Abstract: Wind power prediction can effectively help balance the power supply and load demand of a system, thereby reducing the impact of wind power fluctuations and uncertainty on the stability of the power system. As the proportion of wind power connected to the power system increases, how to accurately predict the power generation of wind farms has become a major challenge, which will affect the grid-connected operation of wind farms. The research progress of domestic and foreign scholars in wind farm power prediction in recent years was systematically summarized. Firstly, a variety of wind power prediction methods according to the classification criteria of time scale, prediction physical quantity, prediction principle, and prediction characteristics was classified and sorted out. The research status of various prediction methods was summarized, its characteristics and application scenarios were explained, and a reference for the selection of prediction models was provided. Then, the evaluation indicators of the prediction effect were summarized from the perspectives of single-point prediction, probability prediction and prediction curve, and some suggestions were given on the selection of evaluation indicators. Finally, the key problems affecting the accuracy and practicability of wind farm power prediction were summarized, and the possible development trends of wind farm power prediction in the future were proposed, which provided a reference for the accurate prediction of wind farm power and the stable grid-connected operation of the power grid.