基于Seasonal Hybrid ESD算法的电力变压器噪声检测方法研究  

Research on Power Transformer Noise Detection Method Based on Seasonal Hybrid ESD Algorithm

作　　者：罗浪 赵泽予 武晓蕊 杨丰帆 荣仪 LUO Lang;ZHAO Ze-yu;WU Xiao-rui;YANG Feng-fan;RONG Yi(Maintenance Company of State Grid Hubei Electric Power Co.,Ltd.,Wuhan 430050 China;School of Electrical Engineering,Nanchang Institute of Technology,Nanchang 330099 China)

机构地区：[1]国网湖北省电力有限公司超高压公司,湖北武汉430050 [2]南昌工程学院电气工程学院,江西南昌330099

出　　处：《自动化技术与应用》2025年第2期105-109,共5页Techniques of Automation and Applications

基　　金：国家自然科学基金项目(62001202)。

摘　　要：为提高在实际仅有少量样本情形下的变压器时序数据异常检测正确率,提出一种基于混合季节性极端偏差检验算法(Seasonal Hybrid ESD,S-H-ESD)的电力变压器噪声信号分析方法。首先通过STL将原始时序数据分解为长趋势、周期和残差分量,再采用S-H-ESD检验残差分量中的异常,将偏离中位数超出阈值的数据点作为变压器噪声信号的异常点,实现原始噪声信号数据中的异常点定位,最后通过对某500 kV变电站主变实测样本集数据,将所提方法与Prophet方法进行对比试验,结果表明:所提方法在小样本数据下异常点检测正确率高达到99.67%,为电力变压器运行状态噪声检测应用提供了小样本场景下的有效方法。In order to improve the accuracy of anomaly detection of transformer timing data when there are only a few samples in practice,this paper proposes a power transformer noise signal based on the Seasonal Hybrid ESD(Seasonal Hybrid ESD,SH-ESD)inspec-tion algorithm.It analysises method to locate the abnormal collection point of the original noise signal.Based on the measured sample sets data of the main transformer of a 500 kV substation,the method in this paper is compared and analyzed with the Prophet method.The results show that the method has a high accuracy of 99.67%in detecting outliers in small sample data,which provides a method for the application of noise in the operating state of power transformers.

关 键 词：电力变压器 噪声信号 S-H-ESD算法 异常点检测 小样本分析 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]