Application of Chaotic Oscillator to Partial Discharge Signal Detection

Abstract

Partial discharge (PD) detection and recognition have become an important tool for the evaluation of insulating degradation in power transformer. How to detect the faint PD signal and suppress interference is the key to PD online measuring. The signal detection based on chaotic oscillator applies the bifurcation behavior of the driven Duffing oscillator. Chaotic system is sensitive to certain signal and immune to noise, the properties of which improve the detection sensitivity and antiinterference capability. In this paper, based on chaotic detection principle, a new PD signal detection method is presented. By Setting the appropriate periodic parameter disturbance to control the chaotic oscillator, the narrowband interference and periodic pulse interference can be accurately discerned and suppressed with the special chaotic system. The results of simulation and experiment prove this method is effective and without any influence to PD pulse.