Name
Title | ||
|---|---|---|
Detection and Location of Aged Cable Segment in Underground Power Distribution System Using Deep Learning Approach |
Abstract | ||
|---|---|---|
Aged cable segments in underground power distribution systems will gradually develop into severe faults. The location of aged cable segment is essential for ensuring system reliability and reducing maintenance cost, but is rarely studied by the existing research. In this article, a novel aged cable segment detection and location method based on wide frequency range transfer function (TF) measurement and deep learning approaches is proposed to fill the gap. The relationship between the measured TF and cable ageing parameters is analyzed, and two models with combined sparse autoencoder and convolutional neural network are trained to estimate the aging location and severity. A transfer learning approach is further developed to improve the model performance for various online situations. Experiments are conducted which prove the effectiveness and generalization ability of the proposed method. Compared with existing approaches, the proposed method shows better performance under different ageing situations, even smooth ageing scenarios. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1109/TII.2021.3056993 | IEEE Transactions on Industrial Informatics |
Keywords | DocType | Volume |
Aged cable segment location,convolutional neural network (CNN),power distribution system,transfer function (TF) measurement,transfer learning | Journal | 17 |
Issue | ISSN | Citations |
11 | 1551-3203 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yang Wu | 1 | 84 | 18.42 |
| Pinjia Zhang | 2 | 0 | 0.68 |
| Guanghui Lu | 3 | 0 | 0.34 |