Procedure development for paramerer optimization of hydrological TANK model using a genetic algorithm

Abstract

    In recent years, there are many conceptual hydrological models introduced to simulate the rainfall-runoff process. But the Tank model is one of them has been widely used in many Asian countries, especially in Vietnam, to model flood forecasting because of these simple concepts and suitable conditions. However, it spends much time and effort to obtain good agreement between the simulated flow and the observed flow due to demand for calibrating many parameters in model. In that sense of this paper, a GA (Genetic Algorithm) has applied under the hydrological Tank model to calibrate the model parameters in the rainfall runoff process. Data used for the calibration are rainfall gauges of the river system upstream of the La Nga River watershed, which is located in southeast Vietnam. The Tank model parameters have been optimized by minimizing the sum of difference squares between simulated and observed values through a GA optimization search. As proved results under this study that GA can obtain efficient and convenience in the hydrology field and in solving of research problems in environmental modeling