Mahsa Hasanpour Kashani - Department of Water Engineering, University of Mohaghegh Ardabili, Ardabil, Iran

River flow forecasting is an important task for water resources management and planning. In this study, three intelligent models namely, artificial neural network (ANN), adaptive neuro -fuzzy inference system (ANFIS), and gene expression programming (GEP) models are applied for river flow forecasting of the Ghare-Soo River located at the Ardabil province using daily lagged discharge data in the period of 2005-2013 collected from the Doostbigloo hydrometric station. Four performance criteria namely, correlation coefficient, root mean square error, Nash-Sutcliff coefficient and bias were used to evaluate and compare results of the models. The results obtained showed that the performances of all the models are satisfactory. However, the gene expression programming model was identified as the most suitable model for flow forecasting of the Ghare-soo River.

Adaptive neuro-fuzzy inference system, Artificial neural networks, Discharge forecasting, Gene expression programming