Partial discharge (PD) detection has been proved as an effective tool for insulation condition monitoring of power equipment. The energy of PD pulses is valuable for studying the characteristics of PD activities. This paper proposes a method for estimating the energy of PD pulses in the presence of white noise and narrowband noise. First, a maximum likelihood (ML) estimator of the pulse energy is derived from the probability distribution of the energy spectral coefficients. To implement the ML method, the sampled data are divided into signal frames and noise frames. The noise frames are then utilized for extracting noise parameters using the 3F-C method. Eventually, these noise parameters are applied to the signal frames to find the ML estimate of the pulse energy. To verify the effectiveness of the proposed method, both simulated data and measured data have been processed using the proposed method and the conventional wavelet packet (WP) denoising method. Compared with the WP denoising method, the proposed method has a higher accuracy and is less susceptible to the lengths of the sampling time windows. The advantage of the method is more significant in unfavorable conditions where the signal-to-noise ratio is low and the accurate lengths of the PD pulses are difficult to determine.

Partial discharge (PD) detection plays a critical role in assessing the insulation condition of gas-insulated switchgear (GIS), a critical component in modern high-voltage power systems. While recent advancements in machine learning have facilitated the development of various models for phase resolved partial discharge (PRPD)-based PD recognition in GIS, conventional approaches predominantly utilize datasets collected under identical experimental conditions. This methodology induces similar spatial characteristic, such as discharge density and sampling rate, between training and testing sets, resulting in high accuracy metrics with limited generalization capabilities. To address this issue, this paper proposes DACNeXt, a domain-adaptive convolutional network designed for PRPD pattern recognition in GIS. The proposed model enhances conventional convolutional networks through three critical modifications: optimized feature extraction layers for PRPD characteristics, large-kernel attention mechanisms for discharge pattern recognition, and adversarial training through domain discriminators. The experimental framework employs a 252kV GIS platform to generate multi-condition datasets, with preprocessing procedures ensuring domain shift simulation between source and target distributions. Compared to previously proposed models for PRPD pattern recognition including GrabCut-GoogleNet, CNN, LBP-SVM, and RNet, DACNeXt achieves average accuracy improvements of 13.5%, 5.4%, 11.5%, and 16.2%, respectively. The network effectively identifies critical defect types including floating potential discharge, corona discharge, and surface discharge. These advancements contribute to improved GIS PD pattern recognition and early fault detection, enhancing the reliability and safety of high-voltage power systems.

Partial discharge (PD) leads to the generation of electrical, acoustic, optical, and thermal signals. The propagation characteristics of optical signals in gas insulated metal-enclosed transmission lines (GIL) are the basis of optical detection research. This paper simulates the propagation of PD optical signals in GIL through modeling GIL with different structures and specification parameters. By analyzing the optical parameters on the probe surface and the detection points when the PD source position is different, the influence of the difference in specifications caused by the voltage level on the propagation of the GIL PD optical signal is studied. The results show that the GIL cavity structure will affect the faculae distribution and the relative irradiance (RI) of the detection surface; the PD source position has a huge impact on the faculae distribution on the detection surface, but has little influence on the RI; as the voltage rises, the faculae distribution on the detection surface becomes more obvious, and the mean of RI decreases. The above results have the reference value for the manufacture of GIL equipment and the research of PD optical detection. When the specular reflection coefficient of surface material is smaller and the diffuse reflection coefficient is larger, the outline of the light spot is clearer, the proportion of brighter parts is larger, and the maximum value of the RI is larger.

Partial discharge (PD) detection has been proved as an effective tool for monitoring the insulation condition of high‐voltage cables. However, the detection can be seriously affected by pulse‐shaped interference from overhead lines (OHLs) when carried out at the cable termination of a combined cable–OHL. In this study, a novel scheme is presented for separating the PD pulses and interference in a combined cable–OHL. First, the characteristics of the currents from different pulse sources are investigated mathematically. On the basis of these characteristics, separation of multiple pulse sources can be realised by projecting the three‐phase power ratios extracted from detected signals onto a ternary plot and then constructing maps using the power ratios of the defective phase. The performance of the scheme is subsequently evaluated by both numerical simulations and field applications. Furthermore, a comparison with the well‐known time–frequency mapping method is provided. The results demonstrate that by using the proposed scheme, different pulse sources can be effectively identified without requiring stable pulse shapes and a better separation could be achieved compared with the conventional waveform‐based approaches.

This paper discusses the application situation of the ultra-high frequency (UHF) method and optical method in the detection of partial discharge of gas insulated switchgear (GIS), and the differences between external and internal UHF sensors. The sensor is a key technology in the PD detection method, and its performance directly affects the validity of the monitoring results. Different partial discharge detection sensors have different characteristics and working environments. UHF sensor has the advantages of wide detection range and convenient layout. Optical sensors have the advantages of high sensitivity and fast response speed. Six main types of internal sensors are evaluated and the performance indicators of internal sensors are discussed which includes disc antenna, loop antenna, dipole antenna, planar spiral antenna, conical antenna, photoelectric composite sensor. Furthermore, some problems need to be solved are proposed for further research, which is of great significance to the safe and stable operation of power equipment.

For 110 kV and above tunnel-installed high-voltage (HV) cross-linked poly-ethylene (XLPE) cable systems, it is a normal procedure to adopt a cross-bonding scheme. The high-frequency current method is frequently used in the cross-bonded cable systems for on-site or online partial discharge (PD) detection by monitoring the signals on the cross-bonding wires. To further study the amplitude distribution characteristics of the PD signals, a parametric characteristic admittance model of a three-phase cable system in a tunnel is established based on Tylavsky’s formulas. The model is used to calculate the amplitude distribution formula of the PD pulse current on the cross-bonding wires. In addition, the influence of cable laying and tunnel environment on the amplitude distribution is also studied. Finally, the correctness of the model and the conclusion are verified by simulation experiments and on-site tests. The results show that the signal amplitude distribution is determined by the ratio of the characteristic admittances. As the distance between the cables and the distance from the inner wall of the tunnel increase, the amplitude difference gradually decreases.