Power-forecasting model is a relevant topic in modern power systems. This is due to the wide integration of Renewable Energy Sources (RESs) and their not predictable production. To face this issue, the paper presents a comparison between the performances of the traditional Autoregressive (AR) statistical method and the more recent Artificial Neural Network (ANN) method. The comparison is applied in forecasting of the most significant RES present in the Sicily zone (Italy), i.e., solar sources. To this aim, using the data of Gestore Servizi Energetici (GSE), both the AR and ANN architectures are built by exploiting the information available in time period 2010 - 2015 in terms of temperatures and solar irradiation, thus obtaining a power forecasting for 2016 for each country of the Sicily zone. The performances of the two different methods are compared on the basis of both Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) values. These two indexes are used to evaluate the accuracy of the proposed models.

On Comparing Regressive and Artificial Neural Network Methods for Power System Forecast / Andreotti, A.; Caiazzo, B.; Di Pasquale, A.; Pagano, M.. - (2021), pp. 1-6. (Intervento presentato al convegno 2021 AEIT International Annual Conference, AEIT 2021 tenutosi a ita nel 2021) [10.23919/AEIT53387.2021.9626938].

On Comparing Regressive and Artificial Neural Network Methods for Power System Forecast

Andreotti A.;Caiazzo B.;Di Pasquale A.;Pagano M.
2021

Abstract

Power-forecasting model is a relevant topic in modern power systems. This is due to the wide integration of Renewable Energy Sources (RESs) and their not predictable production. To face this issue, the paper presents a comparison between the performances of the traditional Autoregressive (AR) statistical method and the more recent Artificial Neural Network (ANN) method. The comparison is applied in forecasting of the most significant RES present in the Sicily zone (Italy), i.e., solar sources. To this aim, using the data of Gestore Servizi Energetici (GSE), both the AR and ANN architectures are built by exploiting the information available in time period 2010 - 2015 in terms of temperatures and solar irradiation, thus obtaining a power forecasting for 2016 for each country of the Sicily zone. The performances of the two different methods are compared on the basis of both Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) values. These two indexes are used to evaluate the accuracy of the proposed models.
2021
978-88-87237-50-4
On Comparing Regressive and Artificial Neural Network Methods for Power System Forecast / Andreotti, A.; Caiazzo, B.; Di Pasquale, A.; Pagano, M.. - (2021), pp. 1-6. (Intervento presentato al convegno 2021 AEIT International Annual Conference, AEIT 2021 tenutosi a ita nel 2021) [10.23919/AEIT53387.2021.9626938].
File in questo prodotto:
File Dimensione Formato  
On_Comparing_Regressive_and_Artificial_Neural_Network_Methods_for_Power_System_Forecast.pdf

solo utenti autorizzati

Tipologia: Documento in Post-print
Licenza: Accesso privato/ristretto
Dimensione 736.59 kB
Formato Adobe PDF
736.59 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

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/11588/873080
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? ND
social impact