Online recognition method of transformer partial discharge based on audio detection
Source abstract
Abstract
This study aims at the problem of low accuracy and poor real-time of transformer partial discharge defection. This paper uses ultrasonic sensors for contactless detection and proposes an Ultra-Lightweight Convolutional Neural Network (UL-CNN). The UL-CNN can extract audio features during partial discharges to achieve the online detection of the transformer. Even in the case of a small number of training samples, the accuracy can reach 98.6%, the online recognition rate is nearly nine times faster than that of MobileNet, and the recognition accuracy and real-time performance are better than those of the classic lightweight network MobileNet.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: High voltage insulation and dielectric phenomena · Power Transformer Diagnostics and Insulation · Electrical Fault Detection and Protection
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
College of Urban Rall Transit and Logistics, Beijing Union University , Beijing 100101, China
Funding
Natural Foundation of Beijing · The Research of Beijing Union University Strategy of Strengthening Schools With Talents · The Research Project of Beijing Union University · The Science and Technology Program of Beijing Municipal Education Commission · Research on intelligent detection and algorithm of traction power supply fault of high-speed rail
References (17)
Show source-provided references
1. Power Syst. Prot. Control · 2017 · 45 · 74
2. High Voltage Eng. · 2015 · 41 · 258310.13336/j.1003-6520.hve.2015.08.011 ↗
3. Power Syst. Clean Energy · 2017 · 33 · 35
4. Power Syst. Clean Energy · 2017 · 33 · 17
5. Indian Standard · 2000 · 13
6. Power Syst. Prot. Control · 2018 · 46 · 55
7. Power Syst. Prot. Control · 2018 · 46 · 25
8. IET Sci., Meas. Technol. · 2018 · 12 · 103110.1049/iet-smt.2018.5020 ↗
9. High Voltage Appar. · 2017 · 53 · 7010.13296/j.1001-1609.hva.2017.05.011 ↗
10. Energies · 2019 · 12 · 248510.3390/en12132485 ↗
11. High Voltage Eng. · 2019 · 45 · 33710.13336/j.1003-6520.hve.20190130001 ↗
12. 2018 · 1
13. High Voltage Eng. · 2018 · 44 · 362510.13336/j.1003-6520.hve.20181031023 ↗
14. 2015 · 1984
15. Software Guide · 2017 · 16 · 198
16. Energies · 2018 · 11 · 120210.3390/en11051202 ↗
17. Sensors · 2018 · 18 · 351210.3390/s18103512 ↗