A New Real Time Forecasting Model for Wind Power
Wind power is the most large-scale development of technical and economic conditions of non-hydro renewable energy. The real time forecasting for wind power is difficult because of the wind power data has nonlinear interaction. A new real time forecasting model for wind power is established. In the model, state space reconstruction is used to transfer the original wind power time series to high dimension space. The input vector and anticipant output vector can be gained by the changed data in the high dimension space. Based on the theory of support vector machine, the real time forecasting model is established with the principle of structural risk minimization of support vector machine. The new model is used for the real time forecasting of wind power. The results prove the efficiency and validity of the new model.