Comparison of Rainfall-Runoff Simulation by Intelligent Techniques and a Conceptual Hydrological Model (A Case Study)

Rainfall-Runoff, as a major component of the hydrologic cycle, plays a key role in water resources management. This paper strives to give a comparison of rainfall-runoff simulation based on Artificial Neural Network (ANN) techniques and MIKE11-NAM model in Qaleh Shahrokh basin located in Iran. Also the best input of ANN models was identified using Gamma Test (GT). The reliability of MIKE11- NAM and ANNs models were evaluated based on performance criteria such as Root Mean Square Error (RMSE), Efficiency Index (EI) and correlation coefficient (R2). The obtained results show ANN models (BFGS-ANN and Conjugate ANN) have better performance than MIKE11-NAM model. Also, the performance BFGS-ANN model were better than other models with R2 value and RMSE equal to 0.92 and 2.01 (m3 / s), respectively. In addition, the results show GT can be used as a new method to determine the best input combination for network training for creating a smooth model by ANN models.