With the increasing of scale of wind power integrated into power system, the influence of wind power fluctuation to the power system is more serious. Wind speed prediction can help formulate electric power grid scheduling and make operational control decision, rational response to wind power fluctuations can reduce the impact of wind power fluctuations on grid safety and stable operation. This paper establishes an equivalent wind speed model considering the geographic distribution of wind turbines and wind speed distribution in time and space. The wind speed relationship between upstream wind turbine and downstream wind turbine is established based on this model, which is used to fix the wind speed of downstream wind turbine. In addition, a wind farm power forecasting method is proposed based on wind speed revision, in order to improve the prediction accuracy of wind power. The feasibility and rationality of the method is verified by simulation examples based on actual wind farms parameters and geographic data.