial terms and conditions. In power calculations over a country with a multitude of wind farms, gathering such information may be challenging.

Nowadays power curves of many WT models can be found in open sources, such as thewindpower.net and wind-turbine-models.com. However, they are still missing for some and especially new WT models. C_T curves, on the other hand, are practically not available in open sources. A statistical solution was developed to approximate both power and C_T curves using as input commonly available WT specifications. The uncertainty in power calculations associated with approximated power and C_T curves was found to be smaller than that associated with the uncertainty in the predicted wind speed.

Wind to power conversion is done for essentially all WTs installed in Finland, to produce an aggregated probabilistic wind power forecast. Wind variability is accounted by applying Gaussian smoothening to power (and C_T) curves. Wake-related losses were calculated applying Katic-Jensen wake model, with missing C_T obtained applying the above method.

How to cite: Atlaskin, E., Lindfors, A., Kallio-Myers, V., and Suomi, I.: Probabilistic forecasting of the aggregated Finnish wind energy based on the MetCoOp ensemble prediction system (MEPS), EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-219, https://doi.org/10.5194/ems2022-219, 2022.

The increasing penetration of the power grid with renewable energy (RE) sources comes with an increase of distributed generation units (DGUs), which exhibit a very heterogeneous structure. Grid operators often face challenges when employing RE due to its large volatility. Especially in exceptional situations, e.g. power blackouts or frequency events, conventional generation units are often preferred to stabilize the grid. Making use of highly volatile power sources requires very reliable knowledge about the near future of the available active power (AAP) signal, hence giving rise to the necessity of accurate and robust RE generation forecasting in short time ranges (nowcasting).

In order to maximize the availability of forecast data to the grid operators the forecasting process can be deployed in a distributed fashion. The presented method makes use of several machine learning (ML) algorithms and provides a probabilistic short-term forecast with little requirements to computational resources on-site. The local availability of forecasts improves the AAP signal and hence enables ancillary services from wind farms such as the provision of a frequency containment or restoration reserve. Also it provides robustness against communication failure between grid operation and forecast provider.

The suggested forecasting procedure is two-fold: On the one hand a data-driven nowcasting (from 0 to 6 hours) approach is pursued within decentralized forecast units, that is designed to be deployable on-site of the respective DGUs. This approach employs local sensor data, i.e. active power, wind measurements from the nacelle anemometer and temperature, as well as the azimuth angle of the nacelle. The various ML models in use are adaptively trained on the latest sensor data and produce ensemble forecasts, from which both the minimally available power and the forecast uncertainty can be deduced. The individual model outputs are then combined by an adaptive genetic algorithm.

This purely data-based nowcast is then enriched with a physical forecast based on numerical weather prediction (NWP) model runs with a time horizon of up to 10 days. It considers actual production data from the wind farm as well as turbine specific control behaviour, which leads to excellent forecast quality. This NWP-based forecast is generated at a central computing centre and can be distributed to the forecasting units at the DGUs. On-site the data-based nowcast and the physics-based forecast are combined to provide good reliability for both time horizons.

In our paper we present the method in detail and describe the infrastructure, for which it was developed. We further conduct a performance assessment of the forecasting procedure and provide performance indicators of expected forecast errors and reliability.

Parts of the presented research have been carried out in the joint research project SysAnDUk (FKZ 03EI4004A) funded by the German Federal Ministry for Economic Affairs and Energy.

How to cite: Holicki, L., Dröse, M., Schürmann, G., and Letzel, M.: Decentralized forecasting of wind energy generation with an adaptive machine learning approach to support ancillary grid services, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-428, https://doi.org/10.5194/ems2022-428, 2022.

Offshore wind turbines induce atmospheric wakes downstream of typically tens of kilometers length while converting wind energy into electrical power. In the recent years, wakes behind single wind farms have been extensively studied. The growing number of offshore wind farms (OWFs) in the German Bight increases the occurrence of interferences between close neighboring OWFs. Previous studies showed that the proximity of neighboring OWFs impacts the performance of downwind turbines and reduces their capacity factor. However, the interactions of wakes from different wind farms is not well understood, in particular concerning the resulting wind speed deficits, turbulence intensities and superimposed wake lengths.

The interaction of wakes from two OWFs generally leads to longer wakes. Also, it was observed frequently that  wakes resulting from 2 OWFs reach a third wind farm. Several conditions can entail interactions of wakes from several OWFs, such as the size of OWF (geometry and density of turbines), atmospheric stability and wind speed and directions. The study focuses on the investigation of the interactions of wakes from two and three OWFs using Synthetic Aperture Radar system (SAR), which is an interesting instrument to observe large spatial areas at resolution of 20 m.  The statistical analysis of a 5-year period of SAR data acquisitions by the  Sentinel-1A and Sentinel1-B satellites revealed that the occurrence of interactions of wakes between OWFs exceeds 75% of all the cases for which wakes were observed. The interaction between OWFs is clearly correlated with a certain wind direction range. Additionally, the geometry of OWFs plays a role in the two-dimensional structure of the wakes and the potential for impacting neighboring wind farms. Obviously, wind directions parallel to the alignment of several OWFs are likely to induce strong interactions of wakes.

SAR data are combined with stability information from atmospheric models and mast measurements to analyse the respective impacts on key parameters of superimposed wakes (e.g. deficit, wake length).  Results are compared with simplified empirical models, which make assumptions about the linearity of the wake superposition process.   

How to cite: Djath, B. and Schulz-Stellenfleth, J.: Study of the interaction of atmospheric wakes from several offshore wind farms as observed by Synthetic Aperture Radar (SAR) system, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-465, https://doi.org/10.5194/ems2022-465, 2022.

Unique airborne in-situ measurements were evaluated to investigate the influence of offshore wind farms on the latent heat flux in the marine boundary layer. 21 of the total 42 measurement flights carried out in the framework of the WIPAFF project over the German Bight in the year 2016 and 2017 enabled such an evaluation under different atmospheric conditions. The airborne data was collected with the research aircraft Dornier 128 belonging to the Technische Universität Braunschweig, Germany. The aircraft is equipped among others with high-resolution sensors for water vapor, three-dimensional  wind vector, temperature and pressure. The measurements of 15 flights showed a significant increase of the vertical upward latent heat flux over the offshore wind farm clusters Amrumbank West, Nordsee Ost, Meerwind Süd/Ost or the wind farm cluster Godewind. For thermally stable conditions, all nine measurement flights except one showed an enhanced latent heat flux downstream of the wind farms at hub height, with an increase of up to +70 W m^-2compared to the undisturbed flow. For flights during unstable thermal conditions, 8 out of 13 cases showed an increase, with the largest difference with respect to undisturbed flow of +400 W m^-2 above the wind farm. The results also suggest that not only the thermal stratification but also the moisture gradient plays a decisive role in whether the influence of the wind farm is visible in the latent heat flux. 

How to cite: Platis, A., Büchau, Y., and Bange, J.: Influence of offshore wind farms on the latent heat flux in the marine boundary layer, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-174, https://doi.org/10.5194/ems2022-174, 2022.

Atmospheric stability is a measure of atmospheric status which determines whether thermodynamically perturbed air will rise, sink, or be neutral. Atmospheric stability has a major impact on the evolution of wind turbine wakes and thus on the yield and performance of offshore wind parks. For estimations of wind park power output and for improving analyses of offshore wind park wakes, a crucial parameter was found to be profiles of atmospheric temperature and stability metrics. Atmospheric temperature profiles can be measured in-situ by balloon-borne sensors, but also estimated from the ground using remote sensing observations.

Ground-based microwave radiometer (MWR) units operating in the 22-30 GHz and 50-60 GHz bands are commonly used to estimate atmospheric temperature and humidity profiles. A handful of MWR profiling types are nowadays available as off-the-shelf commercial products, and a MWR network is currently being established in the framework of EUMETNET E-PROFILE programme (Rüfenacht et al., 2021). This presentation reviews the stability metrics useful for monitoring wind park performances and provides a quantitative assessment of the value of MWR observations to estimate these stability metrics from near surface, either over land or ocean. Results from three different MWR instruments, representing the most common available on the market, will be presented, as obtained during at least three field experiments, both onshore and offshore.

This contribution presents the main outcomes of the Radiometry and Atmospheric Profiling (RAP) scoping study, carried in the framework of the COST Action PROBE (https://www.probe-cost.eu/) and funded by Carbon Trust and the partner companies of the Off-shore Wind Accelerator (OWA) program: (in alphabetical order) EnBW, Equinor, Orsted, RWE, Scottish Power Renewables, Shell, SSE Renewables, Total Energies, Vattenfall.

How to cite: Cimini, D., Gandoin, R., Fiedler, S., Wilson, H., Pospichal, B., Martinet, P., Balotti, A., Gentile, S., and Romano, F.: Assessment of atmospheric stability measurements from microwave radiometer observations for wind energy applications, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-631, https://doi.org/10.5194/ems2022-631, 2022.

It is scientific consensus that the Paris Agreement (COP21) can only be achieved by drastically reducing the emissions of anthropogenic greenhouse gases, in particular those related to burning fossil energy sources. Succesfully providing the required energy with renewable energy sources, such as wind and solar power, is dependant on the weather conditions. Dunkelflauten, periods during which only little energy is available by wind and solar power, can put strain on the energy system based on high shares of variable renewables. Particularly, cold Dunkelflauten with coinciding low renewable energy generation and high heat demand are critical. To mitigate the risk of such critical conditions, energy storage systems are an important component in the future energy system. The appropriate dimensioning of energy storage requires knowledge about the typical duration and frequency of Dunkelflauten.

This study presents a climatological analysis of Dunkelflauten that would have occurred in Germany from 1979 until 2018, with the assumption of a 100% renewable energy scenario. On average four Dunkelflauten are detected per year, with durations of up to nine days. To forecast the chances of a Dunkelflauten event several weeks ahead of time, it is helpful to unveil the long-lasting weather conditions, called weather regimes, during which Dunkelflauten occur. We show that Dunkelflauten are more likely to occur during blocked regimes than during cyclonic regimes. The particularly critical cold Dunkelflaute occurs predominantly during the Greenland blocking. Accordingly, accurate predictions of weather regimes are key to manage energy storage systems and to eventually avoid temporary shortages in energy supply.

How to cite: Mockert, F., Grams, C. M., Brown, T., Neumann, F., and Fallon, J.: Dunkelflauten in Germany: Climatology and Relation to Weather Regimes, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-294, https://doi.org/10.5194/ems2022-294, 2022.

In electricity grids with high shares of solar and wind, storage technologies, such as electric batteries, are suitable to complement renewable generation. Renewable generation is unequally distributed both temporally and spatially. Storage can offset the temporal volatility of wind and photovoltaic (PV) by shifting generation or load if it they do not coincide temporally. In Germany, the power system is additionally affected by a spatial mismatch between load and renewable generation. While wind power generation is predominantly located in the North of Germany, demand is centred in the more densely populated West and South.  Due to insufficient transmission capacity, North-South transmission lines are frequently overloaded in situations with high wind generation. As a result, wind generation is curtailed while fossil fired generation is used to supply loads south of the transmission bottleneck, thus jeopardizing German efforts to increase renewable generation and reducing greenhouse gases. Installing storage capacities can address this spatial mismatch to achieve a more balanced transmission line utilization, thus increasing overall transmission grid utilization, decreasing curtailment and reducing the use of fossil-fuel combustion.  To do so, the allocation of storage within the electricity grid is critical. To investigate the relationship between storage, renewable generation, and the transmission grid, we employ a linear optimization model that allows to analyse optimal allocation of storage and generation technologies within Germany. Our model simulations use high resolution reanalysis meteorological data from the COSMO-REA6 data set developed by the Hans-Ertel-Centre for Weather Research, Climate Monitoring and Diagnostics, and a detailed depiction of electricity grid constraints of Germany. We find that an allocation of storage close to grid bottlenecks is optimal to facilitate wind generation replacing fossil fired generation, thus making efficient use of renewable generation. If grid bottlenecks are not accounted for in the storage allocation, storage cannot be used to relieve grid congestion. Consequently, we find an increase of up to 10% in the overall system operational costs and higher curtailment rates for wind and solar power compared to having storage installed at optimal locations.

How to cite: Czock, B., Sitzmann, A., Zinke, J., and Fiedler, S.: The place beyond the lines - Efficient storage allocation in a spatially unbalanced power system with a high share of renewables, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-196, https://doi.org/10.5194/ems2022-196, 2022.

The use of marine energies from waves and tides have been discussed since long, however sustained application is rare. Nevertheless, in recent years systems harnessing tidal streams in the MW scale have started commercial operation after a year-long test (OES 2019).

Tidal stream flows have – as compared to other renewable energy flows - a remarkably high predictability. However, this predictable flow is characterized by distinct variabilities on various time scales that have to be negotiated when designing their application in energy systems, see e.g. Robins et al. 2015, Lewis at al. 2019, Guillou et al 2020, for discussions on the the handling of fluctuations from hourly to weekly scales. The presence of multi-year fluctuations (variations in long term mean generation) gives the task of a reasonable sizing of storage and generation capacity for integrating the tidal generation into energy supply systems.

This will be analysed for the case of the isolated supply system of the Faroe Islands, for which the integration of tidal power is foreseen for a 2030 carbon neutral scheme (see Trondheim 2021). For this, a 25-year set of predicted tidal stream data using the IOS Tidal Package (IOS 2020) on a 2 month long measured timeseries of the current from the strait of Skopunarfjørður (Niclasen and Simonsen, 2009) is applied.

Under inspection are the storage capacities necessary to assure stable generation for given generation capacities.

References

IOS 2020, https://www.dfo-mpo.gc.ca/science/data-donnees/tidal-marees/index-eng.html, as of 20.04.2020

Guillou et al 2020, Guillou N. , Neill S.P., Thiébot J., Spatio-temporal variability of tidal-stream energy in north-western Europe, Phil. Trans. R. Soc. A 378: 20190493 , 2020 http://dx.doi.org/10.1098/rsta.2019.0493

Lewis at al. 2019, Lewis M, McNaughton J. , Marquez-Dominguez C., Todeschini G.,
Togneri  M.,  Masters I., Allmark M., Stallard Y., Neill S.,
Goward-Brown A., Robins P., Power variability of tidal-stream energy and implications for electricity supply, Energy 1832019

Niclasen B. A.,  and Simonsen K., Current measurements in Skopunarfjørður, 2009, Technical Report, NVDRit 2009:12, University of the Faroe Islands, 2009.

OES 2019, https://www.ocean-energy-systems.org/publications/oes-brochures/document/tidal-current-energy-developments-highlights/ as of 20,04.2020

Robins et al. 2015, Robins P.E., Neill S.P., Lewis MJ., Ward S.L. , Characterising the spatial and temporal variability of the tidal-stream energy resource over the northwest European shelf seas, Applied Energy, 147, 20

Trondheim 2021, Tróndheim H.M.,Niclasen B., Nielsen T., Faria da Silva F. Leth Bak C, 100% Sustainable Electricity in the Faroe Islands: Expansion Planning Through Economic Optimization, IEEE open access Journal Power and Energy 8, 2021

How to cite: Beyer, H. G. and Simonsen, K.: Sizing of generation and storage capacities for tidal stream energy systems in view of the temporal pattern of tidal stream power , EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-91, https://doi.org/10.5194/ems2022-91, 2022.

The share of solar/wind energy in the electricity systems of many countries in the world will reach unprecedented values in the coming decades, fostered by the mitigation of climate change and also by the economic competitiveness of these energies. Accurate regional models of wind/solar power generation are crucial to making a smooth transition to these new energy systems.

In this work, regional solar PV and wind power models for Spain, based on the Random Forest machine learning model, were built and evaluated. Models were obtained for each of the 50 Spanish regions using as input the installed wind/solar capacity of the corresponding region and a set of meteorological variables. Regional installed capacities values were elaborated based on information from wind farms at 506 locations and photovoltaic solar plants at 3854 locations collected from public databases. The study is carried out for the year 2018 at hourly resolution; the model estimates were evaluated based on the actual power generation values provided by the Spanish TSO. This study is part of the Spanish MET4LOWCAR project, that aims at demonstrating the benefits of design low carbon power systems that accounts for the regional climatic patterns of both solar and wind renewable resources, using the Spanish territory as a testbed.

Different studies were undertaken. Firstly, two different modeling approaches were evaluated. In the first one, the meteorological inputs were derived by simple averaging the values at all the grid cells of the corresponding region. In the second approach, meteorological inputs were computed as a weighted average at the solar/wind farms locations using as weights the corresponding installed capacity. Secondly, two different meteorological databases were evaluated: the ERA-5 reanalysis and specific database derived from an ad-hoc integration conducted with the WRF NWP model. This late database has a spatial resolution of 5 km and temporal resolution of 10 minutes. Derived from these databases, up to 11 meteorological variables were used as inputs for the models The relevance of these variables in the models performance was assessed based on the feature importance provided by the Random Forest modeling procedure.

Results show, firstly, that the two modeling approaches perform similarly, being the overall power estimates relative errors of about 20% for Solar PV and 30% for the Wind power. Secondly, that the models performance using the two meteorological databases are also similar, although the WRF databases provide slightly more accurate estimates of the power variability. Thirdly, the most important meteorological input variables for the solar models were the GHI, DNI and temperature, while for the wind models were the wind at 10 and 100 meters above the ground and the wind at the 850 hPa level. The sources of the uncertainty of the power estimates are discussed.

How to cite: Aler-Mur, R., Sánchez-Hernández, G., Jiménez-Garrote, A., López-Cuesta, M., Galván-León, I., and Pozo-Vazquez, D.: Evaluation of machine-learning-based solar PV and wind power regional models for Spain, EMS Annual Meeting 2022, Bonn, Germany, 5–9 Sep 2022, EMS2022-168, https://doi.org/10.5194/ems2022-168, 2022.