Related references
Note: Only part of the references are listed.
Review
Energy & Fuels
Francisco Bilendo et al.
Summary: In the wind energy industry, the power curve is crucial for various key applications such as wind turbine selection, capacity factor estimation, and wind energy assessment. This paper provides a comprehensive review on power curve based applications, anomaly and fault types, data preprocessing and correction schemes, and modeling techniques. More than 100 references were compiled to assess the past, present, and future research directions in this domain.
Article
Energy & Fuels
Soyoung Park et al.
Summary: With a growing global trend to integrate wind power into energy grids, accurate wind-power forecasting is crucial. This study focuses on creating a wind-power generation-forecasting model using a machine-learning algorithm. The model, built with the gradient-boosting machine (GBM) algorithm and based on time-series data from Jeju's wind farms, achieved a best performance with an NMAE value of 5.15%. The improved accuracy of wind-power forecasting and its impact on grid security contribute to the integration of wind energy.
Article
Energy & Fuels
Pawel Piotrowski et al.
Summary: The ability to accurately forecast power generation in large wind farms is crucial for system security and economic savings. Using statistical analysis and innovative prediction methods, the study demonstrates the effectiveness of Numerical Weather Prediction (NWP) forecasts for hourly lags and the use of ensemble methods for improved accuracy. Machine learning solutions and original ensemble methods outperform traditional single methods in reducing errors in energy generation forecasts for wind farms, with the Ensemble Averaging Without Extremes method showing the lowest normalized mean absolute error. The proposed ensemble methods are applicable not only to wind energy but also to other renewable sources like hydropower and photovoltaic systems.
Article
Engineering, Multidisciplinary
Menglin Li et al.
Summary: This article proposes a wind speed correction method to improve the accuracy of short-term wind power forecast. By using a hidden Markov model, the method explores the characteristics of wind speed forecast errors and optimizes the emission probability using kernel density estimation. Experimental results show that the proposed method outperforms traditional methods in wind power forecasting.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2022)
Article
Energy & Fuels
Tatiane C. Carneiro et al.
Summary: With the rapid development of wind and solar power generation, the issue of intermittency becomes more prominent, and ensemble learning methods can improve predictions and be applied to wind and solar data from different locations. The ensemble model achieves better performance in predicting solar data from Brazil and Spain, as well as wind data from Brazil and Spain, compared to individual models.
Article
Energy & Fuels
Abdulelah Alkesaiberi et al.
Summary: This study aims to develop efficient data-driven models to accurately forecast wind power generation. By investigating different machine learning models and incorporating dynamic information, the forecasting performance has been improved. The results demonstrate that considering lagged data and more input variables can lead to better wind power prediction, and the optimized GPR and ensemble models outperform other machine learning models.
Article
Energy & Fuels
Ke Zhang et al.
Summary: Accurate short-term wind speed forecasting is crucial for the development of wind energy. In this study, a variable support segment-based wind speed forecasting model is proposed, which adaptively determines the future wind speed segment using a self-attention mechanism. Experimental results validate the accuracy of the proposed model and the performance of the modified Transformer model.
Article
Engineering, Electrical & Electronic
Zhong Zheng et al.
Summary: In this paper, a novel framework based on variational recurrent autoencoders (VRAEs) is proposed for probabilistic wind speed forecasting. The framework maximizes the likelihood of the complete wind speed sequence to better model the temporal relationship. The proposed method outperforms other benchmarking models in terms of negative CRPS* and achieves better sharpness, overall quality, and reliability of the prediction intervals (PIs).
IEEE TRANSACTIONS ON POWER SYSTEMS
(2022)
Review
Chemistry, Analytical
Lei Kou et al.
Summary: This paper discusses the development and maintenance direction of offshore wind farms in recent years. Research has shown that researching the monitoring, operation, and maintenance of offshore wind farms is of great significance and can bring multiple benefits. The paper mainly summarizes the monitoring, operation, and maintenance of offshore wind farms and proposes future research challenges and directions.
Article
Energy & Fuels
Andi A. H. Lateko et al.
Summary: This study proposes a regression-based ensemble method for day-ahead photovoltaic (PV) power forecasting, which combines multiple forecasting models to improve accuracy. The method consists of three steps: model training, creating optimal weights, and testing. The results show that the proposed method outperforms other methods in terms of mean relative error (MRE), mean absolute error (MAE), and coefficient of determination (R-2).
Review
Energy & Fuels
Xiangfei Sun et al.
Summary: This article reviews the key technologies of transient protection for offshore wind farm transmission lines, including the analysis of fault characteristics, comparison with traditional power systems, and the current research status.
FRONTIERS IN ENERGY RESEARCH
(2022)
Article
Energy & Fuels
Shahram Hanifi et al.
Summary: This paper investigates the obstacles to the integration of wind power into the power grid and how to improve the accuracy of wind power forecasting and the optimization process. By using machine learning models and an optimization algorithm, the long short-term memory (LSTM) model is tuned, and the proposed method is validated using historical data.
Article
Energy & Fuels
Zhenhua Xiong et al.
Summary: In this paper, a hybrid algorithm based on gradient descent and meta-heuristic optimization is proposed to enhance the accuracy of wind power prediction and reduce computational burden. The experimental results demonstrate that the hybrid algorithm outperforms the traditional Back Propagation (BP) algorithm in terms of accuracy, stability, and efficiency.
Article
Energy & Fuels
Guanjun Liu et al.
Summary: This study proposes a new hybrid machine learning model that combines LGB and GPR models to accurately predict wind speed and provide reliable probabilistic predictions. Experimental results show significant improvements in both point forecast accuracy and probabilistic forecast reliability.
Review
Energy & Fuels
Pawel Piotrowski et al.
Summary: This paper focuses on the statistical analysis of forecasting errors in wind generation forecasts and proposes a new error dispersion factor. It assesses the magnitude of forecasting error through quantitative and qualitative analyses and provides recommendations for preferred content in research papers addressing wind farm generation forecasts.
Article
Energy & Fuels
Ju-Yeol Ryu et al.
Summary: This study developed short-term forecasting models for wind power generation using various methods and found that the SVR model has higher accuracy. Additionally, it emphasizes the importance of data curation and weather information in wind power forecasting.
Article
Computer Science, Artificial Intelligence
S. Shobana et al.
Summary: This paper presents a novel coordinated control approach for a microgrid with hybrid AC/DC loads and energy resources, enabling appropriate distribution of power and optimization of control structure through coordination of distributed converters and DC/AC interlinking converters.
INTERNATIONAL JOURNAL OF BIO-INSPIRED COMPUTATION
(2022)
Article
Computer Science, Hardware & Architecture
Aman Samson Mogos et al.
Summary: In this article, a method for accurate five-minute wind speed prediction using machine learning algorithms is proposed, providing important support for monitoring and control of modern energy systems.
IEEE CANADIAN JOURNAL OF ELECTRICAL AND COMPUTER ENGINEERING
(2022)
Article
Energy & Fuels
Jose A. Dominguez-Navarro et al.
Summary: This paper studied the application of wavelet transform in wind speed forecasting methods and found that significant improvements were observed when using wavelet filters. The application of wavelet technique requires a thorough study of the time series in order to select the appropriate family and filter level for optimal results.
Article
Energy & Fuels
Andi A. H. Lateko et al.
Summary: The study introduces a PV power forecasting method based on a stacking ensemble model with recurrent neural network as a meta-learner, utilizing real and forecasted weather data for training and testing. By combining various statistical features and multiple neural network models for prediction, a more accurate forecasting result is achieved.
Article
Energy & Fuels
Upma Singh et al.
Summary: This study compares five optimized robust regression machine learning methods for improving the forecasting accuracy of short-term wind energy generation in Turkish wind farms. Results demonstrate that the algorithm incorporating gradient boosting machine regression shows superior forecasting performance.
Article
Engineering, Multidisciplinary
Yixiao Yu et al.
Summary: This article introduces a nonparametric probabilistic method for regional wind power forecast, which uses quantile regression neural networks (QRNN) and deep quantile regression to handle massive data. The model's performance is enhanced by applying local-connected methods and a ramp function, with test results demonstrating its effectiveness.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2021)
Article
Computer Science, Information Systems
M. S. Hossain Lipu et al.
Summary: Wind energy is growing rapidly worldwide, with AI approaches showing high precision and efficiency in wind power forecasting. Hybrid AI methods are gaining popularity for their precision and performance in this field, and this review aims to explore their applications and progress.
Article
Computer Science, Information Systems
Yi Xuan et al.
Summary: The study proposes a load forecasting method based on deep learning techniques, including a feature selection algorithm using random forest and a hybrid neural network STLF algorithm based on multi-model fusion. Experimental results show that the proposed method outperforms other models in terms of accuracy.
Article
Energy & Fuels
Yuntian Chen et al.
Summary: The study introduces a theory-guided deep-learning load forecasting model, TgDLF, which predicts future load through load ratio decomposition and data-driven models, improving accuracy and robustness.
ADVANCES IN APPLIED ENERGY
(2021)
Article
Computer Science, Information Systems
Zhiqiang Wu et al.
Summary: Accurate forecasting methods for wind and solar power are crucial for power systems. The proposed ensemble neural network model, incorporating LSTM, SVM, BP neural network, and ELM, showed the best prediction accuracy and performance compared to state-of-the-art models.
Article
Computer Science, Information Systems
Zhenhao Tang et al.
Article
Computer Science, Information Systems
Zexian Sun et al.
Article
Computer Science, Information Systems
Ningkai Tang et al.
IEEE INTERNET OF THINGS JOURNAL
(2018)
Article
Computer Science, Interdisciplinary Applications
Seyedali Mirjalili et al.
ADVANCES IN ENGINEERING SOFTWARE
(2017)
Article
Energy & Fuels
Jing Zhao et al.
Article
Energy & Fuels
Shaolong Sun et al.
Article
Energy & Fuels
Huseyin Akcay et al.
Article
Computer Science, Interdisciplinary Applications
Seyedali Mirjalili et al.
ADVANCES IN ENGINEERING SOFTWARE
(2016)
Review
Green & Sustainable Science & Technology
Ye Ren et al.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2015)
