TL-ResNet Load Classification Model Based on Time Series Imaging
Xiaozhuo Guan, Rao Fu, Jia Sun, Zizhang Yu, Tingkai Sun, Genhua Shi · 2024
The process of load classification is similar to that of image classification. The load current is transformed into image through time series imaging for load feature analysis. The image data set used in this paper is constructed based on different time series imaging methods using public data sets. Deep residual neural networks are trained and tested using publicly available data sets. The results show that the trained deep residual neural network can effectively complete the load classification and has a certain generalization ability.