Disclosed is an electric heating subdivision method based on historical load recognition data under a cloud-side collaborative architecture. The method comprises the steps: collecting electrical load historical recognition result sample data; extracting electric heating recognition result data, and summarizing and compiling statistics on same; acquiring recognition data of the following day of a resident user; extracting the electric heating recognition result data, performing statistical calculation on the data, and determining whether the data meets an electric heating subdivision rule; and if so, subdividing an electric heating label of the day of the resident user into a specific label, and updating the data subjected to summarization and statistics compilation such that the data participates in electric heating subdivision of the following day, etc. According to the present invention, in conjunction with characteristics such as the power and running time of an electric kettle, an electric cooker, an electric oven and an electric water heater, and the living habits of residents, an electric heating main class is subdivided into subclasses of electric appliances under a cloud-side collaborative architecture, such that a user can understand the power utilization situation, rationally schedule the starting time of an electric appliance and respond to step tariffs, the electric energy consumption is reduced to the maximum extent, and a corresponding auxiliary judgment can be made for fault diagnosis thereof.
展开▼