Phase Space Reconstruction and Artificial Neural Networks Coupled Model in Mid-Long Term Flow Forecasting

The phase space reconstruction and artificial neural networks (ANN) coupled model is developed for flow forecasting in consideration of chaotic property and nonlinearity of flow series. 50 years of monthly flow data from 1950 to 1999 in Yichang hydrologic station is used for parameter calibration, and 4 years of the data from 2000 to 2003 is used for model validation. The result shows it has high precision and stability in flow forecasting. Compared with the periodic analysis model and the wavelet neural network model in forecasting precision, the phase space reconstruction and ANN coupled model is more satisfactory on qualified rate of forecasting and coefficient of deterministic in flow forecasting.