Author(s): Mohammad Ebrahim Banihabib; Farimah S. Jamali; S. Majid Mousavi
Linked Author(s):
Keywords: Inflow forecasting; Neural network model; Multiple regressionspatial and temporal correlation; Dez reservoir
Abstract: Reservoir inflow forecasting has an important role in water resources engineering. This paper provides a solution for Dez reservoir inflow forecasting using an artificial neural network model (ANN) and a multiple regression model. Moreover it aims to define the best pattern with spatial and temporal correlation, in daily (short-time) and monthly (long-time) time scales, among stations on dam upstream, in two scenarios. According to the correlation coefficient and mean square of errors (MSE), the performances of models were compared and the ANN model was found to model the flows better. In short-time scale, best result is according to two prior days and Tangpanj station; furthermore the longest time scale with best result is one month, using all upstream stations as input data.
Year: 2009