Water demand prediction based on RBF neural network

Neural-network-based method of forecasting is presented in this paper. Simplified rival penalized competitive learning method (SRPCL) to make an adaptive clustering of networkspsila input pattern is developed. The objective function is established to adjust the structure of the neural network dynamically. Thus, the number of the hidden nodes is selected adaptively. The method is applied to water demand prediction in the Yellow river basin. The results of numerical simulations demonstrate the effectiveness of the method.