A Combination Forecasting Model for Wind Farm Output Power
Dai Hui-zhu · Power System Technology · 2009
It is of significance to forecast output power of wind farm for the operation of power grid to which large amount of wind power is connected.By use of BP neural network, radial basis function neural network and support vector machine, a combination forecasting model for output power of wind farm is built.The weights are calculated by three methods, i.e., equal weight average method, covariance optimization combination forecast and time-varying weight combination forecast.Research results show that the forecast accuracy from different methods is diverse one another;even though a method can offer high forecast accuracy in total, at individual point the forecast error of this method may be larger, however combination forecasting model can avoid larger forecast error in each point, so it is favorable to improve forecast accuracy.