The invention belongs to the technical field of wind power plants, and particularly relates to a wind power prediction method based on twenty-four solar terms. An influence on wind power prediction accuracy by season and climate change factors is considered, and the invention puts forward the wind power prediction method based on twenty-four solar terms. By use of the method, from perspectives of prediction model establishment and sample division, a situation that the climate state of a current prediction moment can be more favorably represented by a meteorological data sample set in the same solar term is considered; according to the time nodes of the twenty-four solar terms, sample data is divided, a prediction model is established by wind power plant data in the same solar term, so that the influence by season and climate changes can be reduced to a certain degree, and wind power prediction accuracy is improved.