The accuracy of wind speed and wind turbine power output forecast is important to the power system operation. Based on similarity curves， a back propagation（BP） neural network wind speed forecast method is proposed to improve the accuracy of wind speed forecast by using the seasonal cycle fluctuation of wind speed. Combining with time series analysis and grey forecasting method， a wind speed forecasting method is studied. The wind turbine power prediction is discussed based on the wind speed prediction. Simulation results on the case study of actual wind speed data in some area in China demonstrate that the proposed methodology can improve the accuracy of wind speed forecast and has practical value.