Multi-step Wind Speed Prediction Research Articles

Multistep short‐term wind speed prediction with rank pooling and fast Fourier transformation

AbstractShort‐term wind speed prediction is essential for economical wind power utilization. The real‐world wind speed data are typically intermittent and fluctuating, presenting great challenges to existing shallow models. In this paper, we present a novel deep hybrid model for multistep wind speed prediction, namely, LR‐FFT‐RP‐MLP/LSTM (linear fast Fourier transform rank pooling multiple‐layer perceptron/long short‐term memory). Our hybrid model processes the local and global input features simultaneously. We leverage RP for the local feature extraction to capture the temporal structure while maintaining the temporal order. Besides, to understand the wind periodic patterns, we exploit FFT to extract global features and relevant frequency components in the wind speed data. The resulting local and global features are, respectively, integrated with the original data and are fed into an MLP/LSTM layer for the initial wind speed predictions. Finally, we leverage a linear regression layer to collaborate these initial predictions to produce the final wind speed prediction. The proposed hybrid model is evaluated using real wind speed data collected from 2010 to 2020, demonstrating superior forecasting capabilities when compared with state‐of‐the‐art single and hybrid models. Overall, this study presents a promising approach for improving the accuracy of wind speed forecasting.

Simultaneous forecasting of wind speed for multiple stations based on attribute-augmented spatiotemporal graph convolutional network and tree-structured parzen estimator

The global increase in energy demand and environmental concerns have made the development of wind energy increasingly important. Accurate wind speed prediction is crucial for maximizing the benefits of wind energy. In order to further improve the accuracy of multi-site wind speed prediction, this study adopts an evolutionary algorithm-based deep learning model that fully considers the spatiotemporal relationships among multiple station's wind speed data. Firstly, the mutual information (MI) method is used to select variables with stronger correlations to wind speed as auxiliary input factors. Then, an improved version of Attribute-Augmented Spatiotemporal Graph Convolutional Network (IASTGCN) is employed to process data from multiple stations, taking into account both temporal and spatial factors. Additionally, an MI-based wind speed data relationship matrix between multiple stations is calculated to replace the original distance relationship matrix in the model, enabling the model to better capture and utilize the relationships between stations. Next, the Tree-structured Parzen Estimator (TPE) is used to optimize the hyperparameters of the model. This ultimately achieves multi-site multi-step wind speed prediction. Experimental results demonstrate that the proposed model outperforms baseline models and models that only consider either temporal or spatial factors in various scenarios, exhibiting better predictive performance.

Robust multi-step wind speed forecasting based on a graph-based data reconstruction deep learning method

Against global warming, wind energy has increasingly become a stable form of power supply. Accurate prediction of wind speed is crucial for turbine control and wind farm dispatch, contributing to stable and continued wind energy utilization. However, it is very difficult to accomplish satisfactory wind speed forecasting, especially multi-step forecasting, due to the stochastic, random and volatile characteristic of wind speed series with complex fluctuations. This study is elaborated to propose a novel method based on “data graph” reconstruction and long short term memory (LSTM) network, to achieve accurate and robust multi-step wind speed forecasting. To obtain implicit correlations between wind speed series, the time series data are reconstructed into matrices like “data graphs”. To achieve better computing efficiency and forecasting accuracy, the convolutional neural network (CNN) is adopted to extract the features in the “data graphs”. Then the extracted data graph features are imported into the bidirectional-LSTM (bi-LSTM) network module, which takes the input in forward and backward directions, respectively, to extract more temporal information. For adequate performance assessment, experiments are carried out on data sets of an actual wind farm located in a mountainous region in China. The results show that the proposed method outperforms classic meta-model based and hybrid-model based methods in accuracy, stability and computing efficiency. The results reveal that the data graph reconstruction combined with the CNN is able to extract the hidden features of the wind speed time series data. The proposed CNN-bi-LSTM method effectively improve the multi-step wind speed prediction accuracy.

Multi-step wind speed prediction based on LSSVM combined with ESMD and fractional-order beetle swarm optimization

Accurate and stable wind speed prediction can alleviate the uncertain impacts of wind power generation caused by nonlinear characteristics of wind speed, and then improve the reliability of wind power. In this paper, a hybrid model for wind speed prediction based on mode decomposition, parameter optimization and basic prediction model is proposed. First, the extreme-point symmetric mode decomposition (ESMD) is employed to adaptively decompose the denoised wind speed time series into sub-sequences with different frequencies. Second, a fractional-order beetle swarm optimization (FO-BSO) for parameter optimization of the Least squares support vector machine (LSSVM) is proposed. Through benchmark functions and non-parametric statistical test, the advantages of the FO-BSO in accuracy, stability and convergence speed are verified. Subsequently, the ESMD-FO-BSO-LSSVM prediction model is established, and three groups of wind speed datasets with different sampling locations and sampling frequencies are selected for simulation experiments. The results show that the coefficient of determination of 1-step prediction of the proposed model in three datasets are 0.9856, 0.9713, 0.9940, which has 2.43%, 3.38%, 3.08% average promotion than that of 7 comparative models. And the accuracy and stability of ESMD-FO-BSO-LSSVM model in multi-step wind speed prediction have also achieved better performance than 7 competitors.

Multi-step wind speed prediction based on an improved multi-objective seagull optimization algorithm and a multi-kernel extreme learning machine

Multi-step wind speed prediction based on an improved multi-objective seagull optimization algorithm and a multi-kernel extreme learning machine

Multistep short-term wind speed forecasting using transformer

Wind power can effectively alleviate the energy crisis. However, its integration into the grid affects power quality and power grid stability. Accurate wind speed prediction is a key factor in the efficient use of wind power. Because of its intermittent and nonstationary nature, wind speed forecasting is difficult, and is the topic of much research, especially long-time multistep forecasts. In this paper, the multistep wind speed prediction problem is regarded as a sequence-to-sequence mapping problem, and a multistep wind speed prediction model based on a transformer is proposed. This model is based on an encoder–decoder architecture, where the encoder generates representations of historical wind speed sequences of any length, the decoder generates arbitrarily long future wind speed sequences, and the encoder and decoder are associated by an attention mechanism. At the same time, the encoder and decoder of Transformer are completely based on a multi-head attention mechanism. For easy modeling, a 1-dimensional original wind speed sequence is transformed to a 16-dimensional sequence by ensemble empirical mode decomposition (EEMD), and the multidimensional wind speed data are directly modeled with Transformer. We trained the model with very large-scale (19 years of data) wind speed data averaged at 10-minute intervals, and performed the evaluation over one-year wind speed data. Results show that our one-step forecast model achieved an average mean absolute error (MAE) and root mean square error (RMSE) of 0.167 and 0.221, respectively. To the best of our knowledge, our 3-, 6-, 12-, and 24-hour multistep forecast model achieves a new state of the art in wind speed forecasting, with respective MAEs of 0.243, 0.290, 0.362, and 0.453, and RMSEs of 0.326, 0.401, 0.513, and 0.651. It is believed that performance can be further improved with better model parameter optimization.

Multi-step ahead wind speed prediction based on a two-step decomposition technique and prediction model parameter optimization

Accurate and reliable wind speed prediction is essential for the exploitation and utilization of the wind energy. In this paper, a novel hybrid multi-step wind speed prediction model is proposed based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), robust local mean decomposition (RLMD), and improved whale optimization algorithm (IWOA), and long-term and short-term memory network (LSTM). CEEMDAN is utilized to decompose the wind speed series into a number of intrinsic mode functions, and the RLMD is used to the second step decompose the most non-stationary intrinsic mode function into a series of product functions. After the two-step decomposition, a group of new subsequences is formed. The long-term and short-term memory network (LSTM) prediction model is constructed for every subsequence and an improved whale optimization algorithm (IWOA) is used to optimize key parameters affecting the prediction performance of the LSTM model. And at last, the subsequence prediction results are superimposed to provide the final prediction results. The effectiveness and advancement of the proposed hybrid model is verified by employing wind speed data from two different wind farms. And according to the experimental results and comparison, it can be found that the proposed model can provide better performance than the seven compared models.

Wind speed prediction of unmanned sailboat based on CNN and LSTM hybrid neural network

Wind speed is a key factor for unmanned sailboats, and accurate prediction of wind speed is of great significance to the safety and performance of unmanned sailboats. In this study, a novel hybrid neural network scheme based on convolutional neural network (CNN) and long short-term memory (LSTM) is proposed for multi-step wind speed prediction. The scheme consists of two parts: a data processing module and a model module. We improved the grid search method to determine the selection of learning rate and input length hyperparameters. Simulations were performed on three different data sets and four types of other benchmark models were developed for comparison with the CNN-LSTM, such as recurrent neural network (RNN) and LSTM model, etc. The forecasts are evaluated by looking at the mean absolute error (MAE), root mean squared error (RMSE), correlation coefficient (CC) and R squared (R2). The evaluation metrics showed that the MAE and RMSE of CNN-LSTM are lower than the other benchmark models most of the time, while both CC and R2 are higher than the other models, which means the CNN-LSTM performs better accuracy and stability. It is accurate enough to provide a reliable wind input to the unmanned sailboat control system.

An effective wind speed prediction model combining secondary decomposition and regularised extreme learning machine optimised by cuckoo search algorithm

AbstractWind speed prediction has an important impact on the planning, economic operation and safe maintenance of wind power systems. However, the high volatility and intermittency of wind speed make it difficult to predict accurately. To improve the prediction accuracy, we developed a hybrid multistep wind speed prediction model named EWP‐CS‐RELM. In this model, a secondary decomposition technique of ensemble empirical mode decomposition (EEMD) and wavelet packet transform (WPT) is used, and it is called the EWP decomposition technique. This decomposition technique can achieve an adaptive processing of the data and accurately extract the characteristic components of the signal, avoiding the occurrence of pattern overlap and reducing the mutual interference between components. At the same time, the high and low‐frequency parts of the complex signal (component) can be decomposed into different frequency bands, and the corresponding frequency band can be selected adaptively to match the signal spectrum. The subsequence obtained after EWP decomposition is then predicted using a regularised extreme learning machine (RELM) optimised by the cuckoo search (CS) algorithm with strong global search ability to obtain the results. The hybrid prediction model is validated using four seasons of wind speed data from two wind farms in Shandong, China, and compared with seven other prediction models. Simulation results illustrate that the EWP‐CS‐RELM model outperforms the other seven models with the smallest statistical errors.

DynamicNet: A time-variant ODE network for multi-step wind speed prediction

Wind power is a new type of green energy. Though it is economical to access and gather such energy, effectively matching the energy with consumers’ demand is difficult, because of the fluctuate, intermittent and chaotic nature of wind speed. Hence, multi-step wind speed prediction becomes an important research topic. In this paper, we propose a novel deep learning method, DyanmicNet, for the problem. DynamicNet follows an encoder–decoder framework. To capture the fluctuate, intermittent and chaotic nature of wind speed, it leverages a time-variant structure to build the decoder, which is different from conventional encoder–decoder methods. In addition, a new neural block (ST-GRU-ODE) is developed, which can model the wind speed in a continuous manner by using the neural ordinary differential equation (ODE). To enhance the prediction performance, a multi-step training procedure is also put forward. Comprehensive experiments have been conducted on two real-world datasets, where wind speed is recorded in the form of two orthogonal components namely U-Wind and V-Wind. Each component can be illustrated as wind speed images. Experimental results demonstrate the effectiveness and superiority of the proposed method over state-of-the-art techniques.

A short-term wind speed prediction method utilizing novel hybrid deep learning algorithms to correct numerical weather forecasting

The accuracy of the wind speed prediction is of crucial significance for the operation and dispatch of the power grid system reasonably. However, wind speed is so random and intermittent that the accuracy of wind speed prediction always remains unsatisfactory. Moreover, the coupling relationship between other meteorological variables and wind speed in the time and frequency domains has rarely been studied. Subsequently, a hybrid wind speed prediction model based on weather research and forecasting (WRF) simulation is proposed according to a multivariate data decomposition method and deep learning algorithm optimized by an attention mechanism and a grid search algorithm. Firstly, the WRF simulation is utilized to obtain the predicted wind speed and other meteorological variables are also extracted from WRF different domains. Furthermore, the pearson correlation coefficient (PCC) method is adopted to select principal meteorological variables as the input series. Additionally, the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) method decomposes input series and historical data into respective intrinsic mode functions (IMFs). Then, a new hybrid deep learning model, combining a convolutional neural network (CNN) and a bidirectional long short-term memory network (BLSTM) optimized via an attention mechanism (AM) and a grid search method (GS), is proposed to predict the error and correct the wind speed from WRF innermost domain. Finally, the validation case study is conducted to verify the effectiveness of the proposed model. The results indicate that the proposed model outperforms other comparative models in terms of single-step and multi-step wind speed prediction. Specifically, the values of the mean absolute error (MAE), the mean absolute percentage error (MAPE), and the root mean square error (RMSE) are 0.1042 m/s, 4.63% and 0.1309 m/s after correction, decreased by 94.13%, 91.75% and 93.93%, respectively, compared to those without correction.

Wind speed behaviors feather analysis and its utilization on wind speed prediction using 3D-CNN

Aiming at the local wind speed prediction of each turbine in the wind farm, a wind speed prediction method based on feature analysis of wind speed behavior coupling the time characteristics and spatial feature is proposed, and a three-dimensional convolutional neural network (3D-CNN) multi-output wind speed prediction model with behavioral feature learning is established. Though the overall depth learning of the spatiotemporal behavior of wind speed with the temporal and spatial characteristics information being uncoupled, the function of future wind velocity and history spatiotemporal behavior features of each turbine location was established and furthermore, realized the multi-step wind speed prediction of each turbine location of the whole wind farm. The feasibilities and effectiveness of the proposed model are verified by the obtained SCADA (Supervisory Control and Data Acquisition) data from a wind farm in Hebei Province, China. When the prediction step size is 7 days, comparison results showed that the mean absolute error (MAE) and root mean squared error (RMSE) of the proposed model are 7.51 % and 0.70 %, respectively, which can conduct the wind speed prediction use spatiotemporal feature information effectively. Meanwhile, performances of the proposed model and previous existing prediction models were also compared with its superior been illuminated.

Dynamic time scan forecasting for multi-step wind speed prediction

Multi-step forecasting of wind speed time series, especially for day-ahead and longer time horizons, is still a challenging problem in the wind energy sector. In this paper, a novel analog-based methodology to perform multi-step forecasting in univariate time series, named dynamic time scan forecasting (DTSF), is presented. DTSF is a fast time series forecasting methodology for large data sets. Thus, the proposed method is optimal for forecasting renewable energy features such as wind speed, in which standard statistical and soft computing methods present limitations. A scan procedure is applied to identify similar patterns, named best matches, throughout the time series. As opposed to euclidean distance, more flexible similarity functions, using polynomial regression models, are dynamically estimated and Goodness-of-fit statistics are used to find the best matches. The observed values following the best matches and the fitted similarity functions are used to predict k-steps ahead, as well as forecasting intervals. An ensemble version of the method, named eDTSF, combines different predictions using different set of parameters thus, further improving forecasting performance. Remarkably, eDTSF achieved competitive results for multi-step forecasting of wind speed time series, even in situations of very high variability, as compared to eleven selected concurrent forecasting methods.

A blended approach incorporating TVFEMD, PSR, NNCT-based multi-model fusion and hierarchy-based merged optimization algorithm for multi-step wind speed prediction

Precise wind speed prediction plays an essential role in wind farm planning. To enhance the wind speed forecasting accuracy, a blended approach incorporating time varying filtering-based empirical mode decomposition (TVFEMD), phase space reconstruction (PSR), no negative constraint theory (NNCT)-based multi-model fusion and hierarchy-based merged optimization algorithm is proposed in this paper. Firstly, raw wind speed series is decomposed through TVFEMD into a variety of intrinsic mode functions (IMFs) to decrease its non-stationarity, which are further reconstructed by PSR into input and output vectors of forecasting model. Then, NNCT-based multi-model fusion made up of diverse single forecasting models is established to predict wind speed series, of which the weight coefficients would influence the forecasting results significantly. Subsequently, a hierarchy-based merged optimization algorithm is proposed to optimize the weight coefficients which integrates global elite opposition-based learning strategy and hierarchy-based mechanism with whale optimization algorithm and differential evolution (GHWOADE). Furthermore, the final forecasting results of the raw series are obtained by accumulating the forecasting values of all decomposing subsequences. Eventually, the effectiveness of the proposed approach is verified by comprehensive experiments on wind speed series and benchmark datasets. The corresponding results are summarized as follows: (1) data pretreatment technology based on TVFEMD can effectively reduce the non-stationarity of raw wind speed series; (2) the NNCT-based multi-model fusion strategy can make up the shortcoming that single forecasting model cannot achieve well prediction results for all decomposed subsequences; (3) the proposed GHWOADE algorithm is employed to optimize weight coefficients of corresponding forecasting models, which contributes to promoting wind speed prediction precision.

Multi-step wind speed prediction by combining a WRF simulation and an error correction strategy

The accurate prediction of wind speed is important in satisfying the demands of power grids. However, the prediction of wind speed is challenging because of its randomness and volatility, especially in multi-step cases. This study proposes a novel multi-step wind speed prediction model combining a Weather Research and Forecasting (WRF) simulation and an error correction strategy. First, the WRF model is adopted to predict the wind speed. Variational Mode Decomposition (VMD) is then employed to mine features of the predicted wind speed using the WRF model. The Principal Component Analysis (PCA) method is next used to extract the main components and remove illusive components. Using these principal components and prediction error as the training dataset, Long Short-Term Memory (LSTM) is applied for error correction. The WRF-VMD-PCA-LSTM model is thus developed for the multi-step prediction of wind speed. In a case study of a wind farm located in Sichuan Province, China, the proposed WRF-VMD-PCA-LSTM model outperforms models to which it is compared. The results reveal that the VMD-PCA method effectively extracts features hidden in the numerical WRF output. The proposed model effectively improves the accuracy of multi-step wind speed prediction.

New SVM kernel soft computing models for wind speed prediction in renewable energy applications

This paper proposes a hybrid multi-step wind speed prediction model based on combination of singular spectrum analysis (SSA), variational mode decomposition (VMD) and support vector machine (SVM) and was applied for sustainable renewable energy application. In the proposed SSA–VMD–SVM model, the SSA was applied to eliminate the noise and to approximate the signal with trend information; VMD was applied to decompose and to extract the features of input time series wind speed data into a number of sub-layers; and the SVM model with various kernel functions was adopted to predict the wind speed from each of the sub-layers, and the parameters of SVM were fine-tuned by differential evolutionary algorithm. To investigate the effectiveness of the proposed model, various prediction models are considered for comparative study, and it is demonstrated that the proposed model outperforms with better prediction accuracy.

Joint Point-Interval Prediction and Optimization of Wind Power Considering the Sequential Uncertainties of Stepwise Procedure

To support high-level wind energy utilization, wind power prediction has become a more and more attractive topic. To improve prediction accuracy and flexibility, joint point-interval prediction of wind power via a stepwise procedure is studied in this paper. Firstly, time-information-granularity (TIG) is defined for ultra-short-term wind speed prediction. Hidden features of wind speed in TIGs are extracted via principal component analysis (PCA) and classified via adaptive affinity propagation (ADAP) clustering. Then, Gaussian process regression (GPR) with joint point-interval estimation ability is adopted for stepwise prediction of the wind power, including wind speed prediction and wind turbine power curve (WTPC) modeling. Considering the sequential uncertainties of stepwise prediction, theoretical support for an uncertainty enlargement effect is deduced. Uncertainties’ transmission from single-step or receding multi-step wind speed prediction to wind power prediction is explained in detail. After that, normalized indexes for point-interval estimation performance are presented for GPR parameters’ optimization via a hybrid particle swarm optimization-differential evolution (PSO-DE) algorithm. K-fold cross validation (K-CV) is used to test the model stability. Moreover, due to the timeliness of data-driven GPR models, an evolutionary prediction mechanism via sliding time window is proposed to guarantee the required accuracy. Finally, measured data from a wind farm in northern China are acquired for validation. From the simulation results, several conclusions can be drawn: the multi-model structure has insignificant advantages for wind speed prediction via GPR; joint point-interval prediction of wind power is realizable and very reasonable; uncertainty enlargement exists for stepwise prediction of wind power while it is more significant after receding multi-step prediction of wind speed; a reasonable quantification mechanism for uncertainty is revealed and validated.

Multi-Step Wind Speed Forecasting Based On Ensemble Empirical Mode Decomposition, Long Short Term Memory Network and Error Correction Strategy

It is of great significance for wind power plant to construct an accurate multi-step wind speed prediction model, especially considering its operations and grid integration. By integrating with a data pre-processing measure, a parameter optimization algorithm and error correction strategy, a novel forecasting method for multi-step wind speed in short period is put forward in this article. In the suggested measure, the EEMD (Ensemble Empirical Mode Decomposition) is applied to extract a series of IMFs (intrinsic mode functions) from the initial wind data sequence; the LSTM (Long Short Term Memory) measure is executed as the major forecasting method for each IMF; the GRNN (general regression neural network) is executed as the secondary forecasting method to forecast error sequences for each IMF; and the BSO (Brain Storm Optimization) is employed to optimize the parameter for GRNN during the training process. To verify the validity of the suggested EEMD-LSTM-GRNN-BSO model, eight models were applied on three different wind speed sequences. The calculation outcomes reveal that: (1) the EEMD is able to boost the wind speed prediction capacity and robustness of the LSTM approach effectively; (2) the BSO based parameter optimization method is effective in finding the optimal parameter for GRNN and improving the forecasting performance for the EEMD-LSTM-GRNN model; (3) the error correction method based on the optimized GRNN promotes the forecasting accuracy of the EEMD-LSTM model significantly; and (4) compared with all models involved, the proposed EEMD-LSTM-GRNN-BSO model is proved to have the best performance in predicting the short-term wind speed sequence.

Multi-step wind speed prediction based on turbulence intensity and hybrid deep neural networks

Turbulence intensity of wind speed is a viable uncertainty quantification for wind speed. This paper develops an innovative framework for multi-step wind speed prediction using Wind Speed and Turbulence Intensity-based Recursive Neural Network. In this study, real-time turbulence intensity is measured from actual wind speed, and multi-resolution features of wind speed and turbulence intensity are deployed as input for prediction model. Ensemble recursive neural network is designed to execute prediction on multiple prediction intervals ranging from 10 min to 12 h with two different integrating strategies. Experimental results indicate that: (1) The proposed model dramatically outperforms conventional machine learning models on multi-step wind speed prediction; (2) The reliable wind speed prediction requires that the maximum time-resolution of turbulence intensity should be longer than prediction interval; (3) Turbulence intensity features involved prediction will achieves higher accuracy than the approaches that apply signal processing on raw wind speed, especially on middle-long term prediction. Therefore, this innovative scheme for multi-step wind speed prediction can be of immense utility to apply data-driven methods for accurate long-term wind speed prediction.

A combination forecasting approach applied in multistep wind speed forecasting based on a data processing strategy and an optimized artificial intelligence algorithm

Owing to the complexity and uncertainty of wind speed, accurate wind speed prediction has become a highly anticipated and challenging problem in recent years. Researchers have conducted numerous studies on wind speed prediction theory and practice; however, research on multi-step wind speed prediction remains scarce, which hinders further development in this area. To improve upon the accuracy and stability of multi-step wind speed prediction, this paper proposes a combination model based on a data preprocessing strategy, an improved optimization model, a no negative constraint theory, and several single prediction models. To improve upon forecasting performance, an improved water cycle algorithm based on a quasi-Newton algorithm is proposed to optimize the weight coefficients of the single models. In the empirical research, 10-min and 30-min wind speed data from Shandong Province in China, collected for case studies, were used to assess the comprehensive performance of the proposed combination model. Finally, we used 10-fold cross-validation and multiple error criteria to evaluate the comprehensive performance of the proposed combination model. The simulation results indicate that (a) the quasi-Newton algorithm can effectively increase the diversity of the water cycle algorithm particles, resulting in improved water cycle algorithm optimization performance; (b) the combination model exhibits superior predictive performance to a single model by taking advantage of each single model; and (c) the proposed combination model can effectively improve multi-step wind speed prediction results.