A Machine Learning Approach to Forecasting Hydropower Generation

Sarah Di Grande (), Mariaelena Berlotti, Salvatore Cavalieri and Roberto Gueli
Additional contact information
Sarah Di Grande: Department of Electrical Electronic and Computer Engineering, University of Catania, Viale A. Doria 6, 95125 Catania, Italy
Mariaelena Berlotti: Department of Electrical Electronic and Computer Engineering, University of Catania, Viale A. Doria 6, 95125 Catania, Italy
Salvatore Cavalieri: Department of Electrical Electronic and Computer Engineering, University of Catania, Viale A. Doria 6, 95125 Catania, Italy
Roberto Gueli: Etna Hitech S.C.p.A., Viale Africa 31, 95129 Catania, Italy

Energies, 2024, vol. 17, issue 20, 1-22

Abstract: In light of challenges like climate change, pollution, and depletion of fossil fuel reserves, governments and businesses prioritize renewable energy sources such as solar, wind, and hydroelectric power. Renewable energy forecasting models play a crucial role for energy market operators and prosumers, aiding in planning, decision-making, optimization of energy sales, and evaluation of investments. This study aimed to develop machine learning models for hydropower forecasting in plants integrated into Water Distribution Systems, where energy is generated from water flow used for municipal water supply. The study involved developing and comparing monthly and two-week forecasting models, utilizing both one-step-ahead and two-step-ahead forecasting methodologies, along with different missing data imputation techniques. The tested algorithms—Seasonal Autoregressive Integrated Moving Average, Random Forest, Temporal Convolutional Network, and Neural Basis Expansion Analysis for Time Series—produced varying levels of performance. The Random Forest model proved to be the most effective for monthly forecasting, while the Temporal Convolutional Network delivered the best results for two-week forecasting. Across all scenarios, the seasonal–trend decomposition using the LOESS technique emerged as the most successful for missing data imputation. The accurate predictions obtained demonstrate the effectiveness of using these models for energy planning and decision-making.

Keywords: renewable energy; hydropower; machine learning; forecasting; sustainability; water distribution system (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1996-1073/17/20/5163/pdf (application/pdf)
https://www.mdpi.com/1996-1073/17/20/5163/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jeners:v:17:y:2024:i:20:p:5163-:d:1500387

Access Statistics for this article

Energies is currently edited by Ms. Agatha Cao

More articles in Energies from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().