Method for establishing multi-step wind speed combined prediction model

The invention discloses a method for establishing a multi-step wind speed combined prediction model. The method comprises the following steps: performing grey correlation analysis on wind speed related meteorological attribute factors in an initial data set, and selecting the meteorological attribute factors of which the correlation degree with the wind speed is greater than a set threshold valueas an actual input data set of the prediction model; based on meteorological attribute factors in the actual input data set, performing multi-step primary prediction on the output variable wind speedby adopting a regularization extreme learning machine algorithm improved by a rhododendron optimization algorithm to obtain wind speed primary prediction results under different time scale requirements; and adopting an error correlation linear analysis method to correct the prediction model by solving the function relation between historical prediction errors and an actual input data set in orderto obtain a final predictio