A Study on the Simulation of Rainfall-runoff Process Using Artificial Neural Network (ANN) and HEC-HMS (Case study: Kasilian Basin)
HEC-HMS and Artificial Neural Network (ANN) were applied to simulate the rainfall-runoff process in the Kasilian basin. The study site is located at north of Iran with an area of 68 km2. The ANN has high capability in establishing connection between input and output data and the HEC-HMS model has shown its capability in optimizing simulated hydrographs. Initial loss is a quantitative parameter which is dependent on three main factors including soil، vegetation and Antecedent Moisture Conditions (AMC). In this study after optimizing the initial loss parameter using the HEC-HMS model، this parameter along with incremental rainfall were applied qua inputs in the ANN to simulate runoff or discharge values. Comparison of the obtained results under two scenarios of optimized and non optimized initial loss showed that optimized initial loss highly enhance the accuracy of simulated runoff and flood hydrograph by 100% in some events.