高斯尺度空间周期窄带局部放电带电检测滤波方法研究

Research on Gauss Scale Spatial Periodic Narrow Band Partial Discharge Live Detection and Filtering Method

为去除局部放电检测信号中的周期窄带干扰,本文基于快速傅立叶变换频域阈值算法,提出一种改进算法,即在高斯尺度空间上的FFT周期窄带滤波方法。先对放电信号的频谱进行高斯尺度空间平滑滤波,经过多次投影找到频谱上周期窄带干扰和局放信号所在的局部最大值,再通过峭度提取出周期窄带干扰的中心频率并滤除,最后通过IF FT重构出放电信号。然后对各种噪声强度下的局部放电信号进行评估,仿真和实测信号分析结果表明,在不同噪声水平下,该方法均能在抑制周期窄带干扰的同时保留局部放电信号特征。

In order to effectively eliminate the periodic discrete spectral interference in PD signals,Based on the research of the threshold value method of frequency domain in Fourire transform,an improved algorithm in Gaussian scale space is presented in this paper.Firstly,the spectrum of discharge signal is filtered by Gaussian scale spatial smoothing,and the local maximum of periodic narrow-band interference and partial discharge signal is found through multiple projections.Then the center frequency of periodic narrow-band interference is extracted by kurtosis and filtered.Finally,the discharge signal is reconstructed by IFFT.The simulation and experimental results show that the proposed method can effectively eliminate discrete spectral interference while preserving signals under different noise intensity.