Study on field demonstration of high‐sensitivity SF6 leakage detection method for gas insulated switchgear

SF6 gas is widely used in power equipment such as gas insulated switchgear or gas circuit breaker. However, its global warming potential is 23,500 times higher than that of CO2. In consideration of environmental protection, detection of SF6 gas leakage with high sensitivity is necessary. The authors...

Full description

Saved in:
Bibliographic Details
Published in:Electrical engineering in Japan 2021-03, Vol.214 (1), p.35-41
Main Authors: Tatemi, Masaru, Inami, Hisao, Rokunohe, Toshiaki, Hirose, Makoto
Format: Article
Language:English
Subjects:
Algorithms
Basic converters
Circuit breakers
Conductors
diagnosis
Environmental protection
Field tests
gas pressure
Gas temperature
gas‐insulated switchgear
Leak detection
Machine learning
Sensitivity
slow leakage
sulfur hexafluoride
Surface temperature
Switchgear
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:SF6 gas is widely used in power equipment such as gas insulated switchgear or gas circuit breaker. However, its global warming potential is 23,500 times higher than that of CO2. In consideration of environmental protection, detection of SF6 gas leakage with high sensitivity is necessary. The authors proposed an algorithm to accurately estimate the average gas temperature from the tank surface temperature for establishing a high‐sensitive SF6 slow leak detection technology. Since the gas temperature depends on the temperature of the conductor, the algorithm was characterized by estimating the gas temperature from the tank surface temperature and the load current. The relationship among the tank surface temperature, the gas temperature, and the current was preliminarily investigated by the use of the thermal fluid analysis and stored in the database. By use of this method, the fluctuation of the converted pressure in the basic examination test was suppressed to 20% as compared with that without considering the current. In addition, database was also created from the machine learning of field test result in substitution. In the field test, it was confirmed that the detection time of 0.5%/year slow leak could be detected in about 8 months by considering the load current.
ISSN:0424-7760
1520-6416
DOI:10.1002/eej.23295