Chaotic Forecasting Method of Short-Term Wind Speed in Wind Farm

To improve security and stability of power grid containing a lot of wind power generating units and to reduce operation cost, accurate short-term wind speed forecasting is of significance. On the basis of applying phase space reconstruction to wind speed time series, the wind speed is forecasted by chaotic weighted zero-order local forecasting method; and then to remedy the insufficiency of above-mentioned method, namely the forecasting is carried out by searching near phase point under high embedded dimensions, a modified weighted zero-order local forecasting method, which determines near phase point by correlation degree among phase points, is proposed and a new approach to compute weighted coefficient is put forward to improve forecasting accuracy. Case study results validate the feasibility and effectiveness of the proposed short-term wind speed forecasting method.