Wind power short-term forecast for power system stability and the improvement of power quality has very important significance. According to the existing prediction methods, the paper puts forward a kind of wind power short-term forecast method based on Hilbert Huang transform.First of all,using the continuation inhibition method that putting forward restrains the endpoint of the empirical mode decomposition existed in the original data.then using empirical mode decomposition method?resolve wind power history data and get seven different characteristics of the component,And on the basis of analyzing the characteristics of the components, respectively, sets up different prediction models and the Hilbert transform is performed.Then combining multiple prediction models forecast the wind power history data.Simulation results show that this method makes wind power prediction accuracy is greatly increased, and it has the very good application prospect.