A 1D-CNN-SVM Model for Non-Intrusive Load Identification of Electric Bicycles

Abstract

The illegal charging behavior of electric bicycles leads to frequent fires, threatening the life safety of users. In order to improve the efficiency of load monitoring of electric bicycle charging behavior, a non-intrusive load monitoring method based on ID-CNN-SVM model is proposed in this paper. Firstly, one dimensional convolutional neural network is leveraged to extract load features from the input sequence data; Then, based on the features extracted by ID-CNN, support vector machine is used to identify the electric bicycle and other household appliances. The experiment is conducted with data sampled from data acquisition system, and the comparative experiment is performed with ID-CNN, LSTM, SVM and DT model. The results indicate that ID-CNN-SVM has good performance in the load identification of electric bicycles, verifying the effectiveness of the model.