Flood level forecast model for tidal channel based on the radial basis function-artificial neural network

The radial basis function-artificial neural network(RBF-ANN)is a more excellent neural network,and is applied to flood level forecasting for tidal channel in this paper The parameters of the RBF-ANN are calculated by using the K-mean algorithms and the least square estimation algorithms Compared with the traditional BP algorithm,the RBF-ANN model is fast in convergence,and more valuable in practice Based on character of the tidal channel,the RBF-ANN model with some fore cast lead periods is presented The model is applied to flood level forecasting of Yihe River,and the result shows that the model work is very rapid and the satisfactory results are acquired.