Efficient Short-Term Wind Power Prediction Using a Novel Hybrid Machine Learning Model: LOFVT-OVMD-INGO-LSSVR

Accurate rainfall forecasting is crucial in activities such as agricultural planning, water resource management, and flood preparedness, so failure to accurately forecast rainfall might undermine these activities. This study aimed to improve the accuracy of daily rainfall prediction in Hai district, Kilimanjaro region, Tanzania, a region characterised by a bimodal rainfall pattern, by comparing rainfall prediction accuracies of Traditional Machine Learning (TML) and Deep Learning (DL) models. The 25-year daily rainfall dataset for Hai district was first pre-processed by cleaning, normalisation, and splitting into training (first 60%), validation (next 20%), and test (last 20%) sets. TML models (K-Nearest Neighbours (KNN) through its variant KNN Regressor and Support Vector Machine (SVM) through its variant Support Vector Regressor (SVR)) were developed, trained, and their performances compared with developed and trained DL models (Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU)). Each model’s test performance was evaluated using the Root Mean Squared Error (RMSE) metric. The findings revealed that the DL models, LSTM and GRU, achieved test RMSE scores of 6.35 mm and 6.49 mm, respectively, while the TML models, KNN and SVM, achieved test RMSE scores of 6.61 mm and 10.42 mm, respectively. The study demonstrated that DL models are more effective for accurate rainfall forecasting than TML models in regions with complex rainfall patterns. The insights gained will guide the development of more reliable, location-specific weather prediction systems, ultimately supporting informed decision-making in agriculture, environmental management, and flood risk reduction in Tanzania and other regions facing similar climatic challenges

Very short-term forecasting of wind power generation using hybrid deep learning model

Energy storage scheduling design on friendly grid wind power

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.

Wind energy is an emission free low-carbon energy source that has employed great advances in technology in recent years. The unpredictable and random characteristics of wind, including variables like wind speed, direction, barometric pressure and temperature present distinct challenges for precise wind power prediction. Accurate forecasts of wind power generation are crucial for the effective management of energy grids, ensuring stability, and the seamless integration of wind energy into existing power infrastructures. This study introduces a novel, Transformer-based, Dynamic context-aware power forecasting model tailored to enhance the accuracy of wind power forecasting. This model adapts the Transformer architecture, incorporating innovative modifications to tackle the complications of wind power forecasting. It improves the extraction of contextual features while simplifying the model’s structure. The results demonstrate the Normalized values of Root mean square error (NRMSE) 1.603, 2.219, 2.76 and 2.69 for Dynamic context-aware, Transformer, LSTM and GRU model respectively. Dynamic context-aware model outperforms other models achieving the lowest error rates in accuracy for prediction to provide comprehensive knowledge to decision-makers interested in wind turbines and energy optimization.

A new intelligent method based on combination of VMD and ELM for short term wind power forecasting

The rapid expansion of wind power gives rise to a number of challenges for power system operators and electricity market participants. The key operational challenge is to efficiently handle the uncertainty and variability of wind power when balancing supply and demand in ths system. In this report, we analyze how wind power forecasting can serve as an efficient tool toward this end. We discuss the current status of wind power forecasting in U.S. electricity markets and develop several methodologies and modeling tools for the use of wind power forecasting in operational decisions, from the perspectives of the system operator as well as the wind power producer. In particular, we focus on the use of probabilistic forecasts in operational decisions. Driven by increasing prices for fossil fuels and concerns about greenhouse gas (GHG) emissions, wind power, as a renewable and clean source of energy, is rapidly being introduced into the existing electricity supply portfolio in many parts of the world. The U.S. Department of Energy (DOE) has analyzed a scenario in which wind power meets 20% of the U.S. electricity demand by 2030, which means that the U.S. wind power capacity would have to reach more than 300 gigawatts (GW). The European Union is pursuing a target of 20/20/20, which aims to reduce greenhouse gas (GHG) emissions by 20%, increase the amount of renewable energy to 20% of the energy supply, and improve energy efficiency by 20% by 2020 as compared to 1990. Meanwhile, China is the leading country in terms of installed wind capacity, and had 45 GW of installed wind power capacity out of about 200 GW on a global level at the end of 2010. The rapid increase in the penetration of wind power into power systems introduces more variability and uncertainty in the electricity generation portfolio, and these factors are the key challenges when it comes to integrating wind power into the electric power grid. Wind power forecasting (WPF) is an important tool to help efficiently address this challenge, and significant efforts have been invested in developing more accurate wind power forecasts. In this report, we document our work on the use of wind power forecasting in operational decisions.

The uncertainty of wind power as the main obstacle of its integration into the power grid can be addressed by an accurate and efficient wind power forecast. Among the various wind power forecasting methods, machine learning (ML) algorithms, are recognized as a powerful wind power forecasting tool, however, their performance is highly dependent on the proper tuning of their hyperparameters. Common hyperparameter tuning methods such as grid search or random search are time-consuming, computationally expensive, and unreliable for complex models such as deep learning neural networks. Therefore, there is an urgent need for automatic methods to discover optimal hyperparameters for higher accuracy and efficiency of prediction models. In this study, a novel investigation is contributed to the field of wind power forecasting by a comprehensive comparison of three advanced techniques – Scikit-opt, Optuna, and Hyperopt – for hyperparameter optimization of Convolutional Neural Network (CNN) and Long Short-Term Memory Network (LSTM) models, a facet that, to our knowledge, has not been systematically explored in existing literature. The impact of these optimization techniques on the accuracy and efficiency of the CNN and LSTM models are assessed by comparing the root mean square error (RMSE) of the predictions and the required time to tune the models. The results show that the Optuna algorithm, using a Tree-structured Parzen Estimator (TPE) search method and Expected Improvement (EI) acquisition function, has the best efficiency for both CNN and LSTM models. In terms of accuracy, it is demonstrated that while for the CNN model all the optimization methods achieve similar performances, the LSTM model optimized by the Hyperopt algorithm, based on the annealing search method, results in the highest accuracy. In addition, for the first time in this research, the impact of the random initialization features on the performance of the forecasting models with neural networks is investigated. The proposed structures for deep learning models were examined to determine the most robust structure with the minimal sensitivity to the randomness. What we have discovered from the comparison of advanced hyperparameter optimization methods can be used by researchers to tune the time series-based forecasting models.

ABSTRACTAn accurate prediction of wind power generation is crucial for optimizing the integration of wind energy into the power grid, ensuring energy reliability. This research focuses on enhancing the accuracy of wind power generation forecasts by combining data from mesoscale and reanalysis models with Machine Learning (ML) approaches. We utilized WRF forecast data alongside ERA5 reanalysis data to estimate wind power generation for a wind farm located at Valladolid, Spain. The study evaluated the performance of ML models based on WRF and ERA5 data individually, as well as a combined model using inputs from both datasets. The hybrid model combining WRF and ERA5 data with ML resulted in a 15% improvement in root mean square error (RMSE) and a 10% increase in compared with standalone models, providing a more reliable 1‐h forecast of wind power generation. Additionally, the availability of data over time was addressed: WRF provides the advantage of projecting data into the future, whereas ERA5 offers retrospective data.

A new distributed decomposition–reconstruction–ensemble learning paradigm for short-term wind power prediction

The wind power sector is experiencing rapid growth, which creates new challenges for its electricity grid integration. Accurate wind power forecasting (WPF) is crucial for trading, balancing, and dispatching wind energy. In this paper, we examine the use of aggregated turbine- and farm-level WPFs in the Nordic energy market. The turbine-level WPFs were retrieved from a previous study, while the farm-level WPFs were developed using the same methodology, incorporating inputs from three different numerical weather predictions (NWPs) and implementing both direct and indirect forecasting approaches. In the indirect WPF approach, we explore the impact of using wind direction as an input for the wind farm-level power performance model. The different WPFs are combined into one using weights related to up-to-date forecast errors. An automated and optimized machine-learning pipeline using data from a Norwegian wind farm is used to implement the proposed forecasting methods. The indirect approach, that uses the wind-downscaling model, improves the wind speed forecast accuracy compared to raw forecasts from the relevant NWPs. Additionally, we observed that the farm-level downscaling model exhibited lower error than those developed at the turbine level. The combined use of multiple NWP sources reduced forecasting errors by 8 %–30 % for direct and indirect WPFs, respectively. Direct and indirect forecasting methods present similar performance. Finally, the aggregated turbine-level improved WPF accuracy by 10 % and 15 % for RMSE and MAE, respectively, compared to farm-level WPF.

Unlike other traditional energy resources, wind power outputs depend on natural wind resources that vary over space and time. Accurate wind power forecasting can reduce the burden of balancing energy equilibrium in electrical power systems. In this paper, we propose the short-term probabilistic forecasting of wind energy resources using the enhanced ensemble method. The enhanced ensemble forecasting methods are grouped into two main categories: temporal ensemble and spatial ensemble forecasting. The temporal ensemble forecasting is implemented by autoregressive integrated moving average with explanatory variable model, polynomial regression with time-series data, and analog ensemble for a probabilistic approach. The spatial ensemble forecasting is implemented by geostatistical model and interpolation with geographical property data. In addition, the stochastic approach, analog ensemble is applied to reduce the uncertainty in wind power forecasting and use of Numerical Weather Prediction models for accurate wind power forecasting is considered. We conduct stochastic wind power forecasting using practical data of Jeju power system and evaluate the system reliability on wind power generation variations. As a result, the proposed model shows better performances than single models, while at the same time providing probabilistic forecasts. Based on these forecasts, the grid operators can identify critical operating time points to prepare for problems that can occur in the system due to wind power variations in advance.

Spatial correlation learning based on graph neural network for medium-term wind power forecasting

Long-term wind power forecasting is a challenging endeavor that requires predictions that span years into the future. Accurate forecasting is crucial for optimizing energy production, grid integration, maintenance scheduling, and financial planning. This study attempts to first develop the long short-term memory networks (LSTM) with a seasonal wavelet transform forecasting model for practical long-term wind power forecasting problems with seasonal and regional influences on wind power and the instability of data signals. This model encapsulates wavelet transformation and seasonal decomposition, and employs LSTM for forecasting. The new prediction model adopted seasonal decompositions and two LSTMs to approach low- and high-frequency series datasets, as well as the wavelet synthesis prediction values. Furthermore, the parameters of the LSTM models are selected using stochastic optimization. For a comprehensive evaluation, the proposed LSTM with seasonal wavelet transform is compared with seven methods, including seasonal LSTM (SLSTM), wavelet LSTM (WLSTM), and the seasonal auto-regressive integrated moving average (SARIMA), back propagation neural network (BPNN), generalized regression neural network (GRNN), least square support vector regression (LSSVR), and support vector regression (SVR) were employed for long-term wind power output forecasting of wind farms. The empirical results underscore that the performance of the proposed forecasting model is better than other methods in terms of forecasting accuracy, which could efficiently provide reliable long-term predictions for long-term wind power output forecasting.

The intermittency, uncontrollability, and variability of wind power affect the economical operation and reliable delivery of the power system. To ensure the smooth integration of wind power into the grid, an accurate wind power forecast is essential. In this paper, we propose a short-term wind power prediction model based on decomposition and reconstruction to improve the accuracy of wind power forecasts. This model has the following characteristics: (i) wind power time series are decomposition into multiple components through the application of a decomposition method that combines fully integrated empirical mode decomposition with adaptive noise algorithm; (ii) the components are clustered using K-means clustering based on dynamic time warping. According to the similarity of the complexity and lag of components, they are divided into three classes; (iii) three different prediction models, including seasonal autoregressive integrated moving average model (SARIMA), long short-term memory network (LSTM) and Bi-directional long short-term memory network (BiLSTM), predict three types of components respectively. Finally, to illustrate the capability of the model, we compare its performance with three typical models. The results demonstrate that the proposed method exceeds the baseline models in regard to prediction performance.