局部放电在线监测中小波阈值去噪法的最优阈值自适应选择

小波系数阈值是决定小波阈值去噪法处理后的去噪信号失真度及误差的关键因素。针对局部放电脉冲频谱特征,提出用于局部放电信号去噪的小波最优阈值选择方法,阐述了自适应阈值搜索中的尺度选择方案;采用四种人工模拟加噪信号和两种典型局部放电脉冲仿真信号对该方法的去噪效果进行了定量分析,并采用了自适应阈值法对实测局部放电信号进行了处理,结果表明,该方法在局部放电在线监测中具有良好的去噪效果和实用价值。

Adaptive Optimal Threshold Selection of Wavelet-Based Threshold De-noising for On-line Partial Discharge Monitoring

The thresholds of wavelet coefficients are key factors in close relation with the distortion degree and error of a de-noised signal for wavelet shrinkage methods. Based on the characteristics of partial discharge (PD) frequency spectrums, an optimum method to select wavelet thresholds for de-noising of partial discharge signals is proposed. The scheme to determine wavelet scales for adaptive threshold search is interpreted as well. Four signals to artificially simulate the addition of noises and two typical simulative PD pulses are used to quantitatively analyze the de-noising results of the proposed method. The measured PD signals are processed by adaptive threshold method, the results show that the proposed method possesses well de-noise effect and practicality for online PD monitoring.