liu.seSearch for publications in DiVA
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Hourly Hydropower Production Forecasting with Machine Learning: A Case Study in Linköping, Sweden
Linköping University, Department of Management and Engineering, Energy Systems. Linköping University, Faculty of Science & Engineering.ORCID iD: 0009-0008-6356-2023
Linköping University, Department of Management and Engineering, Energy Systems. Linköping University, Faculty of Science & Engineering. Division of Building, Energy and Environment Technology, Department of Technology and Environment, University of Gävle, Gävle, Sweden.ORCID iD: 0000-0002-0604-3672
Linköping University, Department of Management and Engineering, Energy Systems. Linköping University, Faculty of Science & Engineering.ORCID iD: 0000-0002-6885-6118
Linköping University, Department of Management and Engineering, Energy Systems. Linköping University, Faculty of Science & Engineering.ORCID iD: 0000-0001-7798-0471
2024 (English)In: Proceedings of the 10th World Congress on New Technologies (NewTech'24), 2024Conference paper, Published paper (Refereed)
Abstract [en]

Machine Learning (ML) is frequently utilized in prediction tasks; however, its applications in hydropower forecasting,particularly in forecasting hourly power production, has not been thoroughly investigated. In this paper, two Deep Learning (DL) models,namely an autoregressive neural network and Long Short-Term Memory, are compared to a seasonal autoregressive moving average(SARIMA) model to forecast the hourly power production at a hydropower station situated in Linköping, Sweden. Hyperparameteroptimization algorithms are used to identify suitable DL models and algorithms for automatic model identification of SARIMA modelsare utilized. The three models are evaluated using a rolling origin strategy on a test dataset that consists of 10 months (January – October2023) of hourly power production. The DL models provided similarly accurate forecasts as the SARIMA model according to meansquared error and mean absolute error. However, the DL models are poorly calibrated, resulting in lower coverage compared to theSARIMA model. Furthermore, the models are using a univariate time series (i.e., using historical power production to forecast futurepower production) and future studies need to explore additional variables that may be useful in providing a more accurate forecast.

Place, publisher, year, edition, pages
2024.
Series
ICERT ; 102
Keywords [en]
Machine learning, deep learning, forecasting, time series, hydropower, power production, uncertainty estimation
National Category
Probability Theory and Statistics
Identifiers
URN: urn:nbn:se:liu:diva-207672DOI: 10.11159/icert24.102OAI: oai:DiVA.org:liu-207672DiVA, id: diva2:1898116
Conference
10th World Congress on New Technologies (NewTech'24), Barcelona, Spain, August 25-27, 2024.
Available from: 2024-09-16 Created: 2024-09-16 Last updated: 2024-10-18

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full text

Authority records

Kåge, LinusMilić, VlatkoAndersson, MariaWallén, Magnus

Search in DiVA

By author/editor
Kåge, LinusMilić, VlatkoAndersson, MariaWallén, Magnus
By organisation
Energy SystemsFaculty of Science & Engineering
Probability Theory and Statistics

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 315 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf