The runoff characteristics of Fen river basin was analyzed first. According to the runoff data of 1960-2000, autocorrelation analysis method was used to determine model input variables and then radial basis function artificial neural network was used to recognize the relationship between previous annual runoff and later annual runoff. The developed prediction model was used to predict the annual runoff of 2001 to 2015 of Fen river basin. The result shows that: the predicted annual runoff was similar to the observed annual runoff, which means high prediction accuracy; the developed radial basis function artificial neural network model can be used for annual runoff prediction of Fen river basin. The research results are of great importance for water resource planning and management.
展开▼