Short-term offshore wind speed forecast by seasonal ARIMA - A comparison against GRU and LSTM

Energy, Journal Year: 2021, Volume and Issue: 227, P. 120492 - 120492

Published: March 30, 2021

Language: Английский

---

Wind power forecasting – A data-driven method along with gated recurrent neural network

Renewable Energy, Journal Year: 2020, Volume and Issue: 163, P. 1895 - 1909

Published: Oct. 28, 2020

Language: Английский

---

A Machine Learning-Based Gradient Boosting Regression Approach for Wind Power Production Forecasting: A Step towards Smart Grid Environments

Energies, Journal Year: 2021, Volume and Issue: 14(16), P. 5196 - 5196

Published: Aug. 23, 2021

In the last few years, several countries have accomplished their determined renewable energy targets to achieve future requirements with foremost aim encourage sustainable growth reduced emissions, mainly through implementation of wind and solar energy. present study, we propose compare five optimized robust regression machine learning methods, namely, random forest, gradient boosting (GBM), k-nearest neighbor (kNN), decision-tree, extra tree regression, which are applied improve forecasting accuracy short-term generation in Turkish farms, situated west Turkey, on basis a historic data speed direction. Polar diagrams plotted impacts input variables such as direction examined. Scatter curves depicting relationships between produced turbine power for all methods predicted average is compared real from help error curves. The results demonstrate superior performance algorithm incorporating regression.

Language: Английский

---

Efficient Wind Power Prediction Using Machine Learning Methods: A Comparative Study

Energies, Journal Year: 2022, Volume and Issue: 15(7), P. 2327 - 2327

Published: March 23, 2022

Wind power represents a promising source of renewable energies. Precise forecasting wind generation is crucial to mitigate the challenges balancing supply and demand in smart grid. Nevertheless, major difficulty its high fluctuation intermittent nature, making it challenging forecast. This study aims develop efficient data-driven models accurately forecast generation. Crucially, main contributions this work are listed following elements. Firstly, we investigate performance enhanced machine learning univariate time-series data. Specifically, employed Bayesian optimization (BO) optimally tune hyperparameters Gaussian process regression (GPR), Support Vector Regression (SVR) with different kernels, ensemble (ES) (i.e., Boosted trees Bagged trees) investigated their performance. Secondly, dynamic information has been incorporated construction further enhance models. introduce lagged measurements enable capturing time evolution into design considered Furthermore, more input variables (e.g., speed direction) used improve prediction Actual from three turbines France, Turkey, Kaggle verify efficiency The results reveal benefit considering data better power. also showed that optimized GPR outperformed other

Language: Английский

---

Closed-loop home energy management system with renewable energy sources in a smart grid: A comprehensive review

Journal of Energy Storage, Journal Year: 2022, Volume and Issue: 50, P. 104609 - 104609

Published: April 13, 2022

Language: Английский

---

Advanced hyperparameter optimization of deep learning models for wind power prediction

Renewable Energy, Journal Year: 2023, Volume and Issue: 221, P. 119700 - 119700

Published: Dec. 2, 2023

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

Language: Английский

---

Short-Term Load Forecasting Models: A Review of Challenges, Progress, and the Road Ahead

Energies, Journal Year: 2023, Volume and Issue: 16(10), P. 4060 - 4060

Published: May 12, 2023

Short-term load forecasting (STLF) is critical for the energy industry. Accurate predictions of future electricity demand are necessary to ensure power systems’ reliable and efficient operation. Various STLF models have been proposed in recent years, each with strengths weaknesses. This paper comprehensively reviews some models, including time series, artificial neural networks (ANNs), regression-based, hybrid models. It first introduces fundamental concepts challenges STLF, then discusses model class’s main features assumptions. The compares terms their accuracy, robustness, computational efficiency, scalability, adaptability identifies approach’s advantages limitations. Although this study suggests that ANNs may be most promising ways achieve accurate additional research required handle multiple input features, manage massive data sets, adjust shifting conditions.

Language: Английский

---

Learning based short term wind speed forecasting models for smart grid applications: An extensive review and case study

Electric Power Systems Research, Journal Year: 2023, Volume and Issue: 222, P. 109502 - 109502

Published: June 1, 2023

Language: Английский

---

CNN–LSTM–AM: A power prediction model for offshore wind turbines

Ocean Engineering, Journal Year: 2024, Volume and Issue: 301, P. 117598 - 117598

Published: March 23, 2024

Language: Английский

---

A short-term wind power forecasting method based on multivariate signal decomposition and variable selection

Applied Energy, Journal Year: 2024, Volume and Issue: 360, P. 122759 - 122759

Published: Feb. 6, 2024

Language: Английский

---