Short-term wind speed forecasting based on SVM time-series method
Zhensheng Wu · Electric Power · 2011
Short-term wind speed forecasting is of significance for the operation of grid-connected wind power generation systems.An accurate wind speed forecasting can effectively reduce or avoid the adverse effect of wind farm on power grid and strengthen competition ability of wind farm in electricity market.A new application of the Support Vector Machine(SVM) theory is introduced.The detailed process for SVM usage in wind forecast process is discussed.A SVM wind speed forecast model for wind farm is founded.This model,which only uses historical wind data as input,is simple and effective compared to other models which need additional meteorological data.The comparison of SVM forecast results with other forecast models such as improved fuzzy analytical hierarchy process model,ARMA-ARCH model,EMD-ARMA model and double ARMA models proves the validity of the proposed model.It can be used in the short-term forecast and power generation prediction effectively.