Data Mining Based Partitioning of Dynamic Voltage Control Areas and Contingency Clustering

Partitioning of dynamic voltage control areas (DVCAs) and contingency clustering have attracted increasing attentions in power system planning. In this paper, we propose a data mining based method to recognize behavior patterns of buses and contingencies from offline simulation, so as to identify DV...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE Power & Energy Society General Meeting S. 1 - 5
Hauptverfasser: Wu, Liang, Guan, Lin
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.08.2018
Schlagworte:
ISSN:1944-9933
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Partitioning of dynamic voltage control areas (DVCAs) and contingency clustering have attracted increasing attentions in power system planning. In this paper, we propose a data mining based method to recognize behavior patterns of buses and contingencies from offline simulation, so as to identify DVCAs and group contingencies. The voltage control ability index (VCAI) is defined firstly to evaluate the control effect of a bus with VAR injection subject to a contingency. By traversing all the influencing factors of VCAI, including contingency, controlling bus, and observed bus, a data pool of VCAI is obtained. Behavior patterns of bus and contingency are then extracted from the data pool, respectively. Similarity metric for behavior pattern is defined and the affinity propagation clustering algorithm is adopted to cluster buses and contingencies, so as to form DVCAs and contingency clusters, respectively. Silhouette coefficient analysis is applied to determine a proper clustering scheme. The proposed approach is tested on a modified NE 39-bus system to validate its effectiveness.
ISSN:1944-9933
DOI:10.1109/PESGM.2018.8586496