Electrical insulators are components used to promote electrical isolation and mechanical support for conductors in transmission lines belonging to the electric power systems (EPS). The performance of glass insulators is compromised mainly in the face of high pollution level, causing the amplitude of the leakage current to increase and the formation of superficial discharges. In this work, a methodology for estimating the surface pollution critical level of glass insulators using leakage current parameters and artificial intelligence algorithms is proposed. For this purpose, glass insulators were tested in the laboratory to simulate different levels of pollution, in order to record leakage current, voltage signals and environmental variables (e.g., humidity and temperature). The pollution level was simulated by depositing the pollution solution on the insulators, where each solution was differentiated based on the level of equivalent salt deposit density (ESDD). In the tests, the level of ESDD responsible for causing surface discharges was also recorded, as this phenomenon occurs at high levels of pollution. Pollution modeling was performed using forecasting algorithms such as multilayer perceptron (MLP) and support vectors regression (SVR). Based on the application of the best performance model and using criteria that indicate the occurrence of surface discharges, it was possible to estimate the critical level of pollution in the glass insulators string for 69 kV systems.

This work proposes a methodology for noise removal, separation, and classification of partial discharges in electrical system assets. Partial discharge analysis is an essential method for fault detection and evaluation of the operational conditions of high-voltage equipment. However, it faces several limitations in field measurements due to interference from radio signals, television transmissions, WiFi, corona signals, and multiple sources of partial discharges. To address these challenges, we propose the development of a clustering model to identify partial discharge sources and a classification model to identify the types of discharges. New features extracted from pulses are introduced to model the clustering and classification of discharge sources. The methodology is tested in the laboratory with controlled partial discharge sources, and field tests are conducted in substations to assess its practical applicability. The results of laboratory tests achieved an accuracy of 85% in classifying discharge sources. Field tests were performed in a substation of the Eletrobras group, allowing the identification of at least three potentially defective current transformers.

This work presents a methodology for the generation and classification of phase-resolved partial discharge (PRPD) patterns based on the use of a printed UHF monopole antenna and signal conditioning circuit to reduce hardware requirements. For this purpose, the envelope detection technique was applied. In addition, test objects such as a hydrogenerator bar, dielectric discs with internal cavities in an oil cell, a potential transformer and tip–tip electrodes immersed in oil were used to generate partial discharge (PD) signals. To detect and classify partial discharges, the standard IEC 60270 (2000) method was used as a reference. After the acquisition of conditioned UHF signals, a digital signal filtering threshold technique was used, and peaks of partial discharge envelope pulses were extracted. Feature selection techniques were used to classify the discharges and choose the best features to train machine learning algorithms, such as multilayer perceptron, support vector machine and decision tree algorithms. Accuracies greater than 84% were met, revealing the classification potential of the methodology proposed in this work.

Monitoring the degradation of insulation systems in high-voltage equipment relies on identifying physical, chemical, or electrical phenomena occurring during operation, with partial discharge (PD) activity recognized as a primary indicator of insulation system deterioration. This study introduces a novel methodology for denoising, wavefront identification, and transient segmentation of PD signals collected in laboratory and substation settings using a printed monopole antenna (PMA) operating at ultra-high-frequency (UHF). The proposed approach employs a shift-invariant wavelet denoising technique, where thresholds are estimated using the empirical Bayes method at each decomposition level of the wavelet transform. The denoised signal is subsequently processed through linear zero-phase filtering and post-processing steps to determine the start and end points of the PD event and to mitigate distortion effects introduced by denoising and other disturbances. The proposed methodology enables accurate identification of the first wavefront occurrence and segmentation of the PD event, offering enhanced precision for future studies on PD localization and classification. Experimental results showed substantial improvements in signal-to-noise ratio (SNR), high cross-correlation (above 0.9) between the original and denoised signals, and a significant reduction in normalized mean squared error (NMSE), confirming the robustness of the method under low-SNR conditions.

This work presents a methodology for the generation and classification of phase resolved partial discharges (PRPD) patterns based on the use of a printed UHF monopole antenna and signal con-ditioning circuit to reduce hardware requirements. For this purpose, envelope detection technique was applied. In addition, test objects such as a hydrogenerator bar, dielectric disks with internal cavities in an oil cell, potential transformer, and tip-tip electrodes immersed in oil were used to generate partial discharge (PD) signals. To detect and classify partial discharges, the standard IEC 60270 (2000) method was used as reference. After acquisition of conditioned UHF signals, digital signal filtering threshold technique was used, and peaks of partial discharges envelopes pulses were extracted. Feature selection techniques were used to classify the discharges and choose the best features to train machine learning algorithms, such as multilayer perceptron, support vector ma-chine and decision tree. Accuracies greater than 84% were met, revealing the classification potential of the methodology proposed in this work.