Please use this identifier to cite or link to this item:
Title: Development of Anfis Model with Optimised Inputs to Reduce the Computational Cost and Time for Ground Level Ozone Forecasting
Authors: Gorai, A K
Keywords: ANFIS model
Ozone Forecasting
ANFIS for ozone concentrations
Issue Date: Mar-2016
Publisher: Center for Health and the Global Environment.
Citation: 10th International Conference on Air Quality-Science and Application, Milan, Italy, 14-18 March 2016
Abstract: This  study  aims  to  develop  adaptive  neuro­fuzzy  inference  system  (ANFIS)  for  forecasting  of daily  ozone (O3)  concentrations in the  atmosphere  of  a mega  city. The  ANFIS model predictor considers the value of seven meteorological factors (pressure,  temperature,  relative  humidity,  dew  point,  visibility,  wind  speed,  and  precipitation),  NO2 concentration, and previous day’s ozone concentration in different combinations as  the inputs to predict the 1­day advance and same day ozone concentration. Collinearity  tests  were  conducted  to  eliminate  the redundant  input  variables.  A forward selection  (FS) method is used for selecting the different subsets of input variables. The method  reduces  the  computational  cost  and  time.  The  performances  of  the  models  were  evaluated  on  the  basis  of  four  statistical  indices  [(coefficient  of  determination  (R2),  normalized mean square error (NMSE), index of agreement (IOA), and fractional bias  (FB)].  
Description: Copyright belongs to proceeding publisher
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
2016_Amit_Development.pdf127.8 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.