摘要
为能进一步提高变压器绕组状态检测的可靠性,本文提出一种高压直流作用下通过开关切换产生的暂态激励获得绕组振荡波来分析绕组状态的方法。基于现场试验,本文通过连续小波变换对振荡波信号进行去噪,并分析其时频特性,通过引入图像识别方法与总矩阵相似度量化时频信号特征。结果表明:该方法测试效率高,矩阵相似度可较好反映不同幅值、不同输出侧下振荡波的重复性和规律性,并能较好反映变压器相间绕组的差异性,为研究变压器绕组变形检测提供了参考。
In order to further improve the reliability of transformer winding state detection,this paper presents a method to analyze winding state by using winding oscillation wave under HVDC transient excitation.Based on field experiments,this paper denoises the oscillating wave signal by continuous wavelet transform,and analyses its time-frequency characteristics.By introducing image recognition method and total matrix similarity,the time-frequency signal characteristics are measured.The results show that the method has high testing efficiency,and the matrix similarity can reflect the repeatability and regularity of oscillation waves under different amplitudes and output sides,and can better reflect the difference of transformer windings between phases,which provides a reference for the study of transformer winding deformation detection.
作者
唐伟超
刘红文
王科
Tang Weichao;Liu Hongwen;Wang Ke(Yunnan Wenshan Electric Power Co.,Ltd.,Yunnan Wenshan 663000;Power science research institute of yunnan power grid co.LTD.,Kunming 650217,China)
出处
《云南电力技术》
2020年第1期91-97,共7页
Yunnan Electric Power
关键词
时频变换
变压器
振荡波
连续小波变换
矩阵相似度
time-frequency transformation
transformer
oscillating wave
continuous wavelet transform
matrix similarity