Comparison of non-intrusive load monitoring supervised methods using harmonics as feature

Non-intrusive Load Monitoring (NILM), also known as energy disaggregation, is a useful technique for analyzing energy consumption data, monitored from a single-point source such as a smart meter. In this paper, a three-phase induction motor was designed in SIMULINK to be used for NILM system based on current waveforms and odd-numbered harmonics up to the ninth harmonic. Supervised learning classifiers were proposed including decision tree, KNN, NN, Ensemble, and SVM algorithms to classify the loads with high accuracy. In comparison, results show that the decision tree classifier can classify the loads efficiently for the most loads. Although the ensemble showed a high accuracy but still needs more time for training due to the complexity of the model. Additionally, the more samples obtained the more accuracy of classification, but a high sampling rate has more cost and analysis it takes more time for training.