Method and system for predicting adjustable load of power consumer

The invention relates to a power consumer adjustable load prediction method and system, and the method comprises the steps: obtaining power consumption data of a plurality of power consumers, and carrying out the filling of missing values in the power consumption data of the plurality of power consumers, and obtaining the complete power consumption data of the plurality of power consumers; according to the complete power consumption data of the plurality of power consumers, constructing an incidence matrix between the consumers; wherein the incidence matrix is used for representing the electrical load relevance between different users, and if the electrical load difference between two power users is within 5%, it is considered that the electrical load relevance exists between the two power users; performing QR decomposition on the association matrix to obtain association features; and inputting the associated features into a pre-trained power consumer adjustable load prediction model to obtain a power consume