Wind Speed Prediction with Spatio–Temporal Correlation: A Deep Learning Approach

Research on short-term wind speed prediction based on deep learning model in multi-fan scenario of distributed generation

• The hourly wind speed prediction is an important problem. • A feed-forward (FF) multi-layer perceptron (MLP) artificial neural network (ANN) is used for prediction. • The FF MLP ANN is optimized by 9 state-of-art metaheuristic algorithms. • In the experiments, 38 years of hourly wind data belonging to 5 cities were used. • Grey Wolf Optimizer produced best results in terms of mean square error (MSE). The growing population has tremendously increased the daily energy demand all around the world. India is the second-most crowded nation in the world with approximately 1.4 billion people. New and renewable energy is on the agenda of India and in 2021 India possesses the fourth-largest installed capacity of wind power. Accurate prediction of wind speed is vital in wind farm design and operation. In this work, an hourly wind speed prediction with an artificial neural network optimized by a metaheuristics approach is conducted. A feed-forward (FF) multi-layer perceptron (MLP) artificial neural network (ANN) is used for the prediction of the hourly wind speed. In this study, 38 years of hourly wind data belonging to 5 cities (Ambur, Hosur, Kumbakonam, Nagapattinam, and Pudukottai) were used. These cities have different specific properties such as latitude, longitude, and altitude. The FF MLP ANN is optimized by 9 state-of-art metaheuristic algorithms. In this work, Artificial Bee Colony (ABC), Ant Colony Optimization (ACO), Biogeography Based Optimization (BBO), Evolutionary Strategy (ES), Genetic Algorithm (GA), Grey-Wolf-Optimizer (GWO), Population-Based Incremental Learning (PBIL), Particle Swarm Optimization (PSO), Tree-Seed Algorithm (TSA) have been used to optimize the weights of the ANN. GWO outperforms other metaheuristic algorithms in the prediction of wind speed with a FF MLP ANN model, with a success percentage rate of approximately 3% to 10,000%.

Accurate prediction of wind speed is a prerequisite for the safe and accurate operation of wind power generation, however, WRF models typically do not produce sufficiently accurate wind speed predictions. This study proposed a Seasonal and Temporal Correlation - Deep Forest (STC-DF) model for offshore wind speed prediction. Different from traditional methods, the STC-DF model takes the advantages of the deep forest algorithm to automatically learn complex feature interactions without manual feature engineering. The model is designed to capture the seasonal and temporal characteristics of wind speed variations. To test the effectiveness of the proposed method, we applied the trained STC-DF model to an offshore wind farm in Hainan Province, China. Seven days of data from each season were selected for testing. The results show that the STC-DF model can effectively reduce the error caused by WRF forecast. The error index of the corrected wind speed reduced more than 40%, the accuracy of wind speed forecast increased 15%. And the method passed the multi-model comparison test and robustness experiment. These research results show that the STC-DF model has strong versatility and good correction ability, and is suitable for wind speed forecasting in different regions, which is a feasible method to improve the reliability of offshore wind power generation.

Taiwan, being located on a path in the west Pacific Ocean where typhoons often strike, is often affected by typhoons. The accompanying strong winds and torrential rains make typhoons particularly damaging in Taiwan. Therefore, we aimed to establish an accurate wind speed prediction model for future typhoons, allowing for better preparation to mitigate a typhoon’s toll on life and property. For more accurate wind speed predictions during a typhoon episode, we used cutting-edge machine learning techniques to construct a wind speed prediction model. To ensure model accuracy, we used, as variable input, simulated values from the Weather Research and Forecasting model of the numerical weather prediction system in addition to adopting deeper neural networks that can deepen neural network structures in the construction of estimation models. Our deeper neural networks comprise multilayer perceptron (MLP), deep recurrent neural networks (DRNNs), and stacked long short-term memory (LSTM). These three model-structure types differ by their memory capacity: MLPs are model networks with no memory capacity, whereas DRNNs and stacked LSTM are model networks with memory capacity. A model structure with memory capacity can analyze time-series data and continue memorizing and learning along the time axis. The study area is northeastern Taiwan. Results showed that MLP, DRNN, and stacked LSTM prediction error rates increased with prediction time (1–6 hours). Comparing the three models revealed that model networks with memory capacity (DRNN and stacked LSTM) were more accurate than those without memory capacity. A further comparison of model networks with memory capacity revealed that stacked LSTM yielded slightly more accurate results than did DRNN. Additionally, we determined that in the construction of the wind speed prediction model, the use of numerically simulated values reduced the error rate approximately by 30%. These results indicate that the inclusion of numerically simulated values in wind speed prediction models enhanced their prediction accuracy.

This paper proposes a new combined prediction model for short-term wind speed prediction. The article uses Numerical Weather Prediction (NWP) and actual wind speed as input to the CNN_GRU model. The normalization method is used to solve the problem of the difference in magnitude between different data types. In order to extract the data characteristics between wind direction, temperature, air pressure, numerical weather forecast wind speed and actual wind speed, a continuous data matrix is constructed. The processed data set is divided into training set and test set. First, the characteristics of the data set are extracted using a Convolutional Neural Network (CNN). The fully connected layer then processes the extracted features and inputs them to the GRU network. Finally, the final predicted wind speed is obtained through the output layer. In order to avoid the gradient dispersion caused by the Sigmoid, this paper uses the Relu as the activation function of the network. The CNN_GRU model is compared with the CNN model and the continuous method under the same conditions. The results show that the proposed CNN_GRU model has the best effect in short-term wind speed prediction.

Short‐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.

Accurate wind speed prediction is of importance for long-span cross-sea bridges. To this end, data decomposition techniques are usually employed to promote accuracy of the prediction model. Since wind speed data come sequentially, real-time decomposition should be adopted. However, real-time decomposition may degrade the accuracy due to the end effect. In this paper, a novel scheme of real-time decomposition that is a combination of truncated real-time decomposition and multi-resolution data is developed. Specifically, truncated real-time decomposition firstly denoises the data and eliminates the end effect; high-resolution data are then introduced to compensate for the information loss. Further, a novel wind speed prediction model that consists of the proposed scheme and neural networks is proposed. Specifically, two gated recurrent unit neural networks are employed to extract features from the obtained original-resolution and high-resolution data, respectively, and a multi-layer perceptron is adopted to utilize extracted features and make predictions. The proposed model is validated on the monitoring wind speed data of two long-span cross-sea bridges. Specifically, the mean absolute error and the root of mean square error of the proposed model on the two datasets are 0.334, 0.445 and 0.233 and 0.316 m/s, which are smaller than benchmark models and demonstrate superiority of the proposed model.

Wind speed prediction is crucial for optimizing wind farm and power system operations. The magnitude, nonlinearity, and randomness of wind speed affect the stable operation of wind power systems. This paper proposes a wind speed combined prediction model based on a modified Gated Recurrent Unit (MGRU) and Long Short-Term Memory (LSTM). Firstly, an MGRU model is proposed by simplifying the GRU architecture into a single-gated structure, enhancing prediction accuracy while reducing computational complexity. Subsequently, the initial wind speed data and related features are fed into the MGRU and LSTM models respectively. Finally, the optimal weighted combination algorithm is utilized to determine the weight of the prediction results of the two models, and the final prediction value is obtained. The experimental results show that the MGRU-LSTM model has significantly improved the overall prediction performance with RMSE and MAPE of 0.6551, 0.0685, and time of 150.1s, respectively, which meets the wind speed prediction requirements of wind farms.

Accurate high‐resolution spatial and temporal wind speed data is critical for estimating the wind energy potential of a location. For real‐time wind speed prediction, statistical models typically depend on high‐quality (near) real‐time data from official meteorological stations to improve forecasting accuracy. Personal weather stations (PWS) offer an additional source of real‐time data and broader spatial coverage than official stations. However, they are not subject to rigorous quality control and may exhibit bias or measurement errors. This article presents a framework for incorporating PWS data into statistical models for validated official meteorological station data via a two‐stage approach. First, bias correction is performed on PWS wind speed data using reanalysis data. Second, we implement a Bayesian hierarchical spatiotemporal model that accounts for varying measurement error in the PWS data. This enables wind speed prediction across a target area, and is particularly beneficial for improving predictions in regions sparse in official monitoring stations. Our results show that including bias‐corrected PWS data improves prediction accuracy compared with using meteorological station data alone, with a 5% reduction in prediction error on average across all sites. The results are comparable with popular reanalysis products, but unlike these numerical weather models our approach is available in real‐time and offers improved uncertainty quantification.

Ultra-short-term wind speed prediction is essential for stable control of wind turbines. In this paper, a real-time wind speed prediction model based on online parameter optimization is proposed to improve the speed and accuracy of wind speed prediction simultaneously. Firstly, the extended Kalman filter is employed to address the noise interference in wind speed sequence. By virtue of adaptive parameter optimization algorithm, online updating parameters of the model in actual working conditions can be ensured. Subsequently, online learning BLS network is introduced to update the training data of the model in time. Finally, experimental results have been provided to show to effectiveness and superior of the proposed online wind speed prediction model.

Combinations of physical and statistical wind speed forecasting models are frequently used in wind speed prediction problems arising in wind farms management. Artificial neural networks can be used in these models as a final step to obtain accurate wind speed predictions. The aim of this work is to determine the potential of evolutionary product unit neural networks (EPUNNs) for improving the accuracy and interpretation of these systems. Traditional neural network and EPUNN approaches have been used to develop different wind speed prediction models. The results obtained using different EPUNN models show that the functional model and the hybrid algorithms proposed provide very accurate prediction compared with standard neural networks used to solve this regression problem. One of the main advantages of the application of these EPUNNs has been the possibility of obtaining some interpretation of the non-linear relation predicted by the model, as will be shown in real data of a wind farm in Spain.

The focus is made to develop predictor models for wind speed prediction employing the support vector machine neural models. Basically, support vector machines (SVM) is employed as classifiers, but this contribution models variant of SVM to act as predictors. A developed model of linear support vector machine (LSVM) and proximal support vector machine (PSVM) is proposed to carry out the wind speed prediction using the available real time wind farm data. In developed PSVM predictor, it is modeled for both linear PSVM predictor and non-linear PSVM predictor. The difference between the developed linear and non-linear PSVM predictor models lies in their applicability of kernel functions to perform effective wind speed prediction. The prediction application is implemented for the set of wind farm data with a wind mill height of 50 m in a manner to minimize the mean square error. The training process of the neural network algorithmic flow is done with the developed LSVM, L-PSVM (Linear PSVM) and N-PSVM (nonlinear PSVM) for predicting the wind speed in renewable energy systems. Results computed are compared with the other types of predictors to prove the effectiveness of the proposed variants of SVM predictors. The simulated results presents the effectiveness of the proposed predictors for the real time wind farm data and the applicability of the predictors for the considered datasets.

A new two-stage decomposition and integrated hybrid model for short-term wind speed prediction

As the penetration rate of wind power in the grid continues to increase, wind speed forecasting plays a crucial role in wind power generation systems. Wind speed prediction helps optimize the operation and management of wind power generation, enhancing efficiency and reliability. However, wind speed is a nonlinear and nonstationary system, and traditional statistical methods and classical intelligent algorithms struggle to cope with dynamically updating operating conditions based on sampled data. Therefore, from the perspective of optimizing intelligent algorithms, a wind speed prediction model for wind farms was researched. In this study, we propose the Deterministic Broad Learning System (DBLS) algorithm for wind farm wind speed prediction. It effectively addresses the issues of data saturation and local minima that often occur in continuous-time system modeling. To adapt to the continuous updating of sample data, we improve the sample input of the Broad Learning System (BLS) by using a fixed-width input. When new samples are added, an equivalent number of old samples is removed to maintain the same input width, ensuring the feature capture capability of the model. Additionally, we construct a dataset of wind speed samples from 10 wind farms in Gansu Province, China. Based on this dataset, we conducted comparative experiments between the DBLS and other algorithms such as Random Forest (RF), Support Vector Regression (SVR), Extreme Learning Machines (ELM), and BLS. The comparison analysis of different algorithms was conducted using Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). Among them, the DBLS algorithm exhibited the best performance. The RMSE of the DBLS ranged from 0.762 m/s to 0.776 m/s, and the MAPE of the DBLS ranged from 0.138 to 0.149.

<strong class="journal-contentHeaderColor">Abstract.</strong> The main goals of a wind resource assessment (WRA) at a given site are to estimate the wind speed and annual energy production (AEP) of the planned wind turbines. Several steps are involved in going from initial wind speed estimations of specific locations to a comprehensive full-scale AEP assessment. These steps differ significantly between the chosen tool and the individuals performing the examination. The goal of this work is to compare different WRA simulation tools at the Perdig&atilde;o site in Portugal, for which a large amount of wind measurement data is available, in terms of both accuracy and costs. Results from nine different simulations from five different modellers were obtained via the "IEA Wind Task 31 Comparison metrics simulation challenge for wind resource assessment in complex terrain", consisting of a range of linear models, Reynolds-Averaged Navier-Stokes (RANS) computational fluid dynamics models and Large Eddy Simulations (LES). The wind speed and AEP prediction errors for three different met mast positions across the site were investigated and further translated into relative &ldquo;skill&rdquo; and &ldquo;cost&rdquo; scores, using a method previously developed by the authors. This allowed the most optimal simulation tool in terms of accuracy and cost to be chosen for this site. It was found that the RANS simulations achieved very high prediction accuracy at relatively low costs for both wind speed and AEP estimations. The LES simulations achieved great wind speed prediction for certain conditions, but at a much higher cost, which in turn also reduced the number of possible simulations, leading to a decrease in AEP prediction accuracy. For some of the simulations, the forest canopy was explicitly modelled, which was proven to be beneficial for wind speed predictions at lower heights above the ground, but lead to under-estimations of wind speeds at upper heights, decreasing the AEP prediction accuracy. Lastly, low correlation qualities between wind speed and AEP prediction error were found for each position, showing that accurate wind modelling is not necessarily the only important variable in the WRA process, and that all the steps must be considered.