Institutional Repository of Vilnius University
Daily energy forecasting of solar power plants based on weather forecast data /
Abstract
dc:descriptionThis paper evaluates the performance of three machine learning models - LSTM, SVR, and gradient boosting — across different forecasting horizons (4, 12, 24, and 48 hours) for predicting power generation in a photovoltaic power plant. The models represent different types of forecasting methods: time-series statistical and statistical regression models, as well as an ensemble method. The results show that the gradient boosting excels in short-term forecasting (4 to 12 hours), long short-term memory networks perform best for short and medium-range forecasting (4 to 24 hours), and support vector regression is most effective for long-term forecasting (48 hours). Overall, the gradient boosting model achieved the lowest average RMSE over multiple horizons. External factors affecting forecast accuracy are discussed, and potential solutions are proposed.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Grigaliūnaitė, Aistė,
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD229556514&prefLang=en_US