Fault Diagnosis in Gas Insulated Switchgear Based on Genetic Algorithm and Density- Based Spatial Clustering of Applications With Noise

As a kind of widely used switchgear in power system, the reliability of gas insulated switchgear (GIS) is very important for the safe operation of power systems. However, there is a lack of research on intelligent detection technology of mechanical state of GIS at present. A new method is urgently n...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE sensors journal Ročník 21; číslo 2; s. 965 - 973
Hlavní autoři: Yang, Yuan, Suliang, Ma, Jianwen, Wu, Bowen, Jia, Weixin, Li, Xiaowu, Luo
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York IEEE 15.01.2021
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Circuit faults
Classification
Clustering
Conductors
DBSCAN-based classification
Density
density-based spatial clustering of applications with noise (DBSCAN)
Fault detection
Fault diagnosis
Force
Gas insulated switchgear (GIS)
Gas insulation
genetic algorithm (GA)
Genetic algorithms
Switchgear
Vibrations
ISSN:1530-437X, 1558-1748
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:As a kind of widely used switchgear in power system, the reliability of gas insulated switchgear (GIS) is very important for the safe operation of power systems. However, there is a lack of research on intelligent detection technology of mechanical state of GIS at present. A new method is urgently needed to improve the operability, effectiveness, and accuracy of fault detection in GIS. Aiming at the abnormal vibration signals generated by GIS faults, this article presents a fault diagnosis method (GA-DBSCAN) consisting of a feature selection method based on genetic algorithm (GA) and Density-Based Spatial Clustering of Applications with Noise (DBSCAN), and a fault diagnosis method based on DBSCAN. First, this article analyzes the incentive force of GIS and discusses the characteristic frequency of response signal combining with the non-linear characteristics of a GIS system. Second, GA and DBSCAN are used to screen features for dimension reduction and get the optimized feature space, and DBSCAN-based classification is used to classify faults. Finally, optimized feature space is verified to be superior to the original feature space by typical classification method; the superiority and reliability of DBSCAN-based classification method under optimized feature space is verified by comparing with other classification methods. The proposed GA-DBSCAN approach can substantially increase the performance of the fault diagnosis method, which indicates that the method promotes development of intelligent detection technology of mechanical state in GIS. <graphic position="float" orientation="portrait" xlink:href="wu15-2942618.eps"/>
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1530-437X
1558-1748
DOI:10.1109/JSEN.2019.2942618