Short term wind speed prediction based on CEESMDAN and improved seagull optimization kernel extreme learning machine

Accurate wind speed predictions are crucial for the planning, operation, and energy management of wind farms. In this paper, we propose a novel wind speed prediction model, CEESMDAN-LNR-SOA-KELM. Firstly, we employ the CEESMDAN decomposition method to extract features from the original wind speed da...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Earth science informatics Jg. 18; H. 1; S. 141
Hauptverfasser: Qin, Xiwen, Yuan, Liping, Dong, Xiaogang, Zhang, Siqi, Shi, Hongyu
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Berlin/Heidelberg Springer Berlin Heidelberg 01.01.2025
Springer Nature B.V
Schlagworte:
Algorithms
Convergence
Earth and Environmental Science
Earth Sciences
Earth System Sciences
Energy management
Information Systems Applications (incl.Internet)
Machine learning
Numerical experiments
Ontology
Optimization
Parameters
Prediction models
Simulation and Modeling
Space Exploration and Astronautics
Space Sciences (including Extraterrestrial Physics
Strategy
Weather forecasting
Wind farms
Wind power
Wind speed
Seagull optimization algorithm
Wind speed prediction
Random opposition-based learning
Adaptive decomposition
Kernel extreme learning machine
ISSN:1865-0473, 1865-0481
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Accurate wind speed predictions are crucial for the planning, operation, and energy management of wind farms. In this paper, we propose a novel wind speed prediction model, CEESMDAN-LNR-SOA-KELM. Firstly, we employ the CEESMDAN decomposition method to extract features from the original wind speed data, capturing the underlying characteristics of the data. Secondly, we apply a nonlinear treatment to the convergence factor A of the seagull optimization algorithm (SOA) to better adapt to the complexity and diversity of the problem, thereby enhancing the algorithm's convergence speed. Additionally, we introduce a random opposition-based learning strategy to effectively prevent the SOA algorithm from getting stuck in local optima. We further optimize the parameters of KELM using LNR-SOA. The results of function optimization demonstrate that the proposed improvement strategy significantly enhances the parameter optimization capability of the SOA algorithm. The wind speed data from the Sotavento Galicia wind farm in Spain were used as the subject of the numerical experiments. The experimental results indicate that the model proposed in this paper demonstrates higher accuracy and reliability in wind speed prediction compared to the comparative models. It provides an effective forecasting tool for the wind energy industry and meteorological predictions.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1865-0473
1865-0481
DOI:10.1007/s12145-024-01560-8