BP Neural Network Model Based on Chaos Theory and Application in Ground Water Level Forecasting

By the main component analysis, and maximum Lyapunov index method, this paper analyses chaotic character of ground water level time series. On this basis, combining the reconstruction phase space of chaos theory with BP neural network to set up a BP neural network model based on chaos theory. This paper forecasts ground water level of the Heihu Spring in Jinan by the model. The result shows that the model has a very good forecast accuracy and value. This method can provide a new way for going deep into forecasting Heihu spring discharge.