In this study, a stochastic weather generator and a neuro-fuzzy network were developed to generate precipitation and air temperature (max-mean-min) on a daily basis for meteorological stations of the coastal area of the Basilicata region (Southern Italy). Several simulations were carried out to build an optimal model, whose efficiency was evaluated with the calculation of the root mean square error (RMSE) and the mean absolute error (MAE), which were both obtained comparing simulated and observed values. Subsequently, the developed neuro-fuzzy model was applied to generate other weather variables, such as relative humidity, solar radiation and wind velocity. The simulations showed the good performance of the neuro-fuzzy network in the data filling of the available time series. The evaluation of the performance was made by comparing the values of RMSE and MAE obtained with the neuro-fuzzy method developed in this study and literature methods such as multiple linear regressions and trend line.
Daily Temperature and Precipitation Prediction Using Neuro-Fuzzy Networks and Weather Generators
TELESCA, Vito;CANIANI, Donatella;CALACE, STEFANIA;MAROTTA, LUCIA;MANCINI, Ignazio Marcello
2017-01-01
Abstract
In this study, a stochastic weather generator and a neuro-fuzzy network were developed to generate precipitation and air temperature (max-mean-min) on a daily basis for meteorological stations of the coastal area of the Basilicata region (Southern Italy). Several simulations were carried out to build an optimal model, whose efficiency was evaluated with the calculation of the root mean square error (RMSE) and the mean absolute error (MAE), which were both obtained comparing simulated and observed values. Subsequently, the developed neuro-fuzzy model was applied to generate other weather variables, such as relative humidity, solar radiation and wind velocity. The simulations showed the good performance of the neuro-fuzzy network in the data filling of the available time series. The evaluation of the performance was made by comparing the values of RMSE and MAE obtained with the neuro-fuzzy method developed in this study and literature methods such as multiple linear regressions and trend line.File | Dimensione | Formato | |
---|---|---|---|
Telesca_et_al_Springer_2017.pdf
solo utenti autorizzati
Tipologia:
Documento in Post-print
Licenza:
DRM non definito
Dimensione
1.66 MB
Formato
Adobe PDF
|
1.66 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.