Sinusoidal noise suppression technique for partial discharge measurement of transformers

Partial discharge (PD) measurement system based on high frequency current transducer (HFCT) provides a convenient and cost-effective means for online insulation condition monitoring of transformers. However, it is a difficult task to extract PD signals from the measured signals overwhelmed by noise, which consequently leads to ambiguities in the condition assessment of transformers. This paper develops a novel noise cancellation technique for PD signals extraction during online PD measurement of transformers. This technique is based on an iterative multiplication process, which can suppress sinusoidal noise while maintains PD signals. The proposed technique will be verified through case studies on PD measurements performed on experimental models in laboratory and transformers at substations.