Towards Fault Diagnosis of Covered Conductors through Partial Discharge Classification using Antenna-based Frequency Domain Analysis

In this study, we investigate the use of contactless partial discharge (PD) detection using an antenna-based frequency domain analysis to advance fault classification of covered conductors, which are susceptible to insulation failure due to PDs. First, we conducted experiments to test PDs on a model that simulated PDs on contact using electrodes - needle-needle, needle-sphere, and needle with corona with and without grounding of the antenna. The results showed that the frequency domain analysis can distinguish between different types of contacts. Second, experiments were performed on covered conductors with three types of contacts: copper band, needle, and wire. Statistical analysis revealed a statistically significant difference between the types of contacts in certain frequencies, suggesting the ability to accurately detect and diagnose PDs using the antenna and frequency analysis. The proposed method has the potential to enable detection of different types of fault in covered conductors. These findings open a way to use software-defined radio (SDR) for contactless PD detection in covered conductors, which could lead to more effective, significantly cheaper, and efficient maintenance and monitoring of electrical systems. Also this study highlights the potential of antenna-based frequency domain analysis for advancing classification type of faults in covered conductors.