Steady State Load Decomposition Method Combining Template Matching with K-Nearest Neighbor Algorithms

Abstract

Non-intrusive load monitoring and decomposition technology is one of the most important components of smart grid. It is the basis for in-depth analysis of user electricity information, which is of great importance to improve user quality of life and improve grid services. In this paper, a new load decomposition method combining template matching with k-nearest neighbor algorithm is proposed based on the study of steady-state load decomposition model. It solves the existing problems of unknown load, dependence on training data, and difficult identification of low-power devices. It uses the collected data of the actual operation of the electrical appliances and through adding unknown load of different power, the simulation was carried out by MATLAB to verify that the method in this paper is feasible in practical applications.