Analysis of wind power generation and prediction using ANN: A case study

Many developing nations, such as India have embarked upon wind energy programs for areas experiencing high average wind speeds throughout the year. One of the states in India that is actively pursuing wind power generation programs is Tamil Nadu. Within this state, Muppandal area is one of the identified regions where wind farm concentration is high. Wind energy engineers are interested in studies that aim at assessing the output of wind farms, for which, artificial intelligence techniques can be usefully adapted. The present paper attempts to apply this concept for assessment of the wind energy output of wind farms in Muppandal, Tamil Nadu (India). Field data are collected from seven wind farms at this site over a period of 3 years from April 2002 to March 2005 and used for the analysis and prediction of power generation from wind farms. The model has been developed with the help of neural network methodology. It involves three input variables—wind speed, relative humidity and generation hours and one output variable-energy output of wind farms. The modeling is done using MATLAB toolbox. The model accuracy is evaluated by comparing the simulated results with the actual measured values at the wind farms and is found to be in good agreement.     

Ultra short‐term probability prediction of wind power based on LSTM network and condition normal distribution

Considering the inevitable prediction errors in the traditional point predictions of wind power, in this paper, a new ultra short‐term probability prediction method for wind power is proposed, in which the long short‐term memory (LSTM) network, wavelet decomposition (WT), and principal component analysis (PCA) are combined together for ultra short‐term probability prediction of wind power, a conditional normal distribution model that is developed to describe the uncertainty of prediction errors. First, WT and PCA are jointly used to smooth the original time series, then the point prediction model for subsequence data based on LSTM network is proposed. It is worth pointing out that the input matrix of the model includes many features, such as wind power and wind speed, which will be helpful for improving prediction performance. After optimizing the index of the ultra short‐term probability prediction interval (PI) of wind power by particle swarm optimization (PSO), the conditional normal distribution model of prediction errors is developed. Thus, the ultra short‐term PIs for wind power are obtained. Finally, based on the data of two wind farms in China, simulation results are provided to illustrate the usefulness of the proposed prediction model. It follows from those results that the proposed method can improve the accuracy of prediction, and the reliability of probability prediction for wind power is also improved.     

Comparison between the bivariate Weibull probability approach and linear regression for assessment of the long-term wind energy resource using MCP

A detailed investigation of a measure–correlate–predict (MCP) approach based on the bivariate Weibull (BW) probability distribution of wind speeds at pairs of correlated sites has been conducted. Since wind speeds are typically assumed to follow Weibull distributions, this approach has a stronger theoretical basis than widely used regression MCP techniques. Building on previous work that applied the technique to artificially generated wind data, we have used long-term (11 year) wind observations at 22 pairs of correlated UK sites. Additionally, 22 artificial wind data sets were generated from ideal BW distributions modelled on the observed data at the 22 site pairs. Comparison of the fitting efficiency revealed that significantly longer data periods were required to accurately extract the BW distribution parameters from the observed data, compared to artificial wind data, due to seasonal variations. The overall performance of the BW approach was compared to standard regression MCP techniques for the prediction of the 10 year wind resource using both observed and artificially generated wind data at the 22 site pairs for multiple short-term measurement periods of 1–12 months. Prediction errors were quantified by comparing the predicted and observed values of mean wind speed, mean wind power density, Weibull shape factor and standard deviation of wind speeds at each site. Using the artificial wind data, the BW approach outperformed the regression approaches for all measurement periods. When applied to the real wind speed observations however, the performance of the BW approach was comparable to the regression approaches when using a full 12 month measurement period and generally worse than the regression approaches for shorter data periods. This suggests that real wind observations at correlated sites may differ from ideal BW distributions and hence regression approaches, which require less fitting parameters, may be more appropriate, particularly when using short measurement periods.     

Fuzzy modeling techniques and artificial neural networks to estimate annual energy output of a wind turbine

The purpose of this article is to develop a new method to estimate annual energy output for a given wind turbine in any region which should be easy to use and has satisfactory accuracy. To do this, hourly wind speeds of 25 different stations in Netherlands, output power curve of S47 wind turbine and fuzzy modeling techniques and artificial neural networks were used and a model is developed to estimate annual energy output for S47 wind turbine in different regions. Since this model has three inputs (average wind speed, standard deviation of wind speed, and air density of that region), this model is easy to use. The accuracy of this method is compared with the accuracy of conventional methods and it is shown that this new method performs better. Thereafter, we have shown that by making some small changes to this proposed model, other pitch control wind turbines could be modeled too. As an example, we have modeled E82 wind turbine based on the model developed for S47 and it is shown that this model has still satisfactory accuracy.     

A statistical analysis of wind energy potential at the eastern region of Saudi Arabia

The energy potential of wind for the eastern region of Saudi Arabia is investigated based on measurements of a complete year data at a coastal location in eastern Saudi Arabia. A suitable Weibull distribution is generated and a comparison of this model is made with the Rayleigh distribution of wind power densities. Two horizontal‐axis type of wind energy conversion systems which operate at fixed rpm are considered for the determination of the extractable wind power, and a model of quadratic power output function is used between the cut‐in speed and rated speed. It is shown that small‐scale wind energy systems are suitable in the eastern part of Saudi Arabia for power generation and irrigation purposes. Copyright © 1999 John Wiley & Sons, Ltd.     

Wind resource assessment of an area using short term data correlated to a long term data set

A method of estimating the annual wind energy potential of a selected site using short term measurements related to one year’s recorded wind data at another reference site is presented. The proposed method utilizes the 1-year measured wind speed of one site to extrapolate the annual wind speed at a new site, using an artificial neural network (ANN). In this study, concurrent measurements from target and reference sites over periods of 1-month and 2-month were used to “train” the ANN. Topographical details or other meteorological data are not required for this approach. After derivation of the simulated wind speed time series for the target site, its mean value and its corresponding Weibull distribution parameters are calculated. The derived Weibull distribution of the simulated wind speed is used to make an assessment of the annual wind energy resource in the new area with respect to a particular wind turbine model. Three pairs of measuring stations in the southwest of Ireland were examined, where the wind potential is high and technically exploitable. Analysis of the measurements showed a reasonable cross-correlation coefficient of the wind speed between the sites. Results indicate that with this method, only a short time period of wind data acquisition in a new area might provide the information required for a satisfactory assessment of the annual wind energy resource. To evaluate the accuracy of the method, simulation results of the 1-month and 2-month training periods are compared to the corresponding actual values recorded at the sites. Also, a comparison with the results of a commercial wind energy assessment software package is presented showing similar results.     

Statistical wind direction modeling for the analysis of large scale wind power generation

Understanding the effects of large‐scale wind power generation on the electric power system is growing in importance as the amount of installed generation increases. In addition to wind speed, the direction of the wind is important when considering wind farms, as the aggregate generation of the farm depends on the direction of the wind. This paper introduces the wrapped Gaussian vector autoregressive process for the statistical modeling of wind directions in multiple locations. The model is estimated using measured wind direction data from Finland. The presented methodology can be used to model new locations without wind direction measurements. This capability is tested with two locations that were left out of the estimation procedure. Through long‐term Monte Carlo simulations, the methodology is used to analyze two large‐scale wind power scenarios with different geographical distributions of installed generation. Wind generation data are simulated for each wind farm using wind direction and wind speed simulations and technical wind farm information. It is shown that, compared with only using wind speed data in simulations, the inclusion of simulated wind directions enables a more detailed analysis of the aggregate wind generation probability distribution. Copyright © 2016 John Wiley & Sons, Ltd.     

Comparative performance of AI methods for wind power forecast in Portugal

Because wind has a high volatility and the respective energy produced cannot be stored on a large scale because of excessive costs, it is of utmost importance to be able to forecast wind power generation with the highest accuracy possible. The aim of this paper is to compare 1‐h‐ahead wind power forecasts performance using artificial intelligence‐based methods, such as artificial neural networks (ANNs), adaptive neural fuzzy inference system (ANFIS), and radial basis function network (RBFN). The latter was implemented using three different learning algorithms: stochastic gradient descent (SGD), hybrid, and orthogonal least squares (OLS). The application dataset is the injected wind power in the Portuguese power systems throughout the years 2010–2014. The network architecture optimization and the learning algorithms are presented. An initial data analysis showed data seasonality; therefore, the wind power forecasts were performed according to the seasons of the year. The results showed that ANFIS was the best performer method, and ANN and RBFN‐OLS also showed strong performances. RBFN‐Hybrid and RBFN‐SGD performed poorly. In general, all methods outperformed persistence.     

Wind power forecast using neural networks: Tuning with optimization techniques and error analysis

The increased integration of wind power into the power system implies many challenges to the network operators, mainly due to the hard to predict and variability of wind power generation. Thus, an accurate wind power forecast is imperative for systems operators, aiming at an efficient and economical wind power operation and integration into the power system. This work addresses the issue of forecasting short‐term wind speed and wind power for 1 hour ahead, combining artificial neural networks (ANNs) with optimization techniques on real historical wind speed and wind power data. Levenberg‐Marquardt (LM) and particle swarm optimization (PSO) are used as training algorithms to update the weights and bias of the ANN applied to wind speed predictions. The forecasting performance produced by the proposed models are compared with each other, as well as with the benchmark persistence model. Test results show higher performance for ANN‐LM wind speed forecasting model, outperforming both ANN‐PSO and persistence. The application of ANN‐LM to wind power forecast revealed also a good performance, with an average improvement of 2.8% in relation to persistence. An innovative analysis of mean absolute percentage error (MAPE) behaviour in time and in typical days is finally offered in the paper.     

Surface wind speed reconstruction from synoptic pressure fields: machine learning versus weather regimes classification techniques

This paper tackles a problem of surface wind speed reconstruction based on synoptic‐scale meteorological fields. Specifically, two different approaches are discussed and compared: a pure Machine Learning method, formed by a Support Vector Regression and a genetic algorithm that only considers synoptic pressure as input variable, and a Weather Regimes Classification Technique, based on a k‐means clustering of the main three principal components of the geopotential height field and a simple, but efficient, linear regression between the surface pressure gradient and the observed surface wind. Both algorithms are shown to be accurate enough for wind speed reconstruction at medium latitude regions, even when there are only a few years of observations. These methodologies can also be used for filling gaps in wind speed series and, with some modifications and further research, they could be used for wind speed forecasting. The algorithms proposed are fully described and compared in this paper, and their performance has been comparatively evaluated in several real problems of wind speed reconstruction at three sites (Cabauw (The Netherlands), Capel (Wales, UK) and Kaegnes (Denmark)), obtaining excellent results in terms of wind speed reconstruction with moderate complexity in data processing and algorithms. Copyright © 2014 John Wiley & Sons, Ltd.     

Performance enhancement of the artificial neural network–based reinforcement learning for wind turbine yaw control

The yaw angle control of a wind turbine allows maximization of the power absorbed from the wind and, thus, the increment of the system efficiency. Conventionally, classical control algorithms have been used for the yaw angle control of wind turbines. Nevertheless, in recent years, advanced control strategies have been designed and implemented for this purpose. These advanced control strategies are considered to offer improved features in comparison to classical algorithms. In this paper, an advanced yaw control strategy based on reinforcement learning (RL) is designed and verified in simulation environment. The proposed RL algorithm considers multivariable states and actions, as well as the mechanical loads due to the yaw rotation of the wind turbine nacelle and rotor. Furthermore, a particle swarm optimization (PSO) and Pareto optimal front (PoF)‐based algorithm have been developed in order to find the optimal actions that satisfy the compromise between the power gain and the mechanical loads due to the yaw rotation. Maximizing the power generation and minimizing the mechanical loads in the yaw bearings in an automatic way are the objectives of the proposed RL algorithm. The data of the matrices Q (s,a) of the RL algorithm are stored as continuous functions in an artificial neural network (ANN) avoiding any quantification problem. The NREL 5‐MW reference wind turbine has been considered for the analysis, and real wind data from Salt Lake, Utah, have been used for the validation of the designed yaw control strategy via simulations with the aeroelastic code FAST.     

Forecasting and simulating wind speed in Corsica by using an autoregressive model

Alternative approaches for generating wind speed time series are discussed. The method utilized involves the use of an autoregressive process model. The model has been applied to three Mediterranean sites in Corsica and has been used to generate 3-hourly synthetic time series for these considered sites. The synthetic time series have been examined to determine their ability to preserve the statistical properties of the Corsican wind speed time series. In this context, using the main statistical characteristics of the wind speed (mean, variance, probability distribution, autocorrelation function), the data simulated are compared to experimental ones in order to check whether the wind speed behavior was correctly reproduced over the studied periods. The purpose is to create a data generator in order to construct a reference year for wind systems simulation in Corsica.     

Model‐independent periodic stability analysis of wind turbines

In this work, a new method is proposed for the stability analysis of wind turbines. The method uses input–output time histories obtained by conducting virtual excitation experiments with a suitable wind turbine simulation model. Next, a single‐input/single‐output periodic reduced model is identified from the recorded response and used for a stability analysis conducted according to the Floquet theory. Since only input–output sequences are used, the approach is model independent in the sense that it is applicable to wind turbine simulation models of arbitrary complexity. The use of the Floquet theory reveals a much richer picture than the one obtained by widespread classical approaches based on the use of the multi‐blade coordinate transformation of Coleman. In fact, it is shown here that, for each principal mode computed by the classical approach, there are in reality infinite super‐harmonics of varying strength fanning out from the principal one at multiples of the rotor speed. The relative strength of each harmonic in a fan provides for a way of measuring how periodically one specific fan of modes behaves. The notion of super‐harmonics allows one to justify the presence of peaks in the response spectra, peaks that cannot be explained by the classical time‐invariant analysis. The Campbell diagram, i.e., the plot of system frequencies vs. rotor speed, is in this work enriched by the presence of the super‐harmonics, revealing a much more complex pattern of possible resonant conditions with the per‐rev excitations than normally assumed. Copyright © 2014 John Wiley & Sons, Ltd.     

基于TPNT含风电场发输电组合系统可靠性评估

为克服含风电场可靠性评估中需已知风速分布函数的缺点,提出了一种基于三阶多项式正态变换(TPNT)的非序贯蒙特卡洛模拟法评估含风电场发输电系统的可靠性。在已知风速历史数据或风速分布函数的情况下,通过TPNT构建风速随机变量与标准正态分布变量的关系,进而利用标准正态分布函数的性质产生具有任意数量的具有指定相关性的风速样本,并应用于风电场接入的发输电系统可靠性计算中。通过算例分析验证了TPNT应用于发输电系统可靠性计算中的适用性。在此基础上,从风速相关性、额定容量、风资源强度和风电场位置四个角度分析了风电场接入对可靠性的影响。为含风电场发输电系统可靠性的评估提供了新思路。     

Environmental data processing by clustering methods for energy forecast and planning

This paper presents a statistical approach based on the k-means clustering technique to manage environmental sampled data to evaluate and to forecast of the energy deliverable by different renewable sources in a given site. In particular, wind speed and solar irradiance sampled data are studied in association to the energy capability of a wind generator and a photovoltaic (PV) plant, respectively. The proposed method allows the sub-sets of useful data, describing the energy capability of a site, to be extracted from a set of experimental observations belonging the considered site. The data collection is performed in Sicily, in the south of Italy, as case study. As far as the wind generation is concerned, a suitable generator, matching the wind profile of the studied sites, has been selected for the evaluation of the producible energy. With respect to the photovoltaic generation, the irradiance data have been taken from the acquisition system of an actual installation. It is demonstrated, in both cases, that the use of the k-means clustering method allows data that do not contribute to the produced energy to be grouped into a cluster, moreover it simplifies the problem of the energy assessment since it permits to obtain the desired information on energy capability by managing a reduced amount of experimental samples. In the studied cases, the proposed method permitted a reduction of the 50% of the data with a maximum discrepancy of 10% in energy estimation compared to the classical statistical approach. Therefore, the adopted k-means clustering technique represents an useful tool for an appropriate and less demanding energy forecast and planning in distributed generation systems.     

Application of a control algorithm for wind speed prediction and active power generation

The main objective of the work described in this paper is to offer a new method of prediction of wind speeds, whilst aware that the method develops predictions in time-scales that can vary from a few minutes to an hour. This is needed because wind energy generation is increasing its participation in energy distribution and has to compete with other energy sources that are not so variable in terms of generated active power. It is important to consider that active power demand can vary quite rapidly and different sources of electricity generation must be available. In the case of wind energy, wind speed predictions are an important tool to help producers make the best decisions when selling the energy produced. These decisions are crucial in the electricity market, because of the economic benefits for producers and consequently their profitability, depends on them. The algorithm presented in this paper is based on an artificial neural network and two types of wind data have been used to test the algorithm. In the first, data was collected from a not very windy area; in the second data was collected from a real wind farm located in Navarre (North of Spain), and the values vary from very low to high speeds. Although the algorithm was not tested with typical wind speed values measured on offshore wind farm applications, it can be concluded from the first set of results presented in this paper that the algorithm is valid for estimating average speed values. Finally, a generic algorithm for the active power generation of a wind farm is presented.     

Prediction of the extreme wind speed in the mixed climate region by using Monte Carlo simulation and measure‐correlate‐predict method

The extreme wind speed at an offshore location was predicted using Monte Carlo simulation (MCS) and measure‐correlate‐predict (MCP) method. The Gumbel distribution could successfully express the annual maximum wind speed of extratropical cyclone. On the other hand, the estimated extreme wind speed of tropical cyclones by analytical probability distribution shows larger uncertainty. In the mixed climate like Japan, the extreme wind speed estimated from the combined probability distribution obtained by MCP and MCS methods agrees well with the observed data as compared with the combined probability distribution obtained by the MCP method only. The uncertainty of extreme wind speed due to limited observation period of wind speed and pressure was also evaluated by the Gumbel theory and Monte Carlo simulation. As a result, it was found that the uncertainty of 50 year recurrence wind speed obtained by MCS method is considerably smaller than that obtained by MCP method in the mixed climate. Copyright © 2014 John Wiley & Sons, Ltd.     

A variance analysis of the capacity displaced by wind energy in Europe

Wind energy generation distributed all over Europe is less variable than generation from a single region. To analyse the benefits of distributed generation, the whole electrical generation system of Europe has been modelled including varying penetrations of wind power. The model is chronologically simulating the scheduling of the European power plants to cover the demand at every hour of the year. The wind power generation was modelled using wind speed measurements from 60 meteorological stations for 1 year. The distributed wind power also displaces fossil‐fuelled capacity. However, every assessment of the displaced capacity (or a capacity credit) by means of a chronological model is highly sensitive to single events. Therefore the wind time series was shifted by integer days against the load time series, and the different results were aggregated. The same set of results is shown for two other options, one where the pump storage plants are used more aggressively and the other where all German nuclear plants are shut off. NCEP/NCAR reanalysis data have been used to recreate the same averaged time series from a data set spanning 34 years. Through this it is possible to set the year studied in detail into a longer‐term context. The results are that wind energy can contribute more than 20% of the European demand without significant changes in the system and can replace conventional capacity worth about 10% of the installed wind power capacity. The long‐term reference shows that the analysed year is the worst case for wind power integration. Copyright © 2006 John Wiley &Sons, Ltd.     

A novel application of artificial neural network for wind speed estimation

Providing accurate multi-steps wind speed estimation models has increasing significance, because of the important technical and economic impacts of wind speed on power grid security and environment benefits. In this study, the combined strategies for wind speed forecasting are proposed based on an intelligent data processing system using artificial neural network (ANN). Generalized regression neural network and Elman neural network are employed to form two hybrid models. The approach employs one of ANN to model the samples achieving data denoising and assimilation and apply the other to predict wind speed using the pre-processed samples. The proposed method is demonstrated in terms of the predicting improvements of the hybrid models compared with single ANN and the typical forecasting method. To give sufficient cases for the study, four observation sites with monthly average wind speed of four given years in Western China were used to test the models. Multiple evaluation methods demonstrated that the proposed method provides a promising alternative technique in monthly average wind speed estimation.     

Wind turbine generation performance monitoring with Jaya algorithm

Wind turbine (WT) power curves effectively reflect the generation performance of WTs and depict the relationship between the wind speed and the WT power output. This paper aims at developing an effective method for learning the intrinsic representations of WT power curves, which are robust to external environmental changes. Based on the obtained representations, WT generation performance is monitored. In the proposed approach, data of the supervisory control and data acquisition (SCADA) system is employed to derive the representations. Parametric models of WT power curves are developed using the two‐parameter and four‐parameter logic models. The parameters of these model are identified via Jaya algorithm. To detect the changes of WT power curve model parameters over different time, multivariate control charts are employed. The effectiveness of the proposed WT generation performance monitoring approach is validated based on SCADA data collected from real commercial WTs.