Wind power prediction method based on principal component analysis network

The invention provides a wind power prediction method based on a principal component analysis network, and belongs to the technical field of wind power prediction through the principal component analysis network. The technical problem to be solved is to provide the improvement of the wind power prediction method based on the principal component analysis network. According to the technical scheme adopted for solving the technical problem, the method comprises the steps that the historical wind speed, the wind direction, the cloud cover, the surface air pressure, the environment temperature, the precipitation amount, all meteorological data fluctuation variables and active power state variables of a wind power station are collected and fused to construct sample data; singular point elimination is carried out on each state variable of the sample data, and after the samples are eliminated, data are supplemented in a time sequence with data points eliminated by adopting an interpolation method; performing normaliz