An Artificial Intelligence Approach for Predicting Hydropower Production in the Nordic Power Market

Ali Khosravi*, Ville Olkkonen, Sanna Syri

*Corresponding author for this work

Research output: Contribution to conference without publisher/journalPaperResearchpeer-review

Abstract

Hydropower has historically had an important role in the Nordic power market. In the Nordic power market, hydropower accounts for around 60% of the electricity generated (2010–2019). The share of variable renewable energy sources (VRES) has grown considerably in recent years. Because of the growing awareness about climate change, this tendency is likely to continue in the future. In this paper, an artificial intelligence-based model to forecast hydropower production in different bidding areas in the Nordic power market was developed. Furthermore, the effects of spatial characteristics of VRES production on short-term hydropower production planning are analysed at bidding area level. As predicted, the AI model revealed that inflow and reservoir level are critical for the model's performance prediction. The findings showed that residual demand within the bidding region alone is insufficient to estimate hydropower generation. The model's forecast can be greatly improved by including residual demand for the other bidding areas as an input parameter. The forecast performance of the AI model for hydropower deteriorated as the percentage of non-dispatchable generation increased. However, the model demonstrated its ability to estimate hydropower in the face of the growing amount of variable renewable energy generation in the Nordic power market.
Original languageEnglish
Publication date15. Nov 2023
Publication statusPublished - 15. Nov 2023
EventXLIV Ibero-Latin American Congress on Computational Methods in Engineering - Porto, Portugal
Duration: 13. Nov 202316. Nov 2023

Conference

ConferenceXLIV Ibero-Latin American Congress on Computational Methods in Engineering
Country/TerritoryPortugal
CityPorto
Period13/11/202316/11/2023

Fingerprint

Dive into the research topics of 'An Artificial Intelligence Approach for Predicting Hydropower Production in the Nordic Power Market'. Together they form a unique fingerprint.

Cite this