SVR-Based Current Zero Estimation Technique for Controlled Fault Interruption in the Series-Compensated Transmission Line
This paper presents a support vector regression-based postfault current zero estimation technique for single-line-to-ground faults on series-compensated transmission lines. The same can be used for controlled fault interruption, which, in turn, can help to control the arcing time of the circuit brea...
Gespeichert in:
Veröffentlicht in: | IEEE transactions on power delivery 2013-07, Vol.28 (3), p.1364-1372 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | This paper presents a support vector regression-based postfault current zero estimation technique for single-line-to-ground faults on series-compensated transmission lines. The same can be used for controlled fault interruption, which, in turn, can help to control the arcing time of the circuit breakers (CBs). The proposed technique is applied on a 400-kV series-compensated transmission line with series compensation placed at the middle of the line. Various power system parameters like fault resistance, load angle, fault inception angle, source impedance, fault location, and compensation level are varied to generate a large number of simulation cases involving single-line-to-ground faults. Since the incorrect current zero detection may cause an adverse impact on various CB components, especially contacts and parts of the interrupting chamber, the postfault zero estimation for each test case is performed multiple times using successive postfault sample windows of the faulted phase. The estimation performed in this way bestows correct zero crossing detection for more than 95% of test cases which proves the effectiveness of the proposed algorithm. The blind dataset and cross-validation technique have also been applied for validating the performance of the proposed algorithm. |
---|---|
ISSN: | 0885-8977 1937-4208 |
DOI: | 10.1109/TPWRD.2013.2247425 |