A Robust Transform-Domain Deep Convolutional Network for Voltage Dip Classiﬁcation
Journal article, 2018
This paper proposes a novel method for voltage dip classiﬁcation using deep convolutional neural networks. The main contributions of this paper include: 1) to propose a new effective deep convolutional neural network architecture for automatically learning voltage dip features, rather than extracting hand-crafted features; 2) to employ the deep learning in an effective twodimensional (2-D) transform domain, under space-phasor model (SPM), for efﬁcient learning of dip features; 3) to characterize voltage dips by 2-D SPM-based deep learning, which leads to voltage dip features independent of the duration and sampling frequency ofdiprecordings; and 4) to develop robust automatically-extracted features that are insensitive to training andtest datasets measured from different countries/regions. Experiments were conducted on datasets containing about 6000 measured voltage dips spread over seven classes measured from several different countries. Results have shown good performance of the proposed method: average classiﬁcation rate is about 97% and false alarm rate is about 0.50%. The test results from the proposed method are compared with the results from two existing dip classiﬁcation methods. The proposed method is shown to outperform these existing methods
convolutional neural network.