一种用混沌振子去除局部放电信号中窄带干扰的新方法

A New Suppression Method for Narrow-Band Interference in Partial Discharge Signals With Chaotic Oscillator

如何有效地从现场大量的干扰中真实地提取局部放电信号,至今仍是提高在线监测系统的灵敏度和监测可靠性的难题。本文从非线性理论出发,提出用混沌振子去除局部放电信号中窄带干扰的新思路。讨论了用于判断混沌振子状态变化的Lyapunov指数的计算方法,仿真实验表明混沌振子对局部放电信号中的窄带干扰具有较强的敏感性,对局部放电信号和白噪声具有免疫力,能在保证原始局部放电信号失真较小的同时有效地去除窄带干扰。

On-line PD monitoring is a very powerful tool to detect defects in the insulation structure of electrical apparatus. The strong coupling of various noises greatly reduced the sensitivity of the measured PD and even prevented meaningful measurement. For the purpose of decreasing the distortion of PD pulse caused by noise removal, and suppressing narrow-band interference more effectively, chaos theory is introduced into research work. According to the phase track of chaotic oscillator changes from critical status to great scale period status, narrow-band interference with any frequency can be identified and suppressed. And the algorithm of the Lyapunov exponent is also discussed, which is one of the criteria of chaotic oscillator status variations. In simulation experiments, it is proved that chaotic oscillator is sensitive to narrow-band interference in partial discharge signals, and is immune to partial discharge signals and white noises. It can effectively identify and suppress narrow-band interference without any influence on PD pulse. This method has good application prospect because many kinds of periodic signals can be detected using the property of sensitive dependence on initial conditions of a chaotic oscillator.