Time-domain double-ended phase selection method for AC/DC hybrid transmission system based on model recognition
Author:
Affiliation:

Clc Number:

TM773

Fund Project:

the Xinjiang Uygur Autonomous Region University Scientific Research Key Project (No. XJEDU2021I009)

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments
    Abstract:

    To solve the problem that the traditional single-ended phase selection elements are no longer suitable for AC long lines at the receiving end of the AC/DC hybrid system,a time-domain double-ended phase selection method based on model recognition is proposed. Firstly,the time domain model of thenon-fault phase and fault phase of long AC lines with shunt reactors at the receiving end during fault is studied by using the idea of model identification. It can be seen from the analysis that the differential current and differential voltage of the non-fault phase satisfy the capacitance model but the fault phase does not. Secondly,the capacitance model conformity of each phase is calculated by Spearman rank correlation coefficient,and the fault phase is selected by comparing it with the threshold value. Finally,an AC/DC hybrid system model is built in PSCAD/EMTDC,and the proposed fault phase selection method is simulated and verified from the aspects of fault type,fault location,transition resistance,sampling frequency and noise interference. The results show that the fault phase selection method can quickly and accurately select the fault phase during the fault at each position of the receiving end AC line,without relying on the characteristics of the DC-side power supply,without the influence of the shunt reactors,and with good anti-transition resistance and anti-interference performance.

    Reference
    Related
    Cited by
Get Citation
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:December 09,2022
  • Revised:March 06,2023
  • Adopted:November 21,2022
  • Online: May 19,2023
  • Published: May 28,2023