Partial Discharge Research Articles

Dataset for Antenna-Based Detection of Fault Types in Covered Conductors for 22 kV Voltage Power Lines

This abstract presents a dataset for the detection of fault types in XLPE-covered conductors utilized in 22 kV medium voltage power distribution systems. We employed an antenna-based approach for detecting partial discharges. The dataset encompasses 12 distinct fault categories, ranging from ground phase faults to inter-phase faults, and no-fault case with steel or covered conductor as fault. We also used three different antennas. Each sample is a single measurement from antenna, consisting of 106 data points as floating numbers. The utilization of the antenna-based method offers the potential for a more cost-effective and straightforward installation for the detection of partial discharges. The objective of dataset is to enhance the identification of fault types, thereby promoting broader adoption of covered conductors in overhead power distribution lines. Such adoption proves particularly beneficial in confined areas, including natural parks, where safety is a prime concern. It is noteworthy that this dataset represents an original contribution, as no prior publication has addressed detection for this specific range of fault types and method of detection.

Optical Detection and Characteristics of PCB Partial Discharge in More-Electric Aircraft Under Square Wave Voltage

Optical Detection and Characteristics of PCB Partial Discharge in More-Electric Aircraft Under Square Wave Voltage

A Denoising Method for Partial Discharge Ultrasonic Signals in GIS Based on Ultra-High Frequency Signal Synchronization

A Denoising Method for Partial Discharge Ultrasonic Signals in GIS Based on Ultra-High Frequency Signal Synchronization

Development and Validatiton of Estimation of Breakdown Voltage of Stator Windings of Generator from Partial Discharge Magnitude

Development and Validatiton of Estimation of Breakdown Voltage of Stator Windings of Generator from Partial Discharge Magnitude

A Novel Method to Enhance Partial Discharge Localization Accuracy Using Sectional Winding Analysis and a Neural Network-Based Learning Vector Quantization Model

Timely detection of insulation faults in power transformers is essential to prevent damage, outages, and financial losses. Locating partial discharge (PD) sources in transformers with complex windings poses a significant challenge, leading to insulation failure. This paper presents a novel approach combining a learning vector quantization (LVQ) network with frequency response analysis (FRA) to locate PDs along the windings, even in noisy environments, without using noise suppression techniques. The method obtains sectional winding transfer functions (SWTFs) using vector fitting (VF). A Gaussian function is injected into each SWTF, and the responses are captured as PD reference signals to train the LVQ network. In addition, a 5000 pC PD test pulse from a calibrator is injected into randomly selected laboratory winding sections, while a traveling rectangular wave is injected into the estimated SWTFs, obtained from FRA using vector fitting. The resulting responses are utilized as PD test signals to assess the proposed method across various PD pulse waveforms. The simulation and experimental results demonstrate that the proposed method achieves superior performance, particularly in high-level background noise, leading to significantly improved PD localization accuracy. Finally, the method's efficiency is compared with previous studies, particularly regarding performance in noisy conditions.

Accurate Identification Partial Discharge of Cable Termination for High-Speed Trains Based on S-Transform and Two-Dimensional Convolutional Network Algorithm

Accurate Identification Partial Discharge of Cable Termination for High-Speed Trains Based on S-Transform and Two-Dimensional Convolutional Network Algorithm

Dimensionality Reduction and Clustering Strategies for Label Propagation in Partial Discharge Data Sets

Deep learning approaches have been successfully applied to perform automatic classification of phase-resolved partial discharge (PRPD) diagrams. Under the supervised learning paradigm, however, the performance of classifiers strongly depends on the availability of large and previously labeled data sets. Labeling is an intensive and time-consuming labor, typically involving the manual annotation of a large number of data examples by an expert. In this work, we propose a label propagation algorithm applied to PRPD data sets, aiming to reduce the time necessary to manually label PRPDs. Our basic pipeline is composed of three phases: pre-processing, dimensionality reduction procedures, and clustering. Different configurations of the basic pipeline are tested by using PRPDs obtained from online measurements in hydrogenerators. The performance of each configuration is assessed by using the Silhouette, Caliński–Harabasz, and Davies–Bouldin scores. The clustering of the best three configurations is compared with annotated PRPDs by using the Fowlkes-Mallows index. Results suggest our strategy can substantially decrease the time for manual labeling.

Analysis of electric field and leakage current of glass and porcelain insulators under clean and polluted conditions: A comparative study of three profiles

Electric field (EF) enhancement in insulating systems may occur due to both internal and external factors including inefficient design, manufacturing defects, and depositions of pollutants. Consequently, phenomena like partial discharge, unexpected flashover, and even breakdown of the insulator may take place. In this work, aiming to suggest relatively better performance, a comparative study of porcelain and glass insulators under clean and polluted conditions was conducted using finite element-based software, COMSOL Multiphysics. Three profiles of both the insulators namely, standard, anti-fog, and open were modelled, and the effects of their geometry and varying electric conductivity of the surrounding air were analyzed on EF distribution and leakage current (LC). These parameters were also investigated under the deposition of water droplets and various pollution scenarios. Results of the performed computations demonstrated that by making air conductivity high, LC enhanced significantly while EF was affected the least. The presence of water droplets in the vicinity of live electrode and deposition of pollution near the end fittings created the strongest impact on EF distribution. Among the studied configurations, standard profile of porcelain material appeared to be the most promising for suppressing effects of the simulated stresses.

Online High Frequency Impedance Identification Method of Inverter-Fed Electrical Machines for Stator Health Monitoring

In electric powertrain traction applications, the adopted trend to improve the performance and efficiency of electromechanical power conversion systems is to increase supply voltages and inverter switching frequencies. As a result, electrical machine conductors are subjected to ever-increasing electrical stresses, leading to premature insulation degradation and eventual short-circuits. Winding condition monitoring is crucial to prevent such critical failures. Based on the scientific literature, several methods can be used for early identification of aging. A first solution is to monitor partial discharges. This method requires the use of a specific measurement device and an undisturbed test environment. A second solution is to monitor the inter-turn winding capacitance, which is directly related to the condition of the insulation and can cause a change in the stator impedance behavior. Several approaches can be used to estimate or characterize this impedance behavior. They must be performed on a machine at standstill, which limits their application. In this paper, a new characterization method is proposed to monitor the high-frequency stator impedance evolution of voltage source inverter-fed machines. This method can be applied at any time without removing the machine from its operating environment. The range and accuracy of the proposed frequency characterization depend in particular on the supply voltage level and the bandwidth of the measurement probes. The effects of parameters such as temperature, switching frequency, and DC voltage amplitude on the impedance characteristic were also studied and will be presented. Tests carried out on an automotive traction machine have shown that the first two series and parallel resonances of the high-frequency impedance can be accurately identified using the proposed technique. Therefore, by monitoring these resonances, it is possible to predict the aging rate of the conductor.

Desorption of Antibiotics from Granular Activated Carbon during Water Treatment by Adsorption

Although desorption of adsorbed pharmaceuticals from granular activated carbon (GAC) may inadvertently lead to their partial discharge with adverse effects on aquatic environments, there have only been a few reports of this phenomenon. This study has investigated desorption of antibiotics vancomycin and rifampicin from activated carbon in aqueous media regarding contact time and pH regime. Various characterizations of the three types of GAC were investigated. Then, antibiotics were loaded on them via adsorption. Subsequently desorption and re-adsorption of antibiotics were quantified for a range of contact times and ambient pH values. Within the first hour of a reversed concentration gradient at neutral pH, desorption released 2% to 54% of previously adsorbed antibiotics to water, which were subsequently re-adsorbed within 24 hours to four weeks with less than 1% antibiotics remaining in the liquid phase. Lower desorption was positively associated with higher GAC mesopore content and larger specific surface area. Effects of the ambient pH regime varied between studied adsorbents. The results are evidence that mesopore content and pore size in relation to the kinetic diameter of adsorbate molecules are important determinants of the extent of antibiotic desorption from GAC and the rates of subsequent re-adsorption. Physisorption was the dominant mechanism involved in both processes. Observed proportions and rates of antibiotic desorption suggest that selection of GAC properties should also consider their effects on unintended desorption and the re-adsorption during treatment processes in order to minimize potential pollution discharge or promotion of antibiotic resistance during treatment processes.

A Detailed Review of Partial Discharge Detection Methods for SiC Power Modules Under Square-Wave Voltage Excitation

Silicon carbide (SiC) power modules are increasingly being used in high-voltage and high-frequency applications due to their superior electrical and thermal qualities. However, the issue of the partial discharge (PD) phenomenon raises serious reliability difficulties resulting in insulation failure, performance degradation, and potential device collapse. This paper provides a thorough assessment of the current PD detection strategies in SiC power modules. The issues provided by SiC devices’ distinct operational features, such as high switching frequencies and higher voltage stresses, which hinder PD detection and mitigation, are widely investigated. This review compares the effectiveness, benefits, and limitations of various detection methods, emphasizing the need for better strategies to ensure long-term reliability and performance. This study gives an in-depth overview of the numerous forms of PD phenomena that occur in power modules, including internal and surface discharges, as well as how they appear under various detection systems. It examines the performance of several methods for power module technologies such as SiC. To address these PD issues, this article proposes ways to improve reliability and detection accuracy.

Electrical Tree and Partial Discharge Characteristics of Silicone Rubber Under Mechanical Pressure

Silicone rubber (SIR) is a crucial insulating material in cable accessories, but it is also susceptible to faults. In practical applications, mechanical pressure from bending or shrinking can impact the degradation of SIR, necessitating the study of its electrical tree and partial discharge (PD) characteristics under such pressure. This work presents the construction of a test platform for electrical trees under varying pressures to observe their growth process. A high-frequency current transformer is used to measure PD patterns during tree growth, enabling analysis of the effect of PD on tree initiation and propagation under pressure. The experimental results demonstrate a significant decrease in tree inception probability and increase in PD inception voltage under pressure. The pressure also influences the tree structure and PD during the treeing process, where the longest tree with a branch-like structure appears under 800 kPa. The effect of pressure on electrical tree and PD characteristics can be attributed to changes in free volume, alterations in air pressure within the tree channels, and the affected charge accumulation.

Combining multi-level feature extraction algorithm with residual graph convolutional neural network for partial discharge detection

Partial discharge (PD) identification is critical for the insulation diagnosis of cable conductors; however, there is still scope for enhancement in the existing adaptive extraction capabilities and feature utilization for PD signals. In this context, this paper introduces a method that integrates a one-dimensional convolutional neural network (1D-CNN) with a residual graph convolutional neural network (ResGCN) to recognize PD signals. Noise analysis is performed using various combinations of mother wavelets, and Bayesian optimization is employed to mitigate background noise. Key features of PD signals are progressively abstracted through a 1D-CNN-based multilevel automatic feature learning method, while signal timing attributes are maintained to minimize manual intervention. The graph data is constructed using the signal feature matrix and the signal timing feature similarity matrix. This is followed by the development of a ResGCN utilizing a graph attention mechanism to integrate node feature information and the topology of the PD graph data. This approach aims to fully exploit the correlation between local regions of the feature space and the temporal numerical properties of the signals. Additionally, it jointly optimizes feature extraction and model classification to facilitate adaptive diagnosis. The method is validated with extensive real experimental data obtained from medium voltage overhead power lines. It demonstrates exceptional performance and practicality, achieving an accuracy rate of 97.3% and a recognition rate of 96.1% for PD samples, thus offering reliable theoretical support for effective PD detection.

PD-Free Design of Insulation Systems: An Application to Laminated Busbars

The reliability of components of industrial electrical assets fed by power electronics might be at risk due to the type and extent of electrothermal stresses. The move of power electronics toward higher levels of voltage, switching frequency, slew rate, and specific power increases the risk of partial discharge inception and thus of accelerated extrinsic aging and premature failure. The reaction to this challenge is to embrace the concept of partial discharge-free (PD-free) design and operation. This paper presents a PD-free approach to the design of laminated busbars, considering both AC and DC insulation subsystems, and focusing on surface insulation. The availability of a recently proposed model to estimate the inception field is a key tool. The model is validated through PD measurements performed on a laminated busbar, using new automatic software that can identify the type of source generating PD. Combined with electric field calculations, the model provides estimates of the PD inception voltage which are almost coincident with the measurement results. Inception voltages in the order of 10 kV and 20 kV have been observed for AC and DC excitation, respectively. In the case of DC supply, tests at different ambient temperatures, 25 °C and 60 °C, indicate that the inception voltage does not change significantly with temperature. Disposability, scalability to any voltage/power, and capability to work, potentially, for any other type of insulation system, are interesting features of the proposed approach, which are discussed in the paper.

Review on peak detect and hold circuits and their applicability in partial discharge detection

Abstract This article reviews various designs of electrical peak detect and hold circuits and tests their applicability for the detection of partial discharge pulses. Since partial discharges are very short pulses in the nanosecond range, the peak detector circuit must be very fast and thus work precisely over a wide bandwidth up to the GHz range. The article begins with a comprehensive literature review and gives an overview of the state of the art of peak detector circuits. As a result of the literature research, four different peak detector designs are identified that appear suitable for measuring partial discharges. These four circuits are then simulatively tested for their performance in measuring the pulse height of various partial discharge pulses. It turns out that two of the identified circuits are able to measure nanosecond pulses of small amplitude. These two circuits must be further optimized for partial discharge detection in the future. In addition, the circuit should be tested prototypically with real partial discharge pulses.

Multi-source partial discharge pattern recognition in GIS based on Grabcut-MCNN

Partial discharge (PD) surveillance constitutes a pivotal methodology for diagnosing insulation failures in electrical equipment. Enhancing comprehensively the precision of identifying PD anomalies in Gas Insulated Switchgear (GIS) is of paramount significance for ensuring the steady functioning of power grids. This study introduces a novel framework that integrates Phase-Resolved PD Graph Segmentation (PRPD-Grabcut) with a tailored MobileNets-based Convolutional Neural Network (MCNN) to classify GIS-related PD issues. Leveraging image segmentation via PRPD-Grabcut, crucial features are extracted from PRPD diagrams, which then facilitate the construction of the MCNN model. This model employs depth-wise separable convolutions alongside inverted residual architectures to tackle the vanishing gradient dilemma inherent in Deep Convolutional Neural Networks (DCNNs) during GIS PD pattern discernment. Upon the model's subsequent training and validation, empirical evidence illustrates that the PRPD-Grabcut-MCNN hybrid significantly alleviates the computational load and storage requisites of the model, concurrently enhancing the recognition precision and expediting the training process of the neural network. Relative to diverse established lightweight neural network architectures, MCNN manifests superior performance in terms of recognition accuracy, reduced cross-entropy loss, and expedited training duration.

Estimation of the Partial Discharge Inception Voltage of Electrical Asset Components at Variable Environmental Pressure: A Modelling Approach

Since partial discharges, PD, are a major accelerated degradation mechanism of organic electrical insulation systems, measuring partial discharge inception voltage, PDIV, of electrical asset components in aviation and aerospace is a fundamental tool to cope with life, safety, and reliability requirements. Partial discharge phenomenology and inception voltage depend on pressure, specifically, PDIV decreases with pressure. To avoid PD inception during aircraft or aerospace vehicle operation, the value of PDIV must be known at any pressure that electrical asset components will experience. However, a lack of experimental facilities adequate to test PD on real-size asset components might prevent from having PD-related information at pressures lower (or higher) than standard atmospheric pressure, SAP. This paper presents a heuristic approach, based on physics-derived PD field inception models, that allows the estimation of PDIV at low pressure to be carried out based on measurements made at SAP, when the typology of the defect causing partial discharge is known (from SAP PD measurements), but the precise type, size, and location of the PD-generating defect is unknown. It is shown that PDIV estimates obtained by the proposed models for internal and surface discharges are in good agreement with measured values in a range from SAP to 0.05 bar, testing simple insulation geometries but also real asset components, such as motors.

State primary special standard of impulse current unit in the range from 1.0 to 1.0·105 A GET 202-2024

Metrological assurance of immunity tests to impulse currents of natural and artificial origin is an urgent task in the development and production of aviation, rocket and space equipment and telecommunication equipment. The State primary special standard of lightning discharge impulse current unit GET 202-2012 provided the unity of measurements of impulse current parameters in the amplitude range from 1 to 100 kA and the time range from hundreds of nanoseconds to units of milliseconds. In order to provide metrological support for measuring instruments for impulse currents of electrostatic discharges and switching currents (rise time of the transient response of at least 1 ns, measurement range from 1 to 200 A), GET 202-2012 underwent a modernization cycle in 2020–2023. As a result of the modernization, the State Primary Special Standard of the unit of pulse current in the range from 1.0 to 1.0·105 A GET 202-2024 was approved, the lower limit of the range of reproduction of the unit of pulse current of which was 1 A with a pulse front duration of no more than 1 ns. Thus, the problem of ensuring the uniformity of measurements of electrostatic discharges and power switching impulse currents parameters was solved. To achieve the specified characteristics, GET 202-2024 includes devices for reproducing current pulses based on semiconductor keys and reference current shunts with a short rise time. The composition, structural diagram, technical and metrological characteristics of GET 202-2024 are given. GET 202-2024 ensures the uniformity of measurements of the parameters of pulse currents of electrostatic discharges, partial lightning discharges, as well as pulse currents generated by switching processes when solving problems of electromagnetic compatibility, creating new types of materials and coatings, in geophysical and meteorological research and observations, in healthcare.

Partial Discharge Data Augmentation and Pattern Recognition Method Based on DAE-GAN

Accurate identification of partial discharge (PD) and its types is essential for assessing the operating conditions of electrical equipment. To enhance PD pattern recognition under imbalanced and limited sample conditions, a method based on a Deep Autoencoder-embedded Generative Adversarial Network (DAE-GAN) is proposed. First, the Deep Autoencoder (DAE) is embedded within the Generative Adversarial Network (GAN) to improve the realism of generated samples. Then, complementary PD data samples are introduced during GAN training to address the issue of limited sample size. Lastly, the model’s discriminator is fine-tuned with augmented and balanced training data to enable PD pattern recognition. The DAE-GAN method is used to augment data and recognize patterns in experimental PD signals. The results demonstrate that, under imbalanced and small sample conditions, DAE-GAN generates more authentic PD samples with improved probability distribution fitting compared to other algorithms, leading to varying levels of enhancement in pattern recognition accuracy.

Discharge modes evolution characteristics of metal particle on the spacer surface in AC gas insulated switchgear

Abstract Flashover faults on gas insulated switchgear (GIS) insulators induced by metal particles occur frequently. Previous studies have obtained the characteristics of partial discharges (PDs) induced by metal particles on spacer surfaces, but these characteristics cannot explain the detection failure of ultra-high frequency (UHF) online monitoring. To enable further study of the PD characteristics induced by surface metal particles, the space electric field and a very-high-sensitivity pulse current (PC) measurement system were established. The electric field and PC characteristics of a PD induced by metal particle on the spacer surface of a 126 kV GIS were obtained. Two PD modes were found to be induced by surface metal particles. In addition to the typical pulse discharge (TP), there is a micro-discharge group (MG) mode with low apparent charge and long duration. The average apparent charge of the MG mode is approximately one-tenth of that of the TP at 0.4 pC. Its duration may extend to the millisecond level, causing significant distortion of the spatial electric field while hardly producing UHF signals. Moreover, increasing the applied voltage will increase the proportion of the MG within the total discharge, where this proportion can reach more than 90% before flashover, and the proportion of the discharge pulses that generates UHF signals is as low as 1%. The MG generation mechanism is analysed, the ion group stranded on the spacer surface by the TP changes the local electric field at the tip of the metal particle, which reduces the development length and apparent charge of the MG. Low apparent charge is cleared easily by the background electric field and thus the discharge interval is very short. This paper can provide an important basis for revealing the mechanism of GIS spacer surface discharge induced by metal particles and solving the effectiveness of PD monitoring devices.