Abstract | ||
---|---|---|
Partial Discharge Detection is an important method to find hidden dangers and safety problems of power equipment. The traditional Transient Earth Voltages (TEV) methods of the switchgear and GIS usually consist of four steps, i.e. data gathering, filtering, feature extraction and pattern recognition. These methods generate features using statistical analysis or waveform analysis of the signals, and then conduct the recognition and classification operations. In recent years, the increasingly developing deep learning methods have a huge effect on the pattern recognition, due to its outstanding abilities of recognition and feature extraction, which greatly offset the poor recognition performance of the traditional TEV methods. A novel Convolutional Neural Network (CNN) based TEV detection method is proposed in the paper, which needs no signal feature prepared by human and overcomes the detection problems resulted from an inappropriately selected feature. A CNN model is designed to train and classify the spectral image of the TEV. Although no sophisticated denoising method is adopted in the preprocessing, the proposed method approaches an extraordinary detection performance, which demonstrates the effectiveness of our CNN based TEV detection method. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1109/ISPDC.2016.65 | 2016 15th International Symposium on Parallel and Distributed Computing (ISPDC) |
Keywords | Field | DocType |
partial discharge detection,feature,deep learning,pattern recognition | Noise reduction,Computer science,Convolutional neural network,Real-time computing,Artificial intelligence,Deep learning,Pattern recognition,Feature (computer vision),Filter (signal processing),Feature extraction,Preprocessor,Time–frequency analysis,Machine learning | Conference |
ISSN | ISBN | Citations |
2379-5352 | 978-1-5090-4153-4 | 0 |
PageRank | References | Authors |
0.34 | 1 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yuan Lu | 1 | 1194 | 59.44 |
Ruifeng Wei | 2 | 0 | 0.34 |
Jingrui Chen | 3 | 0 | 0.34 |
Jun Yuan | 4 | 12 | 4.13 |