EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Decomposition aided attention-based recurrent neural networks for multistep ahead time-series forecasting of renewable power generation

Robertas Damaševičius iD, Luka Jovanovic, Aleksandar Petrovic, Miodrag Zivkovic, Nebojsa Bacanin iD, Dejan Jovanovic, Milos Antonijevic

PeerJ Computer Science · 18.01.2024 · 10 · e1795

DOI: 10.7717/peerj-cs.1795 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 18.01.2024. Verfasst von Robertas Damaševičius und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Applied Informatics, Vytautas Magnus University, Kaunas, Lithuania. Die Arbeit wurde bisher 71-mal zitiert (Crossref) und nennt 80 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Renewable energy plays an increasingly important role in our future. As fossil fuels become more difficult to extract and effectively process, renewables offer a solution to the ever-increasing energy demands of the world. However, the shift toward renewable energy is not without challenges. While fossil fuels offer a more reliable means of energy storage that can be converted into usable energy, renewables are more dependent on external factors used for generation. Efficient storage of renewables is more difficult often relying on batteries that have a limited number of charge cycles. A robust and efficient system for forecasting power generation from renewable sources can help alleviate some of the difficulties associated with the transition toward renewable energy. Therefore, this study proposes an attention-based recurrent neural network approach for forecasting power generated from renewable sources. To help networks make more accurate forecasts, decomposition techniques utilized applied the time series, and a modified metaheuristic is introduced to optimized hyperparameter values of the utilized networks. This approach has been tested on two real-world renewable energy datasets covering both solar and wind farms. The models generated by the introduced metaheuristics were compared with those produced by other state-of-the-art optimizers in terms of standard regression metrics and statistical analysis. Finally, the best-performing model was interpreted using SHapley Additive exPlanations.

Schlagwörter

Stichwörter der Autor:innen

AI explainabilityAttention mechanismMetaheuristicsRecurrent neural networksRenawable energy sourcesTime-series forecasting

Autor:innen und Institutionen

  • Department of Applied Informatics, Vytautas Magnus University, Kaunas, Lithuania.
  • Faculty of Technical Sciences, Singidunum University, Belgrade, Serbia.
  • Faculty of Informatics and Computing, Singidunum University, Belgrade, Serbia.
  • College of Academic Studies "Dositej", Belgrade, Serbia.

ORCID-Profile: Robertas Damaševičius, Nebojsa Bacanin

Zitieren

APA

Damaševičius, R., Jovanovic, L., Petrovic, A., Zivkovic, M., Bacanin, N., Jovanovic, D., & Antonijevic, M. (2024). Decomposition aided attention-based recurrent neural networks for multistep ahead time-series forecasting of renewable power generation. PeerJ Computer Science, 10, e1795. https://doi.org/10.7717/peerj-cs.1795

Vancouver

Damaševičius R, Jovanovic L, Petrovic A, Zivkovic M, Bacanin N, Jovanovic D, et al. Decomposition aided attention-based recurrent neural networks for multistep ahead time-series forecasting of renewable power generation. PeerJ Computer Science. 2024;10:e1795. doi:10.7717/peerj-cs.1795

BibTeX
@article{Damaeviius2024,
  author = {Damaševičius, Robertas and Jovanovic, Luka and Petrovic, Aleksandar and Zivkovic, Miodrag and Bacanin, Nebojsa and Jovanovic, Dejan and Antonijevic, Milos},
  title = {Decomposition aided attention-based recurrent neural networks for multistep ahead time-series forecasting of renewable power generation},
  journal = {PeerJ Computer Science},
  year = {2024},
  volume = {10},
  pages = {e1795},
  doi = {10.7717/peerj-cs.1795}
}
Kostenlos starten →