Short-term forecasting of wind energy: A comparison of deep learning frameworks

Mora, Elianne; Cifuentes Quintero, Jenny Alexandra; Marulanda, Geovanny

Publication:
Short-term forecasting of wind energy: A comparison of deep learning frameworks

dc.affiliation.dpto	UC3M. Departamento de Ingeniería Eléctrica	es
dc.affiliation.grupoinv	UC3M. Grupo de Investigación: Redes y Sistemas de Energía Eléctrica (REDES)	es
dc.contributor.author	Mora, Elianne
dc.contributor.author	Cifuentes Quintero, Jenny Alexandra
dc.contributor.author	Marulanda, Geovanny
dc.date.accessioned	2022-05-20T07:36:20Z
dc.date.available	2022-05-20T07:36:20Z
dc.date.issued	2021-12-01
dc.description.abstract	Wind energy has been recognized as the most promising and economical renewable energy source, attracting increasing attention in recent years. However, considering the variability and uncertainty of wind energy, accurate forecasting is crucial to propel high levels of wind energy penetration within electricity markets. In this paper, a comparative framework is proposed where a suite of long short-term memory (LSTM) recurrent neural networks (RNN) models, inclusive of standard, bidirectional, stacked, convolutional, and autoencoder architectures, are implemented to address the existing gaps and limitations of reported wind power forecasting methodologies. These integrated networks are implemented through an iterative process of varying hyperparameters to better assess their effect, and the overall performance of each architecture, when tackling one-hour to three-hours ahead wind power forecasting. The corresponding validation is carried out through hourly wind power data from the Spanish electricity market, collected between 2014 and 2020. The proposed comparative error analysis shows that, overall, the models tend to showcase low error variability and better performance when the networks are able to learn in weekly sequences. The model with the best performance in forecasting one-hour ahead wind power is the stacked LSTM, implemented with weekly learning input sequences, with an average MAPE improvement of roughly 6, 7, and 49%, when compared to standard, bidirectional, and convolutional LSTM models, respectively. In the case of two to three-hours ahead forecasting, the model with the best overall performance is the bidirectional LSTM implemented with weekly learning input sequences, showcasing an average improved MAPE performance from 2 to 23% when compared to the other LSTM architectures implemented.	en
dc.format.extent	26
dc.identifier.bibliographicCitation	Mora, E., Cifuentes, J., & Marulanda, G. (2021). Short-Term Forecasting of Wind Energy: A Comparison of Deep Learning Frameworks. In Energies (Vol. 14, Issue 23, p. 7943). MDPI AG.	en
dc.identifier.doi	https://doi.org/10.3390/en14237943
dc.identifier.issn	1996-1073
dc.identifier.publicationfirstpage	7943
dc.identifier.publicationissue	23
dc.identifier.publicationlastpage	7969
dc.identifier.publicationtitle	Energies (Energies)	en
dc.identifier.publicationvolume	14
dc.identifier.uri	https://hdl.handle.net/10016/34862
dc.identifier.uxxi	AR/0000030537
dc.language.iso	eng	en
dc.publisher	MDPI AG	en
dc.rights	© 2021 by the authors. Licensee MDPI, Basel, Switzerland.	en
dc.rights	Atribución 3.0 España	*
dc.rights.accessRights	open access	en
dc.rights.uri	http://creativecommons.org/licenses/by/3.0/es/	*
dc.subject.eciencia	Estadística	es
dc.subject.eciencia	Ingeniería Mecánica	es
dc.subject.other	Deep learning	en
dc.subject.other	Long short-term memory	en
dc.subject.other	Time series forecasting	en
dc.subject.other	Wind power forecasting	en
dc.title	Short-term forecasting of wind energy: A comparison of deep learning frameworks	en
dc.type	research article	*
dc.type.hasVersion	VoR	*
dspace.entity.type	Publication