Method and system for predicting medium-term power probability of regional wind power plant based on cluster division

The invention discloses a regional wind power plant mid-term power probability prediction method and system based on cluster division. The method comprises the steps of obtaining historical meteorological characteristic parameters of a to-be-predicted region in a set time period; dividing a wind power plant cluster based on a subtraction clustering algorithm on the basis of the historical power sequence of the wind power plant in the corresponding time period; according to a wind power plant cluster division result, establishing a wind power plant mid-term power prediction model based on a light GBM algorithm for each sub-cluster, and obtaining a power prediction result of each sub-cluster; and obtaining a power prediction error of each sub-cluster, calculating probability density distribution of the power prediction errors by adopting a non-parametric kernel density estimation method, and further obtaining a power probability prediction result of the regional wind power plant. According to the invention, the