Cloud edge fusion load adjustment method and system based on improved LSTM

The invention relates to the technical field of power grids, and discloses a cloud edge fusion load adjustment method and system based on improved LSTM, and the method comprises the steps: S1, constructing an influence factor sequence, S2, extracting an effective feature set, and S3, carrying out the power prediction of an improved long and short memory network algorithm; s4, establishing a load regulation model according to a master station instruction; and S5, solving the objective function by using the wolf pack algorithm. According to the method, an influence factor sequence is constructed, data is processed by using a weighted mahalanobis distance, accurate extraction of load power data of the new energy vehicle is realized, feature extraction is performed on the data by using an improved LSTM, so that rapid and accurate prediction of regional loads is realized, and then, the minimum network loss and the minimum peak-valley difference are taken as objective functions, so that the optimal prediction of th