Author(s): Ming-Hsi Hsu; Shu-Horng Lin; Jin-Cheng Fu
Keywords: Stage Profiles forecasting; Flash Flood Routing; Artificial Neural Networks
Abstract: The artificial neural networks were used to predict the stages at gauges in a river. Then the prediction was taken as interior boundary conditions at gauges for correcting the river stage forecasting in a flash flood routing model, which calculates the longitudinal profiles in a river. The flash flood routing model is based on the dynamic wave equations with discretization processes of the four-point implicit finite difference approximation. Eight typhoon events were applied to calibrate and verify the rainfall-stage model and three of those were simulated to demonstrate the proposed model’s capability to forecast the river stage profiles. The results revealed that the flash flood river routing model incorporating predicted stages with artificial neural networks can provide accurate river stages for flood forecasting.