Semi-supervised load identification and decomposition method and system and storage medium

The invention discloses a semi-supervised load identification and decomposition method and system and a storage medium. The method comprises the following steps: acquiring aggregated energy consumption data collected in monitoring equipment; obtaining and storing a group of marked and unmarked decomposed energy consumption data; preprocessing the marked decomposed energy consumption data and the unmarked decomposed energy consumption data, and extracting related features; training a semi-supervised learning model according to the related features to obtain a trained semi-supervised learning model; and substituting the aggregated energy consumption data into the trained semi-supervised learning model to decompose the aggregated energy consumption data to obtain the load of each electric device. The non-intrusive load identification technology is adopted, the semi-supervised learning algorithm is combined, and the method has the following advantages that the method does not depend on a sample library, is high i