International Journal of Advances in Computer Science and Its Applications
Author(s) : HASAN ALMASI, SAEID FAZLI
In this paper, we have utilized ANN (Artificial Neural Network) modeling for forecasting rainfall in Babolsar synoptic station in Iran. To achieve such a model, we have used daily rainfall data from 1978 to 2007 for the synoptic station. A Time Delay Series Neural Network is used in this work. Two hidden layers are considered for the neural network. Inputs are daily rainfall with variable lag to achieve optimal prediction. Using this method as a black box model, we have realized the hidden dynamics of rainfall through the past information of the system. Root Mean Square Error (RMSE) and Correlation Coefficient (r2) are evaluated for comparison purposes. Optimum delay days are calculated for rain forecasting which can be used in climatology application.