Real Wind Farm Data Research Articles

A master-slave adaptive linear neuron-based approach for cost-effective use of battery energy storage systems in wind farms

Nowadays, clean and sustainable energy development is of essential concern, which justifies use of renewable energy sources. Wind energy is an important resource to provide such a demand. Due to the high costs of wind power generation compared to other renewable sources, the wind turbines should be designed in such a way that they usually operate at the highest point of their power. In this case, because of the random and alternating wind speed, the output power of the wind turbine generators and thus the windfarms fluctuate. When the capacity of windfarms is increased, the power injected from the windfarm into the grid can have significant negative effects on the power grid stability. In order to prevent these effects, it is necessary to use BESSs in windfarms. To this end, a waveform with acceptable fluctuations is considered for the windfarm output power. This waveform is compared with the output power waveform and then the difference between the two waveforms is compensated by a BESS. Proper tracking with the minimum delay reduces the required BESS capacity and thus initial investment cost for the windfarms is significantly reduced. In this paper, using real windfarm data, the conventional tracking methods for the windfarm output power waveform are analyzed and compared, first. Then, a novel tracking scheme, namely the MSA, is proposed, which is based on a two-fold master-slave adaptive linear neuron. Acquired results show that compared with the conventional methods, using the MSA scheme can reduce the costs of a 99 MW windfarm by 4.8 million dollars.

Enhancing Wind Farm Efficiency Through Active Control of the Atmospheric Boundary Layer’s Vertical Entrainment of Momentum

In contemporary wind farm design, the primary focus has traditionally been on reducing wake interference to optimize energy capture from horizontal wind flows. However, with the scaling up of wind farms, their interaction with the Atmospheric Boundary Layer (ABL) evolves, making vertical entrainment the main mechanism for the exchange of momentum and energy. This study introduces a methodical approach to augment the efficiency of large-scale offshore wind farms by actively controlling this vertical entrainment of momentum within the ABL. The strategy involves the precise engineering of advection fluxes to alter wind flow dynamics, utilizing turbines as effective vortex generators, toward a process of ”regenerative wind farming.” This setup aims to create a vorticity and vertical flux system akin to those observed in highly unstable ABLs. Expanding upon previous studies that focused on single Vertical Axis Wind Turbines (VAWTs), our research explores the implementation of multi-rotor systems equipped with lift-generating wings. These systems are designed to exert forces perpendicular to the prevailing wind direction, thus creating trailing vortices and directing the flow orthogonally for improved vertical advection. This research is part of a comprehensive investigative framework that combines experiments and multifidelity simulations. The current study extends those findings to wind farm simulations, aiming to assess the impact of ABL control on a full wind farm scale. The first part of the work validates an established analytical wind farm performance model against real wind farm data for thirty-one wind farms in the North Sea and Baltic Sea. The results confirm the predicted trend of decreased performance with increased wind farm size and density. The model is used to calculate the performance of a wind farm for varying regimes of vertical entrainment due to the creation of large-scale circulatory systems. The results are compared against 3D vortex simulations of the full wind farm in ”regenerative wind farming” mode. Our results demonstrate a notable improvement in wind speeds at the turbine hub height and the potential to double the feasible density of wind farms without compromising efficiency compared to traditional setups. These findings suggest a promising pathway towards a more sustainable and profitable future in wind energy, achieved through the strategic manipulation of ABL momentum, regenerating the energy in the wind farm.

Optimization Scheme for Energy Storage Capacity of Large Grid Connected with Wind Farm in New Power System

In this study, our foremost focus was on mitigating the significant impacts on system frequency due to power output variations from large-scale wind farms integrated into the grid. We conducted in-depth simulations using Matlab to meticulously analyze the relationship between the output power of wind turbines and wind speed. Taking into account the power features of wind turbines and the probability distribution of wind velocities, we proposed an innovative calculation method to determine the energy storage requirements to maintain consistent and stable output from wind energy systems over long durations. By rigorously testing with real wind farm data, the feasibility of this method was validated, highlighting its pivotal role as a reference in future wind power project designs.

A Hybrid Short‐Term Wind Power Forecasting Model Considering Significant Data Loss

Accurate wind power forecasting (WPF) is pivotal for the power system dominated by high penetration of renewable energy. Most forecasting techniques require sufficient data samples as a premise for achieving accurate prediction. Due to equipment faults during data collection, complete data is not always available, resulting that the forecasting accuracy is greatly diminished. To address this issue, this paper proposes a novel two‐stage hybrid forecasting approach including data restoration stage and forecasting stage. For the data restoration stage, the bidirectional long short‐term memory (Bi‐LSTM) is integrated into generative adversarial network (GAN) to recover the missing data with consideration of the complex time dynamics and correlations among heterogeneous data. To improve the prediction accuracy, the complete generated wind power sequence is decomposed into multiple time sequences with low volatility based on the enhanced variational mode decomposition (VMD). For the forecasting stage, a hybrid forecasting algorithm that combines convolutional neural network (CNN) and bidirectional gated recurrent unit (BiGRU) with an improved attention mechanism is proposed, strengthening the forecasting performance by assigning optimal weights to key features. The proposed hybrid forecasting method outperforms traditional methods based on real wind farm data with different shares of data loss from Guangxi province in China. © 2023 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

A deep-learning wind speed interval forecasting architecture based on modified scaling approach with feature ranking and two-output gated recurrent unit

Wind power is already an indispensable part of the power supply worldwide, for which wind speed forecasting is technically required as essential work to guarantee its safe and economical generation. To date, the majority of wind speed forecasting models only provide deterministic predictions, which, however, fails in accounting for the uncertainty associated with the wind speed that is critical to practice; the current few models attempting to use interval forecasting to address the uncertainty are always plagued by rigorous hypotheses along with the used interval construction approaches and poor interval quality. To fill the gap, a novel interval forecasting architecture has been developed to better capture the uncertainty. It comprises a modified scaling approach, a two-output Gated Recurrent Unit (GRU), and the efficient feature ranking (FR), getting rid of rigorous hypotheses and constructing reliable prediction intervals. The case study based on real wind farm data sheds light on the advantage of the modified scaling approach, as it outperforms many benchmark approaches and conventional scaling approaches by enhancing the coverage width criterion by 4.07% to 97.76% on average. Further embedding the FR and the deep-learning GRU helps in further narrowing the interval width and reducing the interval deviation.

Performance Improvement of Very Short-term Prediction Intervals for Regional Wind Power Based on Composite Conditional Nonlinear Quantile Regression

Accurate regional wind power prediction plays an important role in the security and reliability of power systems. For the performance improvement of very short-term prediction intervals (PIs), a novel probabilistic prediction method based on composite conditional nonlinear quantile regression (CCNQR) is proposed. First, the hierarchical clustering method based on weighted multivariate time series motifs (WMTSM) is studied to consider the static difference, dynamic difference, and meteorological difference of wind power time series. Then, the correlations are used as sample weights for the conditional linear programming (CLP) of CCNQR. To optimize the performance of PIs, a composite evaluation including the accuracy of PI coverage probability (PICP), the average width (AW), and the offsets of points outside PIs (OPOPI) is used to quantify the appropriate upper and lower bounds. Moreover, the adaptive boundary quantiles (ABQs) are quantified for the optimal performance of PIs. Finally, based on the real wind farm data, the superiority of the proposed method is verified by adequate comparisons with the conventional methods.

Interval Prediction Method for Wind Power Based on VMD-ELM/ARIMA-ADKDE

Wind power is a new energy source, for which the forecasting accuracy is related to the feasibility of various decision-making schemes for the power grid. The random, fluctuating and uncertain characteristics of wind power make accurate point forecasting very difficult and result in low practical application value. Therefore, a novel wind-power interval prediction method is proposed in this paper. First, variational model decomposition (VMD) is applied to the historical time sequence for wind power, followed by using an extreme learning machine (ELM) to establish a prediction model for the intrinsic mode function (IMF). An autoregressive integrated moving average (ARIMA) is used to establish a prediction model for the residual components, and the predicted IMF and residual components are reconstructed to predict the wind power. Finally, iterative self-organizing data analysis technique algorithm (ISODATA) are used with the error sequence to perform clustering segmentation, and adaptive diffusion kernel density estimation (ADKDE) is used to fit the probability density function for the prediction error of each segment and thereby obtain the cumulative distribution function and the final interval prediction under a given confidence level (CL). The results of a simulation based on real wind farm data from Hunan Province, China, from January to April 2020 show that the proposed method can significantly improve interval coverage.

Toward Cost-Oriented Forecasting of Wind Power Generation

Forecasting is considered to be one of the most cost-efficient solutions to integrating wind power into existing power systems. In some applications, unbiased forecasting is necessary, while in others, the forecasting value can be biased for optimal decision making. In this paper, we study optimal point forecasting problems under cost-oriented loss functions, which can lead to a forecasting process that is far more sensitive to the actual cost associated with forecasting errors. Theoretical points of optimal forecasting under different loss functions are illustrated, then a cost-oriented, boosted regression tree method is presented to formulate the optimal forecasting problem under study. Case studies using real wind farm data are conducted. A comparison between cost-oriented forecasting and traditional unbiased forecasting demonstrates the efficiency of the proposed method in maximizing benefits for the decision-making process.

Simulation based risk management for multi-objective optimal wind turbine placement using MOEA/D

Abstract Wake effect and wind uncertainty are the key factors resulting in low efficiency in wind energy extraction. Classic micro-siting approaches focus on reducing the wake effect to determine the best number and positions of the turbines. However, very little literature has addressed the issue of risk due to wind uncertainty which causes the expected production to be distantly deviated from what is actually produced. Multi-objective modeling is of particular interest due to its potential of managing risk. This paper proposes several multi-objective risk management (MORM) models which set the foundation on Monte Carlo simulation to conduct cost, benefit, and risk analyses. We develop an enhanced multi-objective evolutionary algorithm with decomposition (MOEA/D) algorithm by taking advantages of wind farm structure. The experiment result with real wind farm data shows the application differences in gauging the risks with various MORM models. The enhanced MOEA/D is compared with two state-of-the-art algorithms and the former produces the best frontier in the objective space in most of the simulations with mean absolute percentage improvement (API) of 46%. We demonstrate what-if analysis with various risk scenarios to assist the decision maker to realize his/her risk tolerance and to reach quality tradeoff decisions.

Direct Quantile Regression for Nonparametric Probabilistic Forecasting of Wind Power Generation

The fluctuation and uncertainty of wind power generation bring severe challenges to secure and economic operation of power systems. Because wind power forecasting error is unavoidable, probabilistic forecasting becomes critical to accurately quantifying the uncertainty involved in traditional point forecasts of wind power and to providing meaningful information to conduct risk management in power system operation. This paper proposes a novel direct quantile regression approach to efficiently generate nonparametric probabilistic forecasting of wind power generation combining extreme learning machine and quantile regression. Quantiles with different proportions can be directly produced via an innovatively formulated linear programming optimization model, without dependency on point forecasts. Multistep probabilistic forecasting of 10-min wind power is newly carried out based on real wind farm data from Bornholm Island in Denmark. The superiority of the proposed approach is verified through comparisons with other well-established benchmarks. The proposed approach forms a new artificial neural network-based nonparametric forecasting framework for wind power with high efficiency, reliability, and flexibility, which can be beneficial to various decision-making activities in power systems.

Deep learning based ensemble approach for probabilistic wind power forecasting

Due to the economic and environmental benefits, wind power is becoming one of the more promising supplements for electric power generation. However, the uncertainty exhibited in wind power data is generally unacceptably large. Thus, the data should be accurately evaluated by operators to effectively mitigate the risks of wind power on power system operations. Recognizing this challenge, a novel deep learning based ensemble approach is proposed for probabilistic wind power forecasting. In this approach, an advanced point forecasting method is originally proposed based on wavelet transform and convolutional neural network. Wavelet transform is used to decompose the raw wind power data into different frequencies. The nonlinear features in each frequency that are used to improve the forecast accuracy are later effectively learned by the convolutional neural network. The uncertainties in wind power data, i.e., the model misspecification and data noise, are separately identified thereafter. Consequently, the probabilistic distribution of wind power data can be statistically formulated. The proposed ensemble approach has been extensively assessed using real wind farm data from China, and the results demonstrate that the uncertainties in wind power data can be better learned using the proposed approach and that a competitive performance is obtained.

Deep belief network based deterministic and probabilistic wind speed forecasting approach

With the rapid growth of wind power penetration into modern power grids, wind speed forecasting (WSF) plays an increasingly significant role in the planning and operation of electric power and energy systems. However, the wind speed time series always exhibits nonlinear and non-stationary features, making it very difficult to be predicted accurately. Recognizing this challenge, a novel deep learning based approach is proposed for deterministic and probabilistic WSF. The approach is a hybrid of wavelet transform (WT), deep belief network (DBN) and spine quantile regression (QR). WT is employed to decompose raw wind speed data into different frequency series with better behaviors. The nonlinear features and invariant structures of each frequency are completely extracted by layer-wise pre-training based DBN. Then, the uncertainties in wind speed are statistically synthesized via the QR method. Case studies using real wind farm data from China and Australia have been presented. The comparative results demonstrate that the high-level nonlinear and non-stationary feature in the wind speed series can be learned better, and competitive performance can thus be obtained. It is therefore convinced that the proposed method has a high potential for practical applications in electric power and energy systems.

Scalable minimum fatigue control of dispatchable wind farms

As wind energy penetration increases, wind power plants may be required to regulate their power production according to the load-balancing needs of the power system. This presents an opportunity: when a wind farm tracks a power set-point, its wind turbines are free to continuously vary their power production as long as the sum of their productions meets the power demand. Here, we present an intuitive wind turbine coordination policy that uses this flexibility to minimize the aggregate fatigue load on the turbines. An important property is that the policy is scalable enough to be applied to any wind farm size. Specifically, the computational effort required to compute and reconfigure the optimal coordination policy is the same as that of a single stand-alone turbine, and the only centralized information processing needed to implement it is a single averaging operation. The efficiency of the coordination policy is illustrated in a simulation study based on real wind farm data.

Optimal Prediction Intervals of Wind Power Generation

Accurate and reliable wind power forecasting is essential to power system operation. Given significant uncertainties involved in wind generation, probabilistic interval forecasting provides a unique solution to estimate and quantify the potential impacts and risks facing system operation with wind penetration beforehand. This paper proposes a novel hybrid intelligent algorithm approach to directly formulate optimal prediction intervals of wind power generation based on extreme learning machine and particle swarm optimization. Prediction intervals with associated confidence levels are generated through direct optimization of both the coverage probability and sharpness to ensure the quality. The proposed method does not involve the statistical inference or distribution assumption of forecasting errors needed in most existing methods. Case studies using real wind farm data from Australia have been conducted. Comparing with benchmarks applied, experimental results demonstrate the high efficiency and reliability of the developed approach. It is therefore convinced that the proposed method provides a new generalized framework for probabilistic wind power forecasting with high reliability and flexibility and has a high potential of practical applications in power systems.

Remotely monitoring offshore wind turbines via ZigBee networks embedded with an advanced routing strategy

As better wind speeds are available offshore compared to on land, offshore wind power contribution in terms of electricity supplied is higher, thus more and more offshore wind turbines have been and will be deployed. However, the severe offshore conditions make it necessary to develop reliable and cost-effective real-time monitoring system when building offshore wind power farms. This paper proposes an innovative method for designing remote monitoring system for offshore wind turbines based on ZigBee wireless sensor networks. ZigBee networks carrying variety of sensors actively collect dynamic data related to the system operation status, including parameters of the mechanical unit and electrical unit as well as the operation environment. Each wind turbine itself represents a single wireless network, which sends information to remote monitoring center by GPRS module to achieve full wireless communication. To enhance the topologic efficiency and reduce the energy consumption of the networks, an optimized routing algorithm is developed. A physical system based on such method is developed. Analysis and experiment tests with real wind farm data indicate that the developed system works fairly well. The fundamental idea as studied in this work is of great value for building reliable and affordable real-time monitoring systems for wind farms (offshore and on land) with enhanced safety and efficiency.

A model predictive control approach to the problem of wind power smoothing with controlled battery storage

The aim of this study is to design a controller, based on model predictive control (MPC), to smooth the wind power output, which is generated from a wind farm, and subject to a variety of constraints on the system model. In order to employ the model predictive controller, we propose a wind power prediction system, which is used by the controller within its predictive optimization. The proposed controller is capable of smoothing wind power by utilizing inputs from our prediction system, and optimizes the maximum ramp rate requirement and also the state of the charge of the battery under practical constraints. The proposed prediction model is capable of predicting the wind power several steps ahead which is used in the optimization part of the controller. We illustrate the effectiveness of the proposed controller with a simulation example, employing real wind farm data under a variety of hard constraints.