湖南电力 ›› 2025, Vol. 45 ›› Issue (5): 112-116.doi: 10.3969/j.issn.1008-0198.2025.05.015

• 电力人工智能与数字化 • 上一篇    下一篇

基于异常值检测的电压互感器在线监测方法

王尧平1,2, 梁文韬1,2, 谭海波1,2, 龙乙林1,2, 谈丛1,2, 王海元1,2   

  1. 1.智能电气量测与应用技术湖南省重点实验室, 湖南 长沙 410007;
    2.国网湖南省电力有限公司供电服务中心(计量中心), 湖南 长沙 410007
  • 收稿日期:2025-01-16 修回日期:2025-07-21 发布日期:2025-11-11
  • 通信作者: 王尧平(1999),男,硕士,主要从事电压互感器在线监测方面的研究工作。

Online Monitoring Method for Voltage Transformers Based on Anomaly Detection

WANG Yaoping1,2, LIANG Wentao1,2, TAN Haibo1,2, LONG Yilin1,2, TAN Cong1,2, WANG Haiyuan1,2   

  1. 1. Hunan Provincial Key Laboratory of Intelligent Electrical Measurement and Application Technology, Changsha 410007, China;
    2. State Grid Hunan Electric Power Company Limited Power Supply Service Center (Metrology Center), Changsha 410007, China
  • Received:2025-01-16 Revised:2025-07-21 Published:2025-11-11

摘要: 为实现对计量电压互感器状态的有效在线监测,对群体互感器电压比差数据特征进行理论分析。基于3σ准则提出互感器异常检测方法,开发电压互感器在线监测系统,可有效经济地实现对互感器异常的识别。现场验证结果表明:3σ区间内比值差数据样本占比为99.52%,具有正态分布特性;超差互感器异常概率在40%~80%之间波动,显著高于正常互感器。

关键词: 电压互感器, 异常检测, 3σ准则;, 正态分布

Abstract: In order to achieve effective online monitoring of the state of metering voltage transformers, the characteristics of group transformer voltage ratio difference data are theoretically analyzed. An anomaly detection method for voltage transformers is proposed based on the 3σ criterion, and an online monitoring system for voltage transformers is developed, which can effectively and economically realize the identification of transformer anomalies. The field data verification demonstrated that the sample proportion of ratio difference data within the 3σ interval is 99.52%, confirming the gaussian distribution characteristics. The anomaly probability for out-of-tolerance transformers fluctuates between 40% and 80%, significantly higher than that of normal transformers.

Key words: voltage transformer, anomaly detection, 3-sigma rule, gaussian distribution

中图分类号: