Starting from the changing characteristics of the power load
a hybrid model based on machine learning is proposed
and the model is analyzed using a complete set of empirical mode decomposition technology with adaptive noise to decompose the power load.After removing the influence of high-frequency noise
the load sequence is reconstructed
The process of optimizing the extreme learning machine with a multi-objective flower pollination optimization algorithm
and that the model can improve the accuracy and stability of the extreme learning machine prediction
combined with the monthly load forecasting case analysis in Jiangsu Province to verify the effectiveness of the model.