Rainfall Forecasting using Neural Network Fitting Tool (NFTOOL)

Forecasting is the application of science and technology to predict the state of the atmosphere for a future time at a given location. Generating predictions of meteorological events is very complex process, because the atmosphere is unstable. Rainfall Forecasting has become a major tool in numerous applications in meteorology and other environmental areas. This paper investigates to forecast the monthly rainfall of monsoon period between June to September of Ahmedabad district of Gujarat state, India. using Neural network fitting tool. Data fitting is the process of fitting models to data and analyzing the accuracy of the fit. Eight models prepared using different number of Inputs. Meteorological data of 50 years were collected of Ahmedabad District. Neural network data fitting tool analysed models by regression using two-layer feed-forward network trained with Levenberg-Marquardt method and gives results to check accuracy of models. It is found that model-3 with 4 inputs (Wet day frequency, Humidity, Max temperature, Potantial Evapotranspiration,) gives least error. So it concludes that model-3 is the best predicted model and Neural networks fitting tool can be used for the prediction of rainfall for the study area.