Power sale quantity prediction method based on data restoration and decomposition sequence prediction

The invention relates to an electricity sale quantity prediction method based on data restoration and decomposition sequence prediction, and belongs to the technical field of power distribution network distributed energy storage system economic planning, and the method comprises the steps: firstly, employing Box-plot to detect a historical electricity sale data abnormal value; then, interpolating the missing electricity selling data by using the generative adversarial network; decomposing the electricity selling data into subsequences with different frequencies by adopting an improved adaptive noise complete set empirical mode algorithm; and finally, predicting the low-frequency sub-sequence by adopting an LEAR method, predicting the high-frequency sub-sequence by adopting LSTNet, and reconstructing a prediction result into a final electricity sales prediction result. According to the method provided by the invention, the prediction precision is improved from two aspects of data and a prediction model, and co