ABSTRACT

ABSTRACT:   Load identification is a major concept in the field of smart homes and smart grids. Nonintrusive Load Monitoring (NILM) method is applied to solve this problem, which is performed by analyzing the total current and voltage signal of the main distribution board to estimate the energy consumption of individual appliance and turning on/off or other operation. In this paper, we used the theory of NILM to identify household electric load. By analyzing the total current signal, extracting related features, and using Genetic Algorithm (GA) and Support Vector Machine (SVM), we identify different electric loads. We also use the BLUED data set (Anderson et al. 2012) as the experimental data set. Finally, rationality and effectiveness of the proposed method was verified by MATLAB simulation.