Wind power generation has intermittent

random

and uncertain characteristics. Accurate wind power prediction methods are crucial for ensuring the stable operation of the power grid and promoting the consumption of new energy. A machine learning based wind power prediction model can establish a nonlinear and complex mapping between input features and predicted power values through feature mining and complex correlations of a large amount of data. Firstly

the overall framework of wind power prediction based on machine learning is given

and then summarizes five prediction models including deep learning prediction model

ensemble learning prediction model

small sample prediction model

online learning prediction model

and physical data joint prediction model. Targeted research suggestions are proposed

and the five wind power prediction models are compared and analyzed.Finally

the application of large model technology in wind power prediction is briefly analyzed

and the wind power prediction models based on machine learning small and large models are compared.