Predicting the Wind

Due to increasing wind power penetration, the need for and usage of wind power prediction systems have increased. At the same time, much research has been done in this field, which has led to a significant increase in the prediction accuracy recently. With many ongoing research programs in the field of numerical weather prediction (NWP), as well as in the power output prediction models (transforming wind speed into electrical power output), one can expect further improvements in the future. For the time being, three measures are taken as best practices to reduce prediction errors: Combinations of different models can be done with power output forecast models as well as with NWP models (multimodel and multischeme approaches). Reductions in RMSE of up to 20% were shown with intelligent combinations. As expected, a shorter forecast horizon leads to lower prediction errors. However, the organization of the electricity market as well as the conventional generation pool has a large influence on the needed forecast horizon. The forecast error depends on the number of wind turbines and wind farms and their geographical spread. In Germany, typical forecast errors for representative wind farm forecasts are 10-15% RMSE of installed power, while the error for the control areas calculated from these representative wind farms is typically 6-7% and that for the whole of Germany only 5-6%. Whenever possible, aggregating wind power over a large area should be performed as it leads to significant reduction of forecast errors as well as short-term fluctuations. a large area should be performed as it leads to significant reduction of forecast errors as well as short-term fluctuations.     

Analysing wind power ramp events and improving very short-term wind power predictions by including wind speed observations

Though wind power predictions have been consistently improved in the last decade, persistent reasons for remaining uncertainties are sudden large changes in wind speed, so-called ramps. Here, we analyse the occurrence of ramp events in a wind farm in Eastern Germany and the performance of a wind power prediction tool in forecasting these events for forecasting horizons of 15 and 30 min. Results on the seasonality of ramp events and their diurnal cycle are presented for multiple ramp definition thresholds. Ramps were found to be most frequent in March and April and least frequent in November and December. For the analysis, the wind power prediction tool is fed by different wind velocity forecast products, for example, numerical weather prediction (NWP) model and measurement data. It is shown that including observational wind speed data for very short-term wind power forecasts improves the performance of the power prediction tool compared to the NWP reference, both in terms of ramp detection and in decreasing the mean absolute error between predicted and generated wind power. This improvement is enhanced during ramp events, highlighting the importance of wind observations for very short-term wind power prediction.     

Wind farm power prediction: a data‐mining approach

In this paper, models for short‐ and long‐term prediction of wind farm power are discussed. The models are built using weather forecasting data generated at different time scales and horizons. The maximum forecast length of the short‐term prediction model is 12 h, and the maximum forecast length of the long‐term prediction model is 84 h. The wind farm power prediction models are built with five different data mining algorithms. The accuracy of the generated models is analysed. The model generated by a neural network outperforms all other models for both short‐ and long‐term prediction. Two basic prediction methods are presented: the direct prediction model, whereby the power prediction is generated directly from the weather forecasting data, and the integrated prediction model, whereby the prediction of wind speed is generated with the weather data, and then the power is generated with the predicted wind speed. The direct prediction model offers better prediction performance than the integrated prediction model. The main source of the prediction error appears to be contributed by the weather forecasting data. Copyright © 2008 John Wiley & Sons, Ltd.     

考虑历史相似性加权的超短期风电功率组合预测

提出一种基于历史相似性加权的超短期风电功率组合预测方法。首先,采用数值天气预报数据、风电历史数据分别作为极限学习机、长短期记忆网络的输入特征并产生预测数据;然后,通过加权灰色关联算法提取与待预测点特征近似的历史数据,评估并校正两类预测模型的预测结果。采用美国科罗拉多州某风电场实际运行数据进行训练与验证,并使用不同加权方法进行对比。结果表明,基于历史相似性优化方法可改善预测效果,缩小预测误差分布范围,验证了该文方法的有效性。     

Predicting and presenting plausible future scenarios of wind power production from numerical weather prediction systems: a qualitative ex ante evaluation for decision making

This paper is concerned with evaluating techniques to forecast plausible future scenarios in wind power production for up to 48 h ahead, where the term scenario refers to a coherent chronological prediction including the timing, rapidity and size of large changes. Such predictions are of great interest in power systems with high regional wind penetration where a large rapid change in wind power may pose a threat to power system security. Numerous studies have evaluated wind power forecasting methods on ex post statistical measures of forecast accuracy such as root mean square error. Other work has assessed the forecast value by simulating automated decision making for bidding wind generation into particular electricity markets, and in some cases, the ex ante value of a perfect forecast has been assessed. The future, however, will always be uncertain, and decision making always takes place in an ex ante context. This paper discusses how numerical weather prediction (NWP) systems forecasts are produced, with a particular focus on uncertainty and how forecasters might visually present plausible future scenarios for wind power to electricity industry decision makers. It is difficult to quantify the ex ante value of visual wind power forecast information to the complex decision‐making process involved. Consequently, this paper explores qualitative assessments of ex ante value by proposing six desirable attributes for the techniques and the presentation of NWP forecasts to decision makers. It uses these attributes to assess four such methodologies, which include NWP ensemble methods and the recently introduced NWP spatial field approach. Copyright © 2011 John Wiley & Sons, Ltd.     

Leveraging stochastic differential equations for probabilistic forecasting of wind power using a dynamic power curve

Short‐term (hours to days) probabilistic forecasts of wind power generation provide useful information about the associated uncertainty of these forecasts. Standard probabilistic forecasts are usually issued on a per‐horizon‐basis, meaning that they lack information about the development of the uncertainty over time or the inter‐temporal correlation of forecast errors for different horizons. This information is very important for forecast end‐users optimizing time‐dependent variables or dealing with multi‐period decision‐making problems, such as the management and operation of power systems with a high penetration of renewable generation. This paper provides input to these problems by proposing a model based on stochastic differential equations that allows generating predictive densities as well as scenarios for wind power. We build upon a probabilistic model for wind speed and introduce a dynamic power curve. The model thus decomposes the dynamics of wind power prediction errors into wind speed forecast errors and errors related to the conversion from wind speed to wind power. We test the proposed model on an out‐of‐sample period of 1year for a wind farm with a rated capacity of 21MW. The model outperforms simple as well as advanced benchmarks on horizons ranging from 1 to 24h. Copyright © 2016 John Wiley & Sons, Ltd.     

Error analysis of short term wind power prediction models

The integration of wind farms in power networks has become an important problem. This is because the electricity produced cannot be preserved because of the high cost of storage and electricity production must follow market demand. Short-long-range wind forecasting over different lengths/periods of time is becoming an important process for the management of wind farms. Time series modelling of wind speeds is based upon the valid assumption that all the causative factors are implicitly accounted for in the sequence of occurrence of the process itself. Hence time series modelling is equivalent to physical modelling. Auto Regressive Moving Average (ARMA) models, which perform a linear mapping between inputs and outputs, and Artificial Neural Networks (ANNs) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS), which perform a non-linear mapping, provide a robust approach to wind power prediction. In this work, these models are developed in order to forecast power production of a wind farm with three wind turbines, using real load data and comparing different time prediction periods. This comparative analysis takes in the first time, various forecasting methods, time horizons and a deep performance analysis focused upon the normalised mean error and the statistical distribution hereof in order to evaluate error distribution within a narrower curve and therefore forecasting methods whereby it is more improbable to make errors in prediction.     

Implementation of a Model Output Statistics based on meteorological variable screening for short‐term wind power forecast

A combination of physical and statistical treatments to post‐process numerical weather predictions (NWP) outputs is needed for successful short‐term wind power forecasts. One of the most promising and effective approaches for statistical treatment is the Model Output Statistics (MOS) technique. In this study, a MOS based on multiple linear regression is proposed: the model screens the most relevant NWP forecast variables and selects the best predictors to fit a regression equation that minimizes the forecast errors, utilizing wind farm power output measurements as input. The performance of the method is evaluated in two wind farms, located in different topographical areas and with different NWP grid spacing. Because of the high seasonal variability of NWP forecasts, it was considered appropriate to implement monthly stratified MOS. In both wind farms, the first predictors were always wind speeds (at different heights) or friction velocity. When friction velocity is the first predictor, the proposed MOS forecasts resulted to be highly dependent on the friction velocity–wind speed correlation. Negligible improvements were encountered when including more than two predictors in the regression equation. The proposed MOS performed well in both wind farms, and its forecasts compare positively with an actual operative model in use at Risø DTU and other MOS types, showing minimum BIAS and improving NWP power forecast of around 15% in terms of root mean square error. Further improvements could be obtained by the implementation of a more refined MOS stratification, e.g. fitting specific equations in different synoptic situations. Copyright © 2012 John Wiley & Sons, Ltd.     

Comparison of two new short-term wind-power forecasting systems

This paper presents a comparison of two new advanced statistical short-term wind-power forecasting systems developed by two independent research teams. The input variables used in both systems were the same: forecasted meteorological variable values obtained from a numerical weather prediction model; and electric power-generation registers from the SCADA system of the wind farm. Both systems are described in detail and the forecasting results compared, revealing great similarities, although the proposed structures of the two systems are different. The forecast horizon for both systems is 72 h, allowing the use of the forecasted values in electric market operations, as diary and intra-diary power generation bid offers, and in wind-farm maintenance planning.     

Hybridizing the fifth generation mesoscale model with artificial neural networks for short-term wind speed prediction

This paper presents the hybridization of the fifth generation mesoscale model (MM5) with neural networks in order to tackle a problem of short-term wind speed prediction. The mean hourly wind speed forecast at wind turbines in a wind park is an important parameter used to predict the total power production of the park. Our model for short-term wind speed forecast integrates a global numerical weather prediction model and observations at different heights (using atmospheric soundings) as initial and boundary conditions for the MM5 model. Then, the outputs of this model are processed using a neural network to obtain the wind speed forecast in specific points of the wind park. In the experiments carried out, we present some results of wind speed forecasting in a wind park located at the south-east of Spain. The results are encouraging, and show that our hybrid MM5-neural network approach is able to obtain good short-term predictions of wind speed at specific points.     

基于奇异值分解与卡尔曼滤波修正多位置NWP的短期风电功率预测

考虑到数值天气预报网格点位置和系统误差对短期风电功率预测精度的影响,提出一种基于奇异值分解与卡尔曼滤波修正多位置数值天气预报的短期风电功率预测模型。首先通过奇异值分解对多位置数值天气预报数据进行特征提取与降维处理;然后使用卡尔曼滤波方法修正数值天气预报风速数据,降低数值天气预报的系统误差;最后基于极端随机森林算法,利用修正的数值天气预报数据搭建短期风电功率预测模型。通过对某风电场进行仿真,并与单位置、未降维、未修正模型比较,结果表明降维修正模型的预测效果最好,平均误差和均方根误差分别为7.94%和9.96%。     

Improved week-ahead predictions of wind speed using simple linear models with wavelet decomposition

Simple linear methods are widely used for time series modelling and prediction and in particular for the forecast of wind speed variations. Linear prediction models are popular for their simplicity and computational efficiency, but their prediction accuracy generally deteriorates beyond a few time steps. In this paper we demonstrate that the prediction accuracy of simple auto-regressive (AR) models can be significantly improved, by as much as 60.15% for day-ahead predictions and up to 18.25% for week-ahead predictions, when combined with suitable time series decomposition. The comparison with new reference forecast model (NRFM) also shows similar accuracy gain of week ahead predictions. The combined model is capable of forecasting wind speed up to 7 days ahead with an average root mean square error less than 3 m/s. We also compare the performance of AR and f-ARIMA models in wind speed prediction and observe that the f-ARIMA model is no better than the AR model when used in combination with time series decomposition.     

运用广义回归神经网络预测风电场功率

运用广义回归神经网络对风电场出力提前了24h预测。对引入数值气象预报信息与不引人数值气象预报信息两种情况的预测结果进行了比较分析。首先,对前15d的风功率数据进行训练,通过交叉验证,建立模型,预测了未来一天的风电场出力。然后加入历史风速数据,对历史风速和风功率进行训练,利用数值气象预报信息,预测未来1d的风功率。通过算例表明,使用广义回归神经网络模型预测未来1d的风电场出力,预测结果能够跟踪实际风功率,同时加入数值气象预报信息的预测结果较不加入数值气象预报信息的神经网络预测,精度有所提高。     

Statistical post‐processing of turbulence‐resolving weather forecasts for offshore wind power forecasting

Accurate short‐term power forecasts are crucial for the reliable and efficient integration of wind energy in power systems and electricity markets. Typically, forecasts for hours to days ahead are based on the output of numerical weather prediction models, and with the advance of computing power, the spatial and temporal resolutions of these models have increased substantially. However, high‐resolution forecasts often exhibit spatial and/or temporal displacement errors, and when regarding typical average performance metrics, they often perform worse than smoother forecasts from lower‐resolution models. Recent computational advances have enabled the use of large‐eddy simulations (LESs) in the context of operational weather forecasting, yielding turbulence‐resolving weather forecasts with a spatial resolution of 100 m or finer and a temporal resolution of 30 seconds or less. This paper is a proof‐of‐concept study on the prospect of leveraging these ultra high‐resolution weather models for operational forecasting at Horns Rev I in Denmark. It is shown that temporal smoothing of the forecasts clearly improves their skill, even for the benchmark resolution forecast, although potentially valuable high‐frequency information is lost. Therefore, a statistical post‐processing approach is explored on the basis of smoothing and feature engineering from the high‐frequency signal. The results indicate that for wind farm forecasting, using information content from both the standard and LES resolution models improves the forecast accuracy, especially with a feature selection stage, compared with using the information content solely from either source.     

Detecting,categorizing and forecasting large ramps in wind farm power output using meteorological observations and WPPT

The Wind Power Prediction Tool (WPPT) has been installed in Australia for the first time, to forecast the power output from the 65MW Roaring 40s Renewable Energy P/L Woolnorth Bluff Point wind farm. This article analyses the general performance of WPPT as well as its performance during large ramps (swings) in power output. In addition to this, detected large ramps are studied in detail and categorized. WPPT combines wind speed and direction forecasts from the Australian Bureau of Meteorology regional numerical weather prediction model, MesoLAPS, with real‐time wind power observations to make hourly forecasts of the wind farm power output. The general performances of MesoLAPS and WPPT are evaluated over 1 year using the root mean square error (RMSE). The errors are significantly lower than for basic benchmark forecasts but higher than for many other WPPT installations, where the site conditions are not as complicated as Woolnorth Bluff Point. Large ramps are considered critical events for a wind power forecast for energy trading as well as managing power system security. A methodology is developed to detect large ramp events in the wind farm power data. Forty‐one large ramp events are detected over 1 year and these are categorized according to their predictability by MesoLAPS, the mechanical behaviour of the wind turbine, the power change observed on the grid and the source weather event. During these events, MesoLAPS and WPPT are found to give an RMSE only roughly equivalent to just predicting the mean (climatology forecast). Copyright © 2007 John Wiley & Sons, Ltd.     

Benefits of solar forecasting for energy imbalance markets

Short term electricity trading to balance generation and demand provides an economic opportunity to integrate larger shares of variable renewable energy sources in the power grid. Recently, many regulatory market environments are reorganized to allow short term electricity trading. This study seeks to quantify the benefits of solar forecasting for energy imbalance markets (EIM). State-of-the-art solar forecasts, covering forecast horizons ranging from 24 h to 5 min are proposed and compared against the currently used benchmark models, persistence (P) and smart persistence (SP). The implemented reforecast of numerical weather prediction time series achieves a skill of 14.5% over the smart persistence model. Using the proposed forecasts for a forecast horizon of up to 75 min for a single 1 MW power plant reduces required flexibility reserves by 21% and 16.14%, depending on the allowed trading intervals (5 and 15 min). The probability of an imbalance, caused through wrong market bids from PV solar plants, can be reduced by 19.65% and 15.12% (for 5 and 15 min trading intervals). All EIM stakeholders benefit from accurate forecasting. Previous estimates on the benefits of EIMs, based on persistence model are conservative. It is shown that the design variables regulating the market time lines, the bidding and the binding schedules, drive the benefits of forecasting.     

Adaptive post‐processing of short‐term wind forecasts for energy applications

We present a new method of reducing the error in predicted wind speed, thus enabling better management of wind energy facilities. A numerical weather prediction model, COSMO, was used to produce 48 h forecast data every day in 2008 at horizontal resolutions of 10 and 3 km. A new adaptive statistical method was applied to the model output to improve the forecast skill. The method applied corrective weights to a set of forecasts generated using several post‐processing methods. The weights were calculated based on the recent skill of the different forecasts. The resulting forecast data were compared with observed data, and skill scores were calculated to allow comparison between different post‐processing methods. The total root mean square error performance of the composite forecast is superior to that of any of the individual methods. Copyright © 2010 John Wiley & Sons, Ltd.     

Data efficient measure-correlate-predict approaches to wind resource assessment for small-scale wind energy

The feasibility of predicting the long-term wind resource at 22 UK sites using a measure-correlate-predict (MCP) approach based on just three months onsite wind speed measurements has been investigated. Three regression based techniques were compared in terms of their ability to predict the wind resource at a target site based on measurements at a nearby reference site. The accuracy of the predicted parameters of mean wind speed, mean wind power density, standard deviation of wind speeds and the Weibull shape factor was assessed, and their associated error distributions were investigated, using long-term measurements recorded over a period of 10 years. For each site, 120 wind resource predictions covering the entire data period were obtained using a sliding window approach to account for inter-annual and seasonal variations. Both the magnitude and sign of the prediction errors were found to be strongly dependent on the season used for onsite measurements. Averaged across 22 sites and all seasons, the best performing MCP approach resulted in mean absolute and percentage errors in the mean wind speed of 0.21 ms⁻¹ and 4.8% respectively, and in the mean wind power density of 11 Wm⁻² and 14%. The average errors were reduced to 3.6% in the mean wind speed and 10% in the mean wind power density when using the optimum season for onsite wind measurements. These values were shown to be a large improvement on the predictions obtained using an established semi-empirical model based on boundary layer scaling. The results indicate that the MCP approaches applied to very short onsite measurement periods have the potential to be a valuable addition to the wind resource assessment toolkit for small-scale wind developers.