Non-intrusive electric heating load identification method based on Bayesian classification

This paper provides a non-intrusive electric heating load identification technique based on Bayesian classification. The customer side is an important component of the smart grid. In winter, a large number of electric heating devices at the customer side can affect the quality of power supply. So it is essential to pay attention to customer heating type data for providing better electric energy service. Main work of this paper: according to the historical data of several commonly used electric heating equipment, a data set is established and the electricity consumption characteristics of various types of electric heating equipment are analyzed. And the data set is preprocessed and trained to establish a Bayesian classification model. Three types of steady-state features with low requirements for sampling frequency are selected to achieve recognition with high accuracy.