Resident load prediction method and device based on neural network and dynamic mirror image descent

The invention relates to a resident load prediction method based on a neural network and dynamic mirror image descent, and the method comprises the following steps: obtaining a resident load related data set at a plurality of moments before a to-be-predicted moment, carrying out the regularization processing of the resident load related data, carrying out the segmentation of the resident load related data set after the regularization processing, and carrying out the segmentation of the resident load related data set; obtaining a test set and a training set; constructing a deep long-short-term memory neural network model, and training the deep long-short-term memory neural network model by using the training set to obtain a trained deep long-short-term memory neural network model; obtaining an error adjustment variable by using an improved dynamic mirror image descent method so as to correct a prediction error; using the test set and the trained deep long short-term memory neural network model to obtain an int