| Autor: |
Liming Tan, Xianggen Yin, Jian Qiao, Yikai Wang, Wen Xu, Wei Chen |
| Jazyk: |
angličtina |
| Rok vydání: |
2022 |
| Předmět: |
|
| Zdroj: |
Energy Reports, Vol 8, Iss , Pp 1437-1445 (2022) |
| Druh dokumentu: |
article |
| ISSN: |
2352-4847 |
| DOI: |
10.1016/j.egyr.2022.02.194 |
| Popis: |
The frequent occurrence of ground faults in distribution networks seriously affects the power supply reliability, and the accurate fault section location method is the foundation of automatic distribution systems. Since there are location blind areas in distribution networks with the similarity characteristics distortion of the zero-mode current waveforms, the application of the existing location methods based on the zero-mode current similarity degree are limited. However, when a ground fault occurs in the location blind areas, the polarity relationship characteristics of the zero-mode current remains unchanged. But the location method only based on polarity relationship is seriously affected by the reverse connection of the CTs. Maximum mean discrepancy (MMD) method can comprehensively reflect the waveform similarity degree and polarity relationship between zero-mode currents. Based on the current relationship of the fault zero-mode equivalent circuit, it is concluded that the MMD value of the fault section is larger than that of the normal section in the fault line. On this basis, a small-current ground fault section location method for distribution networks based on MMD is constructed. Simulation results show that the proposed method can effectively eliminate the location blind areas, which can accurately locate the fault section under different line types, fault points and fault conditions. Moreover, it has strong robustness, which is less affected by arc faults and measurement noises. |
| Databáze: |
Directory of Open Access Journals |
| Externí odkaz: |
|
