Wind power output short-term prediction method and system

The invention relates to the technical field of new energy output prediction, and discloses a wind power output short-term prediction method and system, and the method comprises the steps: collecting wind power generation influence parameters of various feature types through employing a plurality of data collection channels, and enabling one data collection channel to correspond to a plurality of wind power generation influence parameters of one feature type; performing global feature extraction and local feature extraction on the plurality of wind power generation influence parameters of each data acquisition channel; performing time sequence feature extraction on a plurality of wind power generation influence parameters of which the feature type of one data acquisition channel is historical wind power output; and fusing the extracted global feature information, local feature information and time sequence feature information to obtain a wind power output short-term prediction result. According to the method,