Gas insulated composite electrical appliances (GIS) have the advantages of small footprint, compact structure, good electromagnetic compatibility, safety and reliability, so they are increasingly used in power systems. However, due to its complex structure, once GIS equipment malfunctions, the maintenance work is more complex and the cycle is longer than open equipment. GIS faults are mostly insulation faults, so insulation condition detection and fault diagnosis of GIS are of great significance for its safe and reliable operation. At present, the acceptance and debugging work for GIS before operation in China has been very complete, while the insulation condition detection of GIS during operation only adopts a single method of SF6 micro water detection. This method cannot directly reflect the insulation status of GIS. Although a large amount of research has been conducted domestically and internationally on partial discharge and SF6 decomposition product detection, and many achievements have been made, the reliability of the detection results still needs to be improved due to on-site interference, pattern recognition, and other reasons. Therefore, under the current technical conditions, in order to improve the level of GIS status detection and fault diagnosis, this article aims to study a comprehensive monitoring method that combines GIS partial discharge detection, SF6 decomposition product detection, and SF6 micro water detection, and guide actual production accordingly

This article first provides a detailed analysis of typical defect types in GIS and the current research status of GIS live detection at home and abroad. After comprehensive analysis and comparison of the advantages and disadvantages of various partial discharge methods, this article decides to use the ultra-high frequency method as a GIS partial discharge detection method, and provides a detailed introduction to the principle of the ultra-high frequency method and the device used in this article. This method has the advantages of high detection sensitivity, high fault recognition rate, and being able to detect early mild partial discharges. However, it cannot quantitatively detect partial discharges, and there are still certain blind spots in signal recognition. The content of SF6 decomposition products can reflect the insulation status and internal defects of GIS, and it can reflect the internal discharge and overheating faults of GIS. By conducting long-term quantitative analysis of the characteristic gases in the decomposition products of SF6, the development trend of the insulation status of the equipment can be monitored, and potential faults of the equipment can be identified. The paper also provides a detailed introduction to the most widely used SF6 micro water detection method. Considering the advantages and disadvantages of various GIS live detection methods, this article combines three live detection methods: ultra-high frequency, SF6 decomposition products, and micro water to comprehensively evaluate and diagnose the insulation status of GIS. This method avoids the shortcomings of a single method and improves the reliability of detection results