LAUSR

Abstract Non-intrusive load monitoring technology is the key technology to realize smart grid construction. The identification and classification of similar electrical appliances and low-power small-current electrical appliances in non-intrusive load monitoring has always been the focus of research and the difficulty of breakthrough. This paper uses current harmonic amplitude as classification data and based on the composition characteristics of harmonic amplitude data, and the current situation of fuzzy classification of similar electrical appliances, pointing out a method for enhancing the characteristic difference of harmonic amplitude data. This method can improve the current situation that the low-power electric load is difficult to identify when the low-power electric load and the high-power electric load work together. The status quo of identification has a good effect on the classification of low power and similar electrical load. In this paper, a non-intrusive load monitoring system is built, the bus current data is collected, and a feature database is built, and the network is used to classify the feature data and finally obtain the classification result. The experimental results show that the proposed method has a significant effect on the accuracy of classification and recognition of similar electrical appliances with similar eigenvalues.