Article Details

Abstract

To identify the key risk factors influencing faults in GIS busbars of smart substations, this study constructs a 2×2×2 factorial design based on operational monitoring data from 300 GIS busbars across five ultra-high voltage substations of a provincial power company in East China over an 18-month period. A 2×2×2 factorial design is constructed, and the chi-square test, Mantel-Haenszel stratified analysis, and logistic regression are employed to systematically quantify the independent effects and interactions of three factors: voltage fluctuation, ambient temperature, and load level. The results indicate that voltage fluctuation is the most significant factor affecting GIS busbar faults (OR=3.23), followed by load level (OR=2.08) and ambient temperature (OR=1.88). Significant interactions exist among the three factors, with the combination of “high fluctuation + high temperature + high load” exhibiting a fault rate as high as 52.6%, which is 9.7 times higher than that of the optimal combination. Stratified analysis effectively controlled the confounding effect of load, and the main-effect logistic model achieved an AUC of 0.701, explaining 86.6% of the deviance, thereby meeting the requirements for primary on-site screening. Based on the evidence, a precision maintenance strategy of "voltage priority, load peak shaving, and temperature compensation" is proposed, providing a quantitative basis for condition-based maintenance of GIS equipment in smart substations.

Keywords

GIS busbar; Chi-square test; Logistic Regression; multi-factor interaction; smart substation

References

1. [1] Liu Gencai, Jin Xiangchao, Xie Zhiyang, etc Modeling and Analysis of Busbar Heat Transfer Model for Gas Insulated Switchgear [J]. Journal of Wuhan University (Engineering Edition), 2012, 45 (5): 643-647
2. [2] Huang Mengdan, Xiao Chun, Yang Zhining, etc Research progress on heating of GIS equipment [J]. High Voltage Apparatus, 2020, 56 (12): 24-33
3. [3] Wang Xiaoni Research on Thermal Fault Identification of Electrical Equipment in Intelligent Substations Based on Infrared Technology [J]. Automation Technology and Application, 2025, 44 (3): 168-170175
4. [4] Xing Zhihui Overall Design of Smart Grid Equipment Monitoring and Operation Big Data Platform [J]. Mechanical and Electronic, 2023, 41 (4): 41-45, 50
5. [5] Mei Shengwei, Wang Yingying, Chen Laijun Review of the Current Status and Prospects of Information Security Research in Smart Grid from the Perspective of Complex Networks [J]. High Voltage Technology, 2011, 37 (3): 672-679
6. [6] Tang Yi, Chen Qian, Li Mengya, etc A review of network attacks in the environment of power information physical fusion system [J]. Power System Automation, 2016, 40 (17): 59-69
7. [7] Liu Chang, Zheng Tao, Wang Zhihua, etc Real time identification algorithm for AC sampling anomalies in intelligent substations based on DBSCAN [J]. Power System Protection and Control, 2024, 52 (24): 140-148
8. [8] Wang Xiaoli, Sun Xiaoli Research on Fault Diagnosis and Classification of Relay Protection Based on Machine Learning [J]. Electric Transmission Automation, 2021, 43 (4): 17-20, 52
9. [9] Han Yu, Dong Bo A mechanical fault diagnosis method for high-voltage circuit breakers based on coil current [J]. High Voltage Apparatus, 2019, 55 (9): 241-246
10. [10] Xu Jiayan, Wu Fangying, Li Rui, etc Dynamic identification technology for secondary virtual circuit failure risk of intelligent substation relay protection [J]. Power System Protection and Control, 2025, 53 (1): 160-170
11. [11] Congratulations, Sun Wenkai, Jia Yafei, etc GIS equipment fault recognition technology based on deep learning [J]. Information Technology, 2021 (9): 57-61
12. [12] Yuan Wenhai, Liu Biao, Xu Hao, etc GIS partial discharge fault type recognition based on composite neural network [J]. Journal of Electric Power Science and Technology, 2021, 36 (4): 157-164

Funding

National Natural Science Foundation of China (11701274); Nanjing Institute of Technology Graduate Course Ideological and Political Demonstration Course (2025KCSZ08); Excellent Teaching Case of Graduate Students at Nanjing University of Engineering (2025JXAN05); Degree and Graduate Education Reform in Jiangsu Province (2025YJYJG10)

About the authors

Ru Zhengliang (1980-), male, Han nationality, from Yancheng, Jiangsu, associate professor, master's student, research direction: mathematical statistics.