A power-based fault direction estimation method for active distribution networks

This paper presents a strategy for estimating fault direction in active distribution networks, focusing on the high penetration of distributed energy resources, changing operating conditions, untransposed lines, and unbalanced loads. The proposed strategy introduces a power-based directional criteri...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Results in engineering 2024-03, Vol.21, p.101754, Article 101754
Hauptverfasser: Muñoz-Arango, Giovanni, Mora-Flórez, Juan, Pérez-Londoño, Sandra
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper presents a strategy for estimating fault direction in active distribution networks, focusing on the high penetration of distributed energy resources, changing operating conditions, untransposed lines, and unbalanced loads. The proposed strategy introduces a power-based directional criterion derived from analysing active power within three-phase unbalanced systems known for their inherent unidirectional nature. The method enables rapid direction estimation by measuring the difference between pre-fault average active power and during-fault average active power values, utilising two moving windows. Notably, this difference is positive for forward faults and negative for backward faults, offering clear directionality insights. The strategy is validated using a modified IEEE 34-bus test system, where it demonstrates feasibility under several fault scenarios, including all fault types, varying fault impedance, operating conditions, noise levels, and sampling frequencies. Additionally, the proposed algorithm is compared with contemporary and conventional approaches, highlighting its effectiveness. •The proposed method is suitable for unbalanced distribution systems.•The directional judgement is made immediately after fault detection (approximately 1 ms).•The method uses local measurements.•The method guarantees a low computational burden.
ISSN:2590-1230
2590-1230
DOI:10.1016/j.rineng.2024.101754