Hourly Forecasting of SO2 Pollutant Concentration Using an Elman Neural Network
- Autori: BRUNELLI U; V PIAZZA; L PIGNATO; F SORBELLO; VITABILE S
- Anno di pubblicazione: 2006
- Tipologia: eedings
- Parole Chiave: Neural networks ; Recurrent neural networks ; Models
- OA Link: http://hdl.handle.net/10447/15496
Abstract
In this paper the first results produced by an Elman neural network for hourly SO2 ground concentration forecasting are presented. Time series has been recorded between 1998 and 2001 and are referred to a monitoring station of SO2 in the industrial site of Priolo, Syracuse, Italy. Data has been kindly provided by CIPA (Consorzio Industriale per la Protezione dell'Ambiente, Siracusa, Italia). Time series parameters are the horizontal and vertical wind velocity, the wind direction, the stability classes of Thomas, the base level of the layer of the atmospheric stability, the gradient of the potential temperature and the difference of the potential temperature of reference.