Electric power material number prediction method and device, computer equipment and storage medium

Abstract

The invention relates to an electric power material number prediction method and device, computer equipment and a storage medium, and relates to the technical field of electric power material prediction. The method comprises the steps of determining characteristic parameters of a material prediction model based on the material type of each electric power material in an electric power enterprise and the historical material number of each electric power material in each historical time period; fitting a material prediction model based on the feature parameters and the historical material number to obtain a fitted material prediction model; the fitted material prediction model is determined based on a model evaluation value corresponding to the test data; and inputting the historical material number of each electric power material in the historical time period in the electric power enterprise and the to-be-predicted time of each electric power material into the fitted material prediction model to obtain a number prediction result of each electric power material. By adopting the method, the prediction accuracy of the power resources can be improved.

Images