Abstract: When the existing method is used to predict the wind speed of the short-term VSCF wind power generation system,the output reactive power,output active power and short-term wind speed of the system cannot be accurately predicted due to its failure to analyze the operating characteristics of the wind turbine,and the average absolute error and mean square error of the prediction results are also large. To this end,the wind speed prediction method of the VSCF wind power generation system is proposed in this paper.Firstly,the operation characteristics of the wind turbine are analyzed, and then the support vector machine regression algorithm is used to construct the wind speed prediction model.Finally,the wind speed prediction model is used to complete the short-term prediction of wind speed of the VSCF wind power generation system. The experimental results show that the proposed method can accurately predict the output reactive power,output active power and short-term wind speed of the system,and the average absolute error and mean square error of the prediction are small,which verifies the overall effectiveness of the proposed method.