Power Quality Disturbances Classification Using A TCN-CNN Model

Accurate classification of different disturbance signals plays an important role in improving power quality. Existing methods either ignore the time dependence of the power quality disturbance signals or miss the latent features, which limits their performance. In this paper, a new power quality disturbances classification model is proposed to account for temporal dependence and latent features simultaneously. Specifically, the proposed model consists of a convolutional neural network (CNN) and a temporal convolutional network (TCN). The TCN is used to capture temporal dependencies and the CNN is employed to mine latent features. Then, a hybrid structure is presented to integrate these two parts. The simulation results show that the proposed model outperforms other models at different sampling frequencies and different levels of noises.