A Novel Ensemble Model Based on an Advanced Optimization Algorithm for Wind Speed Forecasting

Concerning the vision of achieving carbon neutral and peak carbon goals, wind energy is extremely important as a renewable and clean energy source. However, existing research ignores the implicit features of the data preprocessing technique and the role of the internal mechanism of the optimization...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Energies (Basel) Ročník 16; číslo 14; s. 5281
Hlavní autoři: Wang, Yukun, Zhao, Aiying, Wei, Xiaoxue, Li, Ranran
Médium: Journal Article
Jazyk:angličtina
Vydáno: Basel MDPI AG 01.07.2023
Témata:
Accuracy
Algorithms
Alternative energy
Analysis
Artificial intelligence
Buildings and facilities
Carbon
Data mining
egret population optimization algorithm
Energy industry
Energy resources
ensemble model
Forecasting
Green technology
Kalman filters
Lagrange multiplier
Machine learning
Mathematical optimization
Methods
Neural networks
Optimization algorithms
selection of parameters
Support vector machines
Time series
Wind power
wind speed forecasting
China
ISSN:1996-1073, 1996-1073
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Concerning the vision of achieving carbon neutral and peak carbon goals, wind energy is extremely important as a renewable and clean energy source. However, existing research ignores the implicit features of the data preprocessing technique and the role of the internal mechanism of the optimization algorithm, making it difficult to achieve high-accuracy prediction. To fill this gap, this study proposes a wind speed forecasting model that combines data denoising techniques, optimization algorithms, and machine learning algorithms. The model discusses the important parameters in the data decomposition technique, determines the best parameter values by comparing the model’s performance, and then decomposes and reconstructs the wind speed time series. In addition, a novel optimization algorithm is used to optimize the parameters of the machine learning algorithm using a waiting strategy and an aggressive strategy to improve the effectiveness of the model. Several control experiments were designed and implemented using 10-min wind speed data from three sites in Penglai, Shandong Province. Based on the numerical comparison results and the discussion of the proposed model, it is concluded that the developed model can obtain high accuracy and reliability of wind speed prediction in the short term relative to other comparative models and can have further applications in wind power plants.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1996-1073
1996-1073
DOI:10.3390/en16145281