Prediction of Reservior Runoff Using RBF Neural Network-Grey System United Model

At present, classic methods are used to predict reservoir runoff, but the result is not ideal. Due to the shortages of neural network and grey system, in this paper, a grey neural network model is set up based on grey and neural network theory. The data got from the GM(1,4) on the factors affecting the reservoir runoff is used as the input of the neural network and the origin data of reservoir runoff are used as the output of neural network which was trained to get the optimal structure of neural network. The results show that the model had highly fitting and predicting precision advantages than other model had. The case study shows that the model is quite accurate in prediction reservoir runoff, which has some project referential value.