The present invention discloses a power network planning method taking the consideration of wind power plant output power correlative characteristics. On the basis of historical output data of wind electric plants, an non-parametric estimation method is employed to determine edge distribution characteristics of random output of each wind electric plant. A maximum likelihood method is employed to calculate parameters in each common Copula function, and appropriate Copula functions are selected from the common Copula functions to perform quantification of correlative characteristics among the random output of the wind electric plants according to a rule of the minimum squared Euclidean distance in squared Euclidean distances between each common Copula function and experience Copula distribution. The method provided by the invention is simple and efficient.