Energy storage aging state estimation method based on transposition attention model

The invention discloses an energy storage aging state estimation method based on a transposition attention model. The method comprises the following steps: extracting a health factor representing battery aging from a lithium battery charging curve for predicting capacity aging; determining the relationship between the extracted health factors and capacity aging by using a Pearson's correlation coefficient, and selecting the health factors with correlation higher than 0.9 for input of an attention model; a transpose-based attention model is established, and health factors with correlation higher than 0.9 are used as input of the attention model composed of a transpose module, a position coding module, a coder module and a full-connection neural network module; a battery data set is selected and divided into a training set and a test set according to the proportion of 1: 4, the training set is used for training parameters of the attention model, and the prediction effect of the attention model is tested in the