基于EEMD与KICA的局部放电窄带干扰抑制

Suppressing Periodic Narrowband Noise in Partial Discharge Signal Using EEMD and KICA

窄带干扰严重时可能完全淹没局部放电信号,给局部放电检测工作带来巨大困难。针对局放信号经验模态分解模态混叠现象,提出采用集合经验模态分解与核独立分量分析抑制局部放电窄带干扰的方法。在局放信号叠加不同频率、不同幅值的窄带干扰情况下,该方法能较好地提取出局放信号,有效提高信噪比且信号失真小。仿真和现场测试信号的处理结果均验证了所提方法的有效性。

It is difficult to measure the partial discharge(PD) signals when the narrowband noises are severe. Aiming at the mixed mode phenomena of the empirical mode decomposition method of PD, this paper puts forward a new method that ensemble empirical mode decomposition with kernel independent component analysis for de-noising in order to extract the PD signals from narrowband noise. Under the condition of adding different amplitude or different frequency narrowband interference, the method is introduced to separate the PD signals from the narrowband interferences and improves signal to noise ratio(SNR) and the signal distortion is small. The proceeding result of simulation model and signal from field test proves its effectiveness.