Generated power prediction method and system based on canonical correlation analysis

The invention provides a generation power prediction method and system based on canonical correlation analysis. The method comprises the following steps: acquiring historical predicted weather and historical actual weather; based on the historical predicted weather and the historical actual weather, training to obtain a canonical correlation analysis model; acquiring historical typical predicted weather and historical actually measured power; training to obtain a power prediction model based on the historical typical prediction weather and the historical actually measured power; obtaining the predicted weather of the to-be-predicted time period; inputting the predicted weather into the canonical correlation analysis model, and outputting canonical predicted weather by the model; inputting the typical predicted weather into the power prediction model, and outputting predicted power by the model; according to the method, canonical correlation analysis is introduced into power prediction, so that the problem of