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.     

Dynamic models of wind farms with fixed speed wind turbines

The increasing wind power penetration on power systems requires the development of adequate wind farms models for representing the dynamic behaviour of wind farms on power systems. The behaviour of a wind farm can be represented by a detailed model including the modelling of all wind turbines and the wind farm electrical network. But this detailed model presents a high order model if a wind farm with high number of wind turbines is modelled and therefore the simulation time is long. The development of equivalent wind farm models enables the model order and the computation time to be reduced when the impact of wind farms on power systems is studied. In this paper, equivalent models of wind farms with fixed speed wind turbines are proposed by aggregating wind turbines into an equivalent wind turbine that operates on an equivalent wind farm electrical network. Two equivalent wind turbines have been developed: one for aggregated wind turbines with similar winds, and another for aggregated wind turbines under any incoming wind, even with different incoming winds.The proposed equivalent models provide high accuracy for representing the dynamic response of wind farm on power system simulations with an important reduction of model order and simulation time compare to that of the complete wind farm modelled by the detailed model.     

A simple analytical wind park model considering atmospheric stability

The analytical top‐down wind park model by Emeis and Frandsen 1 is enhanced by consistently making both the downward momentum flux and the momentum loss at the rough surface dependent on atmospheric stability. Specifying the surface roughness underneath the turbines in a wind farm in the model gives the opportunity to investigate principal differences between onshore and offshore wind parks, because the roughness length of the sea surface is two to three orders of magnitude lower than the roughness length of land surfaces. Implications for the necessary distance between single turbines in offshore wind farms and the distance between neighbouring wind parks are computed. It turns out from the model simulations that over smooth surfaces offshore the wind speed reduction at hub height in a wind farm is larger than over rough onshore surfaces given the same density of turbines within the park. Mean wind profiles within the park are also calculated from this model. Offshore wind farms must have a larger distance between each other in order to avoid shadowing effects of the upstream farm. Copyright © 2009 John Wiley & Sons, Ltd.     

Effective short-cut modelling of wind park efficiency

A short-cut model suitable for the estimation of the overall wind park efficiency was developed. Its parameters were estimated by fitting the predictions of a detailed explicit kinematic wake model for a wide range of wind velocities, number of turbines utilized, relative distances of turbines and various types of commercial wind turbines. An empirical model was used to represent the turbine power curve. The turbine grid adopted was the uniform one, where each individual machine is equally placed between its neighbours. The proposed short-cut model represented the full model in a quite satisfactory way and it was presented in a suitable formulation for dealing with wind park preliminary design problems. The effect of wind direction was also taken into consideration and was properly analysed.     

Deep learning with knowledge transfer for explainable anomaly prediction in wind turbines

The last decade has witnessed an increased interest in applying machine learning techniques to predict faults and anomalies in the operation of wind turbines. These efforts have lately been dominated by deep learning techniques which, as in other fields, tend to outperform traditional machine learning algorithms given sufficient amounts of training data. An important shortcoming of deep learning models is their lack of transparency—they operate as black boxes and typically do not provide rationales for their predictions, which can lead to a lack of trust in predicted outputs. In this article, a novel hybrid model for anomaly prediction in wind farms is proposed, which combines a recurrent neural network approach for accurate classification with an XGBoost decision tree classifier for transparent outputs. Experiments with an offshore wind turbine show that our model achieves a classification accuracy of up to 97%. The model is further able to generate detailed feature importance analyses for any detected anomalies, identifying exactly those components in a wind turbine that contribute to an anomaly. Finally, the feasibility of transfer learning is demonstrated for the wind domain by porting our “offshore” model to an unseen dataset from an onshore wind farm. The latter model achieves an accuracy of 65% and is able to detect 85% of anomalies in the unseen domain. These results are encouraging for application to wind farms for which no training data are available, for example, because they have not been in operation for long.     

Aggregated models of permanent magnet synchronous generators wind farms

This paper evaluates the responses of three aggregated models of a wind farm consisting of variable speed permanent magnet synchronous generator wind turbines when wind fluctuations or grid disturbances occur. These responses are compared with those of the detailed wind farm model, in order to verify the effectiveness of the studied aggregation methods for this type of wind farms. The equivalent wind farm models have been developed by adapting different aggregation criteria that already exist in technical literature and had been applied to other technologies. In this work, these methods have been modified to suit them to the permanent magnet synchronous generator technology. The results show that the three aggregated models provide very similar results to the detailed model, both in the evolution of active power when fluctuations in wind speed occur, and in the active power and DC-link voltage during the two simulated voltage dips. Notably, the aggregated model with an approximate mechanical torque offers excellent results.     

基于PAM-SSD-LSTM的短期风速预测

为提高短期风速预测的准确性,提出一种基于PAM聚类、奇异谱分解（SSD）和LSTM神经网络的组合预测模型来预测短期风速,以解决上述问题。首先,为提高神经网络的学习效率,采用PAM算法对原始风速数据进行相似日聚类;其次,SSD具有抑制模态混叠和虚假分量产生的优点,使用SSD分解风速序列,提取多尺度规律;最后,由于LSTM神经网络捕捉长时间依赖的序列的波动规律的能力较强,使用LSTM神经网络对分解后的风速分量进行预测,将各分量预测值叠加得到最终预测结果。实验结果表明,基于PAM-SSD-LSTM的组合预测模型可有效提高风速短期预测的准确率。     

Power quality impact of a sea located hybrid wind park

The power quality impact of a 2.5 MW wind park has been measured and analyzed. The turbines in the park operate at variable speed at low wind speeds and at fixed speed at higher wind speeds. The harmonic current injection is low and the calculated flicker contribution is seldom above 0.25. The wind park also has the potential to maintain the voltage on the supplying grid below a certain limit, in this case 11 kV     

Wind turbines type and number choice using combinatorial optimization

The paper addresses the problem associated with the optimal wind park design. A combinatorial optimization model for wind turbines type and number choice and placement considering the given wind conditions and wind park area is developed. The wind park investment costs and the total power relation as function of wind turbines number and type are used as optimization criteria. The optimization problem is formulated as a single criterion mixed-integer nonlinear discrete combinatorial task. The different wind park conditions are introduced into optimization tasks formulation as variables relations and restrictions. Two basic wind directions cases are taken into consideration – uniform and predominant wind directions for two wind park area shapes – square and rectangular. The developed wind park design approach was tested numerically by solving of different optimization tasks formulations based on wind turbines real parameters data. The numerical testing shows the applicability of the developed optimization approach. Using it will help to find mathematically reasoned wind turbines choice as contradiction to the heuristic approaches.     

基于SSA-PSO-ANFIS的短期风速预测研究

针对风速具有强非线性的特点,提出一种奇异谱分析和改进粒子群优化自适应模糊推理系统的短期风速预测模型。该方法采用奇异谱分析将原始序列分解为趋势和谐波分量,对各分量分别建立模糊神经网络模型,最后将各分量预测结果叠加得到预测风速值。为提高预测精度,改用改进粒子群算法对自适应模糊推理系统的隶属度函数进行优化。以河北某风电场实测数据进行仿真并与传统的神经网络对比分析,结果表明将风速重构后分别预测再叠加降低了原始问题的复杂度,同时提高了预测精度,在不同时间间隔的风速序列预测中该模型显著降低了多步实时预测中的误差。     

风-水电联合优化运行分析

以风-水电联合运行后的风电场效益最大为目标,提出了将概率遍历搜索和局部弹性搜索相结合的混合遗传算法来进行风-水电联合优化运行的仿真分析.仿真结果表明风-水电联合供电不但提高了风电场的收益,同时也平滑了风电场的功率输出,这将有利于提高风电在电力系统中的份额,促进我国风电产业的发展.     

Control Algorithm for Coordinated Reactive Power Compensation in a Wind Park

The penetration level of wind energy is continuously growing, and it is especially relevant in European countries such as Denmark, Germany, and Spain. For this reason, grid codes in different countries have been recently revised, or are now under revision in order to integrate this energy in the network taking into account the security of supply. This paper is related to reactive compensation, which is one aspect usually included in these codes. On the other hand, a great number of installed wind parks are formed by fixed speed wind turbines equipped with induction generators. The typical scheme for reactive compensation in this kind of wind parks is based on capacitor banks locally controlled in each machine. This configuration makes very difficult to follow the requirements of the new grid codes. To overcome this problem, a configuration with a central controller that coordinates the actuation over all the capacitor steps in the wind park is proposed in this paper. A central controller algorithm that is based on a dynamic programming is presented and evaluated by means of simulation. At this time, the proposed scheme has been installed at the Sotavento Experimental Wind Park (Spain) and it is currently being tested.      

Robust gain scheduling baseline controller for floating offshore wind turbines

The possibility of a pitch instability for floating wind turbines, due to the blade‐pitch controller, has been discussed extensively in recent years. Contrary to many advanced multi‐input‐multi‐output controllers that have been proposed, this paper aims at a standard proportional‐integral type, only feeding back the rotor speed error. The advantage of this controller is its standard layout, equal to onshore turbines, and the clearly defined model‐based control design procedure, which can be fully automated. It is more robust than most advanced controllers because it does not require additional signals of the floating platform, which make controllers often sensitive to unmodeled dynamics. For the design of this controller, a tailored linearized coupled dynamic model of reduced order is used with a detailed representation of the hydrodynamic viscous drag. The stability margin is the main design criterion at each wind speed. This results in a gain scheduling function, which looks fundamentally different than the one of onshore turbines. The model‐based controller design process has been automated, dependent only on a given stability margin. In spite of the simple structure, the results show that the controller performance satisfies common design requirements of wind turbines, which is confirmed by a model of higher fidelity than the controller design model. The controller performance is compared against an advanced controller and the fixed‐bottom version of the same turbine, indicating clearly the different challenges of floating wind control and possible remedies.     

Estimation of target station data using satellite data and deep learning algorithms

In this study, an innovative model has been developed for wind speed estimation through the Deep Learning method using hourly wind speed data from the measurement stations of the General Directorate of Meteorology in Van and Hakkari provinces in Turkey in conjunction with simultaneous satellite images from Eumetsat. Obtained satellite images were used during the introduction of the model, while wind speed data were used at the output stage. As a result of the findings, it was found that 85% accuracy performance could be achieved to provide sufficient insight for systems that are widely established worldwide. The model, developed as a result of the study, eliminates the need to install wind measuring stations for any region on earth within the satellite field in terms of determining wind potential. Since the field of view of the Meteosat 7 satellite covers the whole of Eastern Europe, it was determined that it could predict a high rate of up to 6 hours later by the method used in image analysis. The systems to be controlled with this method will be able to examine the weather events instantly at each point in the satellite field of view and make more accurate decisions. Also, companies will be able to perform a more detailed and rapid field scan compared to existing limited methods, and reduce initial investment costs and operating costs in terms of renewable energy resources investments.     

Decision analysis on generation capacity of a wind park

The investment decision on generation capacity of a wind park is difficult when wind studies or data are neither available nor sufficient to provide adequate information for developing a wind power project. Although new measurement is possible but it is definitely time consuming. To determine the optimum capacity, decision analysis techniques are proposed in this paper to cope with uncertainties arising from wind speed distribution and power–speed characteristics. The wind speed distribution is modeled from the measured data, the Rayleigh distribution, and the Weibull distribution. The power–speed curve of a wind turbine from cut-in speed to rated speed is modeled by using linear, parabolic, cubic, and quadratic characteristics. The optimization model is formulated as a mixed-integer nonlinear programming problem. The constraints are considered as interval bounds so that a set of feasible solutions is obtained. The optimum solution can be determined by using the profit-to-cost and profit-to-area ratios as performance metrics of investment. Decision analysis rules are then applied to overcome the uncertainty problem and to refine the investment plan. The proposed procedure has been tested with the wind power project of the Electricity Generating Authority of Thailand.     

Life cycle assessment of a wind farm and related externalities

This paper concentrates on the assessment of energy and emissions related to the production and manufacture of materials for an offshore wind farm as well as a wind farm on land based on a life cycle analysis (LCA) model. In Denmark a model has been developed for life cycle assessments of different materials. The model is able to assess the energy use related to the production, transportation and manufacture of 1 kg of material. The energy use is divided into fuels used in order to estimate the emissions through the life cycle. In the paper the model and the attached assumptions are described, and the model is demonstrated for two wind farms. The externalities for the wind farms are reported, showing the importance of life cycle assessment for renewable energy technologies.     

How does market power affect the impact of large scale wind investment in 'energy only' wholesale electricity markets?

In the short run, it is well known that increasing wind penetration is likely to reduce spot market electricity prices due to the merit order effect. The long run effect is less clear because there will be a change in new capacity investment in response to the wind penetration. In this paper we examine the interaction between capacity investment, wind penetration and market power by first using a least-cost generation expansion model to simulate capacity investment with increasing amounts of wind generation, and then using a computer agent-based model to predict electricity prices in the presence of market power. We find the degree to which firms are able to exercise market power depends critically on the ratio of capacity to peak demand. For our preferred long run generation scenario we show market power increases for some periods as wind penetration increases however the merit order counteracts this with the results that prices overall remain flat. Returns to peakers increase significantly as wind penetration increases. The market power in turn leads to inefficient dispatch which is exacerbated with large amounts of wind generation.     

Nonlinear model predictive control of wind turbines using LIDAR

LIDAR systems are able to provide preview information of wind disturbances at various distances in front of wind turbines. This technology paves the way for new control concepts in wind energy such as feedforward control and model predictive control. This paper compares a nonlinear model predictive controller with a baseline controller, showing the advantages of using the wind predictions in the optimization problem to reduce wind turbine extreme and fatigue loads on tower and blades as well as to limit the pitch rates. The wind information is obtained by a detailed simulation of a LIDAR system. The controller design is evaluated and tested in a simulation environment with coherent gusts and a set of turbulent wind fields using a detailed aeroelastic model of the wind turbine over the full operation region. Results show promising load reduction up to 50% for extreme gusts and 30% for lifetime fatigue loads without negative impact on overall energy production. This controller can be considered as an upper bound for other LIDAR assisted controllers that are more suited for real time applications. Copyright © 2012 John Wiley & Sons, Ltd.     

Mitigating subsynchronous resonance and damping power system oscillation in a series compensated wind park using a novel static synchronous series compensator control algorithm

This paper presents a novel damping control algorithm for static synchronous series compensator (SSSC) in a series compensated wind park for mitigating subsynchronous resonance (SSR) and for damping power system oscillations. The sample test system, adapted from the IEEE first benchmark model on SSR replacing the synchronous generator, is employed aggregating wind park based self‐excited induction generator. Consequently, it investigates the SSR phenomena and the damping power system oscillation while integrating large wind park based on SEIG. The potential occurrence and mitigation of the SSR caused by induction generator effects as well as torsional interactions, in a series compensated wind park, are investigated. The auxiliary subsynchronous damping control loops for the SSSC based on a novel design procedure of non‐linear optimization are developed. The performance of the controller is tested in steady state operation and in response to system contingencies, taking into account the impact of short circuit ratios (SCRs). The simulation results are presented to demonstrate the capability of the controllers for mitigating the SSR, damping the power system oscillation and enhancing the transient stability margin in response to different SCRs. Copyright © 2011 John Wiley & Sons, Ltd.     

A linearized numerical model of wind‐farm flows

A fast and reasonably accurate numerical three‐dimensional wake model able to predict the flow behaviour of a wind farm over a flat terrain has been developed. The model is based on the boundary‐layer approximation of the Navier–Stokes equations, linearized around the incoming atmospheric boundary layer, with the assumption that the wind turbines provide a small perturbation to the velocity field. The linearization of the actuator‐disc theory brought additional insights that could be used to understand the behaviour, as well as the limitations, of a flow model based on linear methods: for instance, it is shown that an adjustment of the turbine's thrust coefficient is necessary in order to obtain the same wake velocity field provided by the actuator disc theory within the used linear framework. The model is here validated against two independent wind‐tunnel campaigns with a small and a large wind farm aimed at the characterization of the flow above and upstream of the farms, respectively. The developed model is, in contrary to current engineering wake models, able to account for effects occurring in the upstream flow region, thereby including more physical mechanisms than other simplified approaches. The conducted simulations (in agreement with the measurement results) show that the presence of a wind farm affects the approaching flow far more upstream than generally expected and definitely beyond the current industrial standards. Despite the model assumptions, several velocity statistics above wind farms have been properly estimated providing an insight into the transfer of momentum inside the turbine rows. Copyright © 2016 John Wiley & Sons, Ltd.