连续小波变换变压器绕组变形故障检测

Detecting Transformer Winding Deformation Fault Types Based on Continuous Wavelet Transform

利用脉冲频率响应法在线检测电力变压器绕组变形故障时,传统频率响应曲线并不能刻画暂态信号的时域特性,且快速傅里叶变换的局限性可能造成有用信息的缺失。本文提出利用连续小波变换处理脉冲在线注入法获得的暂态信号,研究利用小波时频图检测绕组变形故障类型。从理论推导入手,构建了单绕组仿真模型,开展了不同故障类型仿真分析;进行了绕组故障试验研究,分析检测信号时频特性,采用矩阵相似度量化评判检测信号的差异性。仿真分析与实验测试的数据处理结果,均初步证实了本方法的可行性。

When online detecting transformer winding deformation with impulse frequency response analysis, the time domain characteristics of transient signal could not be depicted by conventional frequency response curves, and the limitations of Fast Fourier Transform could result in deletion of available information. In this paper, continuous wavelet transform is used to process transient signal, which is obtained by online impulse signal injection method, time-frequency spectrum of wavelet transform is studied to detect transformer winding deformation fault types. A single winding simulation model is constructed after theoretical deduction, the simulation analysis of different fault types is performed, the typical winding fault test is conducted, the time-frequency characteristic of detection signal is analyzed, the matrix similarity is used to quantize the diversity of the detection signal. Both the simulation analysis and experimental studies show the feasibility of the proposed method.