Digital Library

Forecasting energy consumption in energy clusters using machine learning methods

• Piotr Jurek
• Anna Plichta

(2024). ECMS 2024, 38th Proceedings
Edited by: Daniel Grzonka, Natalia Rylko, Grazyna Suchacka, Vladimir Mityushev, European Council for Modelling and Simulation.
DOI: http://doi.org/10.7148/2024
ISSN: 2522-2422 (ONLINE)
ISSN: 2522-2414 (PRINT)
ISSN: 2522-2430 (CD-ROM)
ISBN: 978-3-937436-84-5
ISBN: 978-3-937436-83-8 (CD) Communications of the ECMS Volume 38, Issue 1, June 2024, Cracow, Poland June 4th – June 7th, 2024

https://doi.org/10.7148/2024-0500

Piotr jurek, Anna plichta (2024). Forecasting Energy Consumption in Energy Clusters using Machine Learning Methods, ECMS 2024, Proceedings Edited by: Daniel Grzonka, Natalia Rylko, Grazyna Suchacka, Vladimir Mityushev, European Council for Modelling and Simulation. doi:10.7148/2024-0500

The rising demand for secure, sustainable energy generation led to the emergence of the concept of Energy Cluster in Poland. Renewable energy sources depend on such factors as changeable weather conditions, which makes them unstable. Therefore, electrical energy consumption forecasting is necessary to balance the needs of every Energy Cluster member. 

This article examines the efficacy of various machine learning models, such as Linear Regression, Decision Tree Regression, K-neighbors regression, Exponential Smoothing, or Support Vector Regression in predicting such data. Deep learning models based on Long Short-Term Memory and convolution were also considered. Depending on factors such as the seasonality of the dataset or the presence of a trend, each examined model performed differently. 

Overall, LSTM turned out to be the most universal method, working well with various datasets and balancing learning speed with accuracy. In some specific cases, however, Exponential Smoothing proved more efficient, suggesting that an entity-by-entity approach may be appropriate.