Wind power forecasting based on chaos and wavelet packet theory
Jiandong Mao, Xiaojing Zhang, Juan Li · 2018
With increasingly serious energy crisis and environment problem, as a green energy, wind power has exhibited its huge potential in substituting the traditional energy, and also people are paying more and more attentions to wind power. In wind generation the power forecasting is a key technology in power system dispatching. A novel ultra-short term wind power forecasting technology based on chaos and wavelet packet theory is present and the chaos and wavelet pocket theory is described. Some experiments are performed to verify the feasibility of the forecasting technology. The results show that when weighted one order local region prediction method was employed, wavelet function of sym6 was selected and 5 layer wavelet pocket decomposition was used, the technology is capable to predict accurately power output of future 4 hours and the forecasting value were in good agreement with the practical power.