摘要：Flood management is one of the biggest issues in urban areas. While hydrological modeling is widely applied for the estimations of watershed's hydrological response to precipitation, application of GEP, in this study, exhibit an inventive framework is implemented for assessment of rainfall- runoff purpose. GEP designed for regression, classification, time series estimation and logical synthesis with using character linear chromosomes composed of genes structurally organized in a head and a tail. An increment in hydrological data (i.e. rainfall, evapotranspiration) is generally expected to increase the accuracy of flood forecasts. However, in this study, it is noteworthy that GEP can predict floods and capture time of peaks using only precipitation data over a basin with a high correlation between flow rate and rainfall. Rainfall-runoff predictions are conducted using two storms. The results illustrate that GEP model can capture and predict runoff without topographic and hydrological data (with the exception of precipitation) for the rainfall events over a sub-tropical catchment. The addition of evapotranspiration and/or groundwater data sets to the simulations does not have leading reaction to acquire flow fluctuations with exception of increment in the width of the peak flow in the hydrograph.
