In order to reduce the energy consumption of the router

an improved back propagation（ BP） neural network method is proposed to predict the flow of the router. This method adds a learning rate adaptive algorithm based on the traditional BP neural network

it can improve the prediction precision and speed of training. MATLAB platform is used to build BP neural network for network traffic modeling

training and forecasting. The simulation results show that the improved BP neural network has higher precision and faster training speed than the traditional prediction model. The router can adjust the working state more quickly and accurately

so as to realize effective energy-saving. This method has a good application prospect.