Logo: to the web site of Uppsala University

Accurate, time-resolved installed capacity data are crucial for forecasting and analysing wind-power production. The time series of installed capacity is often only approximately known in regions with rapid wind power development. Public wind power databases may only be updated yearly and installation dates may not be differentiated within wind farms. Errors in the magnitude or timing of added capacity can significantly impact the analysis of a production time series. The cumulative maximum of the power production can be used as a robust approach to estimate installed capacity but requires monotonically increasing capacity and relies on frequent high wind events. Limited information exists in the literature on advanced techniques to determine the installed capacity time series from measured data for a region. Here, we present a method to find the most probable time series of installed capacity for some measured wind power production, using a simulation of capacity factor time series and a quadratic optimization. Results show a 27.2% reduction in normalized mean absolute error when quantifying the installed capacity after a new wind farm is connected. It is also shown that the day-ahead forecast mean absolute error and root mean square error are reduced by 2.0% and 2.3%, respectively, when using the method to normalize production data before training of a forecasting model, compared to using the cumulative maximum for normalization. An advantage of the proposed method is that it can be used without the assumption of monotonically increasing installed capacity. It is also computationally cheap and thus suitable for usage in automated forecasting pipelines.

Effective energy trading requires probabilistic forecasts to quantify uncertainty and manage financial risk. In this paper, we describe our approach, which combines separate wind and solar models using a state-of-the-art stacked CatBoost framework. The effectiveness of this method was validated in the HEFTCom2024 competition, where it was the winning entry for forecasting and trading the combined generation of a 1200 MW offshore wind and 2400 MW solar portfolio in England. Key factors contributing to our success include the use of separate wind and solar power models, the incorporation of three different weather forecast datasets, no missed submissions (benchmark fills), and effective handling of a long-lasting cable issue for the offshore wind farm. Although the main focus was on the forecasting model, we also won the trading track. This is attributed mainly to our forecast accuracy, but our trading score exceeded expectations based on the trend in trading vs. the forecasting scores of co-competitors.

Output from meteorological reanalyses are used extensively in both academia and industry for modelling wind power. Recently, the first batch of the new ERA5 reanalysis was released. The main purpose of this paper is to compare the performance of ERA5 and MERRA-2 (a commonly used reanalysis today) in terms of modelling i) the aggregated wind generation in five different countries and ii) the generation for 1051 individual wind turbines in Sweden. The modelled wind power generation was compared to measurements. In conclusion, ERA5 performs better than MERRA-2 in all analysed aspects; correlations are higher, mean absolute and root mean square errors are in average around 20% lower and distributions of both hourly data and changes in hourly data are more similar to those for measurements. It is also shown that the uncertainty related to long-term correction (using one year of measurements and reanalysis data to predict the energy production during the remaining 1–5 years) is 20% lower for ERA5. In fact, using one year sample data and ERA5 gives slightly more accurate estimates than using two years of sample data and MERRA-2. Additionally, a new metric for quantifying the system size and dispersion of wind farms is proposed.

In wind integration studies, sub-hourly, load synchronous wind data are often preferable. These datasets can be generatedby a hybrid approach, combining hourly measurements or output from meteorological models with a stochastic simulationof the high-frequency fluctuations. This paper presents a method for simulating aggregated intra-hourly wind power fluc-tuations for a power system, taking into account the time-varying volatility seen in measurements. Some key elements inthe modelling were transformations to stationarity, the use of frequency domain techniques including a search for appropri-ate phase angles and an adjustment of the resulting time series in order to get correct hourly means. Generation data fromDenmark and Germany with 5 and 15 min temporal resolution were used for training models. It is shown that the distribu-tion and non-stationarity of simulated deviations from hourly means closely follow those of measurements. Power spectraldensities and step change distributions agree well. Of particular importance is that the results are good also when the train-ing and objective power systems are not the same. The computational cost is low in comparison with other approaches forgenerating high-frequency data.

We show that Swedish wind turbines constructed before 2007 lose 0.15 capacity factor percentage points per year, corresponding to a lifetime energy loss of 6%. A gradual increase of downtime accounts for around one third of the deterioration and worsened efficiency for the remaining. Although the performance loss in Sweden is considerably smaller than previously reported in the UK, it is statistically significant and calls for a revision of the industry practice for wind energy calculations. The study is based on two partly overlapping datasets, comprising 1,100 monthly and 1,300 hourly time series spanning 5–25 years each.

When performing wind integration studies, synthetic wind power forecasts are key elements. Historically, data from operational forecasting systems have been used sparsely, likely due to the high costs involved. Purely statistical methods for simulating wind power forecasts are more common,but have problems mimicking all relevant aspects of actual forecasts. Consequently, a new approach to obtain wind power forecasts for integration studies is proposed, relying on long time series of freely and globally available reforecasts. In order to produce synthetic forecasts with similar properties as operational ditto, some processing (noise addition and error reduction) is necessary. Validations with measurements from Belgium and Sweden show that the method is adequate; and distributions, correlations, autocorrelations and power spectral densities of forecast errors correspond well. Furthermore, abrupt changes when forecasts are updated and the existence of level and phase errors are reproduced. The influence from terrain complexity on error magnitude is promising, but more data is necessary for a proper validation.

The correlations between wind power generation in different countries are important for quantifying the reductions in variability when electrically interconnecting the countries. Hourly, country-wise time series of wind power output were generated for all European countries using MERRA reanalysis data. By comparing the model output with actual measurements, it is shown that this approach is appropriate for studying correlations. In order to deepen the analysis, correlation coefficients were not only computed for these time series, but also for the one hour step changes and for band-pass filtered data. The general pattern is that correlations reduce with separation distance in an exponential fashion and are highest for the long-term components (T > 4 months) and lowest for step changes and short-term components (T < 2 days). Interesting deviations from this pattern however exist. When comparing to earlier results for individual farms, the exponential decay is slower, in particular for step changes.

A previously developed model based on MERRA reanalysis data underestimates the high-frequency variability and step changes of hourly, aggregated wind power generation. The goal of this work is to restore these fluctuations. Since the volatility of the high-frequency signal varies in time, machine learning techniques were employed to predict the volatility. As predictors, derivatives of the output from the original “MERRA model” as well as empirical orthogonal functions of several meteorological variables were used. A FFT-IFFT approach, including a search algorithm for finding appropriate phase angles, was taken to generate a signal that was subsequently transformed to simulated high-frequency fluctuations using the predicted volatility. When comparing to the original MERRA model, the improved model output has a power spectral density and step change distribution in much better agreement with measurements. Moreover, the non-stationarity of the high-frequency fluctuations was captured to a large degree. The filtering and noise addition however resulted in a small increase in the RMS error.