Nowadays, green energy is considered as a viable solution to hinder CO2 emissions and greenhouse effects. Indeed, it is expected that Renewable Energy Sources (RES) will cover 40% of the total energy request by 2040. This will move forward decentralized and cooperative power distribution systems also called smart grids. Among RES, solar energy will play a crucial role. However, reliable models and tools are needed to forecast and estimate with a good accuracy the renewable energy production in short-term time periods. These tools will unlock new services for smart grid management. In this paper, we propose an innovative methodology for implementing two different non-linear autoregressive neural networks to forecast Global Horizontal Solar Irradiance (GHI) in short-term time periods (i.e. from future 15 to 120min). Both neural networks have been implemented, trained and validated exploiting a dataset consisting of four years of solar radiation values collected by a real weather station. We also present the experimental results discussing and comparing the accuracy of both neural networks. Then, the resulting GHI forecast is given as input to a Photovoltaic simulator to predict energy production in short-term time periods. Finally, we present the results of this Photovoltaic energy estimation discussing also their accuracy.

Non-linear Autoregressive Neural Networks to Forecast Short-Term Solar Radiation for Photovoltaic Energy Predictions / Aliberti, Alessandro; Bottaccioli, Lorenzo; Cirrincione, Giansalvo; Macii, Enrico; Acquaviva, Andrea; Patti, Edoardo (COMMUNICATIONS IN COMPUTER AND INFORMATION SCIENCE). - In: Smart Cities, Green Technologies and Intelligent Transport Systems / Donnellan B., Klein C., Helfert M., Gusikhin O., Pascoal A.. - [s.l] : Springer, 2019. - ISBN 978-3-030-26632-5. - pp. 3-22 [10.1007/978-3-030-26633-2_1]

Non-linear Autoregressive Neural Networks to Forecast Short-Term Solar Radiation for Photovoltaic Energy Predictions

Aliberti, Alessandro;Bottaccioli, Lorenzo;Cirrincione, Giansalvo;Macii, Enrico;Acquaviva, Andrea;Patti, Edoardo
2019

Abstract

Nowadays, green energy is considered as a viable solution to hinder CO2 emissions and greenhouse effects. Indeed, it is expected that Renewable Energy Sources (RES) will cover 40% of the total energy request by 2040. This will move forward decentralized and cooperative power distribution systems also called smart grids. Among RES, solar energy will play a crucial role. However, reliable models and tools are needed to forecast and estimate with a good accuracy the renewable energy production in short-term time periods. These tools will unlock new services for smart grid management. In this paper, we propose an innovative methodology for implementing two different non-linear autoregressive neural networks to forecast Global Horizontal Solar Irradiance (GHI) in short-term time periods (i.e. from future 15 to 120min). Both neural networks have been implemented, trained and validated exploiting a dataset consisting of four years of solar radiation values collected by a real weather station. We also present the experimental results discussing and comparing the accuracy of both neural networks. Then, the resulting GHI forecast is given as input to a Photovoltaic simulator to predict energy production in short-term time periods. Finally, we present the results of this Photovoltaic energy estimation discussing also their accuracy.
2019
978-3-030-26632-5
978-3-030-26633-2
Smart Cities, Green Technologies and Intelligent Transport Systems
File in questo prodotto:
File Dimensione Formato  
main.pdf

Open Access dal 29/07/2020

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 1.81 MB
Formato Adobe PDF
1.81 MB Adobe PDF Visualizza/Apri
versione_editoriale.pdf

non disponibili

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 3.66 MB
Formato Adobe PDF
3.66 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2746455
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo