Kiyoumars Roshangar - Assistant professor, Faculty of Civil Engineering, Univ. of Tabriz, Tabriz, Iran
Reyhaneh Eskandari - M.Sc. graduate student, Faculty of Civil Engineering, Univ. of Tabriz, Tabriz, Iran
Fatemeh Vojoudi - M.Sc. graduate student, Faculty of Civil Engineering, Univ. of Tabriz, Tabriz, Iran

Accurate flow discharge prediction is very important in planning, designing, operating and maintenance of water resources structures. Various models have been developed so far to identify the relation between discharge and stage. In this study, artificial intelligent approach and conventional flow discharge rating curve models are considered for predicting flow discharge (FD) in a natural river. Discharge and stage data obtained from Ahar Chai River in northwest Iran. The accuracy of the ANN models is compared with conventional model. The determination coefficient (DC), coefficient of correlation (R2) and mean normalize error (MNE) statistics are used for evaluating the accuracy of the models. Based on the comparison results, the ANN models are found to be superior alternative to the conventional model

Flow discharge prediction, intelligent approach, conventional model, determination coefficient, coefficient of correlation, mean normalize error