Forecasting method study on chaotic load series with high embedded dimension

The current chaos forecasting methods, which apply Euclidean distance to determine the nearest points in state space to forecast chaos time series with high dimensions, are not so effective. In this paper, one new idea based on incidence degree instead of Euclidean distance is firstly put forward to determine the nearest point in phase space. In the mean time, the method is developed to improve the character of the full Lyapunov exponential spectrum in reconstruction space with high dimensions. The test result of short-term load forecasting series shows that the precision of load forecasting is greatly improved by means of the new method when the embedded dimension is high compared with the method used recently.