Photovoltaic Energy Production Forecasting Using LSTM and Cross-Validation
View/ Open
Author
Gonçalves, Nairon Augusto Monari
Martins, Antonio Cesar Germano
Fourmaux, Nicolas
Metadata
Show full item recordAbstract
The recent water shortage faced in Brazil and being the seventh largest in terms of population size according to the United Nations (UN), requires efficient and sustainable energy management strategies. Photovoltaic (PV) energy is a promising renewable source to meet this demand, but accurately forecasting its production remains a critical challenge. This study proposes the implementation of Long Short-Term Memory (LSTM) neural networks, a variant of recurrent neural networks (RNNs), for reliable prediction of photovoltaic production. To ensure the robustness and reliability of the predictive model, this work will use the Cross-validation statistical method. This technique helps validate model performance by dividing the dataset into training and testing sets, thereby providing insights into its generation and predictive capabilities. This is crucial to help energy suppliers in their decision-making processes and improve the regulation of photovoltaic production.
https://doi.org/10.53316/cbgd2023.026
Collections
The following license files are associated with this item: