The construction of ubiquitous power Internet of things makes it possible to provide users with more diversified and personalized service. In recent years, with the rapid development of the coal-to-electricity project, how to identify the coal-to-electricity users through the user load data has become a research hotspot. Firstly, this paper deeply analyzes the achievements and problems of the coal-to-electricity project at this stage. Using big data technology and ubiquitous power Internet of things technology can solve the contradictions and problems in the process of coal-to-electricity. Taking the load characteristics of coal-to-electricity users in a certain area as an example, the change of load characteristics after heating with regenerative electric boiler is described. By constructing the support vector machine model after particle swarm optimization, the typical daily power load data of a certain area in winter are identified and classified. Through the verification of the test set, the model established in this paper has high recognition accuracy, with an average accuracy rate of 98%, which has a certain practical value.