Identification of Partial Discharge Sources by Feature Extraction from a Signal Conditioning System.

This paper addresses the critical challenge of detecting, separating, and classifying partial discharges in substations. It proposes two solutions: the first involves developing a signal conditioning system to reduce the sampling requirements for PD detection and increase the signal-to-noise ratio. The second approach uses machine learning techniques to separate and classify PD based on features extracted from the conditioned signal. Three clustering algorithms (K-means, Gaussian Mixture Model (GMM), and Mean-shift) and the Support Vector Machine (SVM) method were used for signal separation and classification. The proposed system effectively reduced high-frequency components up to 50 MHz, improved the signal-to-noise ratio, and effectively separated different sources of partial discharges without losing relevant information. An accuracy of up to 93% was achieved in classifying the partial discharge sources. The successful implementation of the signal conditioning system and the machine learning-based signal separation approach opens avenues for more economical, scalable, and reliable PD monitoring systems.

Design and Application of a Metamaterial Superstrate on a Bio-Inspired Antenna for Partial Discharge Detection through Dielectric Windows.

The adaptation of dielectric windows as metamaterial superstrate over a bio-inspired Printed Monopole Antenna (PMA) was evaluated in order to improve the detection sensitivity of Ultra High Frequency (UHF) sensors designed for Partial Discharge (PD) measurement. For this purpose, rectangular and circular Split Ring Resonators (SRR) structures were designed and evaluated aiming to achieve a metamaterial superstrate that improves the characteristics of the bio-inspired PMA as the gain, bandwidth, and radiation pattern. Measurements of the PMA with metamaterial superstrate were carried out in an anechoic chamber and compared to the simulations performed. The results show that the metamaterial superstrate insertion did not impact the original operating bandwidth, covering most of the characteristic frequency range of PD activity. Moreover, this insertion resulted in a mean gain enhancement of 0.7 dBi regarding the reference PMA, resulting in an antenna with better sensitivity for PD detection (mean gain of 3.61 dBi). The PMA-metamaterial set PD detection sensitivity was evaluated through laboratory tests with a point-to-plane PD generator setup and in field with measurements from a 230 kV current transformer. The developed PMA-metamaterial set was able to detect, successfully, the activity of PD for both tests, being classified as an optimized sensor for PD detection through dielectric windows.