Akademik Veri Yönetim Sistemi
IEEE Sponsored ELECO 2023, Bursa, Türkiye, 30 Kasım - 02 Aralık 2023, ss.74-79
Renewable energy sources play a pivotal role in contemporary
distributed energy generation, owing to their significance in
reducing energy costs and mitigating carbon emissions. Ensuring predictability in energy demand and production is crucial for effective future planning, wherein intuitive predictions
for renewable energy sources are indispensable. In this study,
we propose a novel method for predicting power generation
in solar power plants. We develop a hybrid prediction model
by combining prevalent machine learning models trained with
meteorological data, yielding superior results compared to individual model outcomes. Through analysis, we evaluate the
performance of the models trained with real meteorological
and production data, while emphasizing the advantages of the
proposed hybrid approach. The proposed method offers valuable insights into enhancing the predictability of solar power
plant generation, thereby contributing to the advancement of
renewable energy utilization.