Partial Discharge Patterns Research Articles

Current sensor optimization based on simulated transfer function under partial discharge pulses

The time measurement efficiency of the partial discharge (PD) relies on the signal-to-noise ratio (SNR) and gain of the high-frequency current transformer (HFCT) sensor. However, the PD's time measurement efficiency decreases with the noise coupled to the HFCT in onsite measurements. To overcome that setback, this paper proposes one pre-processing, through modelling and simulation, considering the physical effects, features of the electrical circuit and coil construction parameters of the HFCT. The main goal is to reach reasonable high SNR under the strong influence of background noises. This investigation aims to validate the hypothesis of improvement or deterioration of the HFCT signal response through a transfer function optimization. This research effort's contributions are threefold: 1. Generation of PD pulse signal and noise addition; 2. HFCT modelling, simulation, and frequency response analysis; and 3. Models performance evaluation and validation of hypothesis. In conclusion, the pre-processing approach stands out as a means to robustify and provide freedom to the electric utility, making up for an eventual need to redefine the physical and geometrical parameters of the HFCT sensor under specific background noise for maintenance tests purpose. According to a cyber-physical system framework, experiments corroborate the project's goals to contribute to the PD pattern monitoring in onsite measurements and incorporate robustness to signals with low SNRs.

Autonomous tracking of partial discharge pattern evolution based on optical flow

Measurement of partial discharges (PD) is one of the key indicators of high voltage insulation integrity. They long term effect may lead to insulation deterioration and breakdown. This paper presents a novel technique for autonomous tracking of partial discharge image sequence evolution. The approach is based on motion detection utilizing optical flow algorithm. The goal of the presented research was to automate pattern analysis, leading towards intelligent PD diagnostics. Experiments have been performed on various classes of PD images representing monitoring sequence, partial discharges in stator bars representing electrical machine insulation, effect of harmonics and development of corona phases. Optical flow based detection was successfully applied to the sequence of PD images. Presented approach of autonomous tracking of partial discharge pattern evolution based on optical flow may be especially suitable for monitoring applications, however it can be also applied to off-line diagnostics. Motivation to apply motion analysis to a PD pattern is based on the conviction and direction of future autonomous AI-supported partial discharge diagnostic systems.

Dielectric Strength and Patterns of Partial Discharges in Nanocomposites Insulation of Three-Core Belted Power Cables

Dielectric Strength and Patterns of Partial Discharges in Nanocomposites Insulation of Three-Core Belted Power Cables

Classification of Superimposed Partial Discharge Patterns

Phase resolved partial discharge patterns (PRPD) are routinely used to assess the condition of power transformers. In the past, classification systems have been developed in order to automate the fault identification task. Most of those systems work with the assumption that only one source is active. In reality, however, multiple PD sources can be active at the same time. Hence, PRPD patterns can overlap and cannot be separated easily, e.g., by visual inspection. Multiple PD sources in a single PRPD represent a multi-label classification problem. We present a system based on long short-term memory (LSTM) neural networks to resolve this task. The system is generally able to classify multiple overlapping PRPD by while only being trained by single class PD sources. The system achieves a single class accuracy of 99% and a mean multi-label accuracy of 43% for an imbalanced dataset. This method can be used with overlapping PRPD patterns to identify the main PD source and, depending on the data, also classify the second source. The method works with conventional electrical measuring devices. Within a detailed discussion of the presented approach, both its benefits but also its problems regarding different repetition rates of different PD sources are being evaluated.

A Novel Diagnosis Method for Void Defects in HVDC Mass-Impregnated PPLP Cable Based on Partial Discharge Measurement

Mass Impregnated PPLP cable, which is applied to various high-voltage direct current (HVDC) projects due to its excellent dielectric and temperature properties, has a problem wherein voids are formed inside the butt-gap due to cavitation. However, there has been no previous research into technology for void defect identification and insulation diagnosis on HVDC MI-PPLP cables. In this paper, to propose an insulation diagnosis method for void defects in HVDC MI-PPLP cable, the direct current (DC) void discharge patterns were analyzed according to the specimen temperature and the magnitude of applied voltage using the pulse sequence analysis method. In addition, to confirm the pre-symptoms of dielectric breakdown in MI-PPLP cable due to DC void discharge, partial discharge patterns were analyzed continuously until dielectric breakdown occurred. From the experimental results, DC void discharge patterns of the same shape were obtained regardless of the specimen temperature and the magnitude of applied voltage. In addition, it was confirmed that new insulation aging patterns were generated as electrical and thermal aging occurred due to the continuous DC void discharge. Therefore, it is demonstrated that identification and insulation diagnosis of void defects in HVDC MI-PPLP cable is possible through the obtained DC void discharge and insulation aging patterns.

Application of the Gaussian Mixture Model to Classify Stages of Electrical Tree Growth in Epoxy Resin.

In high-voltage (HV) insulation, electrical trees are an important degradation phenomenon strongly linked to partial discharge (PD) activity. Their initiation and development have attracted the attention of the research community and better understanding and characterization of the phenomenon are needed. They are very damaging and develop through the insulation material forming a discharge conduction path. Therefore, it is important to adequately measure and characterize tree growth before it can lead to complete failure of the system. In this paper, the Gaussian mixture model (GMM) has been applied to cluster and classify the different growth stages of electrical trees in epoxy resin insulation. First, tree growth experiments were conducted, and PD data captured from the initial to breakdown stage of the tree growth in epoxy resin insulation. Second, the GMM was applied to categorize the different electrical tree stages into clusters. The results show that PD dynamics vary with different stress voltages and tree growth stages. The electrical tree patterns with shorter breakdown times had identical clusters throughout the degradation stages. The breakdown time can be a key factor in determining the degradation levels of PD patterns emanating from trees in epoxy resin. This is important in order to determine the severity of electrical treeing degradation, and, therefore, to perform efficient asset management. The novelty of the work presented in this paper is that for the first time the GMM has been applied for electrical tree growth classification and the optimal values for the hyperparameters, i.e., the number of clusters and the appropriate covariance structure, have been determined for the different electrical tree clusters.

The effect of silica content to partial discharge characteristic of low-density polyethene and natural rubber blend as the electrical insulator

The dielectric properties of low-density polyethylene natural rubber (LDPE-NR) biopolymeric insulating materials can be improved by adding the silica nanoparticles in a certain percentage of weight (w%). In the present study, four types of bio-nano polymeric samples were prepared. To each sample, the nanosilica particles with wt% 1.5%, 3%, 4.5% and 6%. As one characteristic of dielectric, the partial discharge (PD) characteristics, each sample has been tested for 1 hour under AC high voltage field, and the pulses were counted for each sample and grouped into positive and negative pulses. The PD pattern was also plotted based on X-Y axes, namely Φ-q-n pattern. It was found that the number of positive and negative partial discharge (PD) pulses for each silica sample after 60 minutes of testing varied for all samples. It is also found that samples with a higher percentage of nanosilica had fewer PD pulses. The PD pattern in lower w% of silica was identified in the 90 degrees mostly in containing This indicates that w% of nanosilica particles can improve the PD resistance or the insulation quality of LDPE-NR insulation materials.

Investigation of Free-Moving Particles under AC Electric Field in Different Insulating Gas Mixtures

Free-moving conductive particles may affect the dielectric reliability of gas-insulated systems, e.g., gas-insulated lines (GIL) or gas-insulated switchgear (GIS). This paper reports on investigations on the behavior of free-moving particles under AC electric field, using different insulating gas mixtures. Multiple particles were positioned inside a GIL test setup with each particle placed within the same distance to a particle trap. To ensure reproducible and realistic particle geometry, cuboid particles were used. In this paper, a width correction factor was empirically determined for a well-known formula from the literature, which is used to calculate the lifting field strength for cuboid particles. While increasing the applied voltage, the movement of the particles was observed with a camera and further evaluated using ultra-high frequency (UHF) partial discharge (PD) measurements. The UHF phase-resolved partial discharge (PRPD) patterns of steel and aluminum particles in SF6, as well as in different alternative insulating gas mixtures, are compared and discussed. To determine the effectiveness of particle traps for different gas mixtures, further parameters are evaluated, such as the duration of each particle's movement and the number of particles reaching the inner conductor.

Online recognition method of partial discharge pattern for transformer bushings based on small sample ultra-micro-CNN network

Oil impregnated paper bushing is the key equipment connecting a transformer and a power grid. Insulation deterioration may cause partial discharge, which poses a great threat to the safe operation of power systems. In order to realize the online diagnosis for partial discharge and automatic identification of insulation defects of transformer bushings, an ultra-micro-convolutional neural network with only more than 3000 parameters is designed, which adaptively extracts partial discharge characteristics based on small samples, so as to judge the defect category and the reasons. The accuracy rate can reach 97.1%, the computational complexity is lower, the real-time performance is stronger, and it can be easily deployed on various embedded platforms.

Study of Oil/Pressboard Creeping Discharges under Divergent AC Voltage—Part 1: Fundamental Phenomena and Influencing Factors

This paper presents an experimental study of the fundamental phenomena of progressive creeping discharge as well as the influences of pressboard ageing and voltage amplitude under divergent AC voltages. A ramp-stress test is used to determine the partial discharge inception voltage (PDIV) and the flashover voltage of oil/pressboard interface versus pressboard ageing. Constant-stress tests are carried out to investigate the partial discharge (PD) behaviors, the pressboard surface temperature, and the oil-dissolved gases of progressive creeping discharges. It is found that the pressboard ageing affects only the interface PDIV. The progressive creeping discharges concentrate at the triple-junction and comprises probably oil corona discharges and surface discharges in pressboard dimples. If no pressboard damage is present, the progressive discharges exhibit in general a constant PD pattern but a diminishing intensity, and nevertheless can be sustained for the severely-aged pressboards under strong voltages. The underlying causes are correlated with the competing effects of pressboard ageing on interface moisture and the triple-junction field intensification. Gas analyses indicate the hydrogen is the principal faulty gas and the Duval's Triangle diagnoses the progressive creeping discharges as either low-energy or high-energy discharge fault. No pressboard overheating is observed during the discharges.

Study on the characteristics of elongation at break and tensile strength of photovoltaic insulating backsheets subjected to partial discharge degradation

As widely distributed vast reserves of emerging green energy, solar energy has gained widespread attention in recent years. Photovoltaic (PV) power generation is one of the most scientific ways to use solar energy. In PV systems, polyethylene terephthalate (PET) is the chief material of PV backsheets. It is well known that PV backsheets will encounter various challenges, among which partial discharge (PD) is one of these that cause deterioration in insulating and mechanical properties of PET. It has severely affected the service life of PV backsheets. In this paper, the PD characteristic of PET samples at different PD durations and temperatures was studied through phase-resolved PD patterns. In addition, the change mechanism of mechanical properties including elongation at break and tensile strength of PET after PD in 30, 60, and 120 min at 30 and 55 °C was also studied. It can be found that the insulating and mechanical properties of PET are reduced after PD degradation at different aging durations and temperatures. This study can provide a reference for the performance improvement of PV PET backsheets, which will also help to evaluate the service life of the PV module and PV power generation systems.

부분방전 데이터 처리 방법에 따른 CNN 기반 패턴 분류기의 비교 연구

This study is focused on solution to the problem that may occur due to noise signals when using the image obtained through the conventional partial discharge preprocessing methods as inputs to the CNN-based partial discharge pattern classifier. To solve such problem, a new data preprocessing method is proposed by considering a projection technique. In the data information obtained through the conventional partial discharge preprocessing methods, data information of useless noise signal leads to the problem of performance degradation when constructing highly qualified pattern classifier based on the data information of partial discharge signal. The proposed preprocessing method called as ‘Projection’ in this study is designed to solve this problem and improve a classification performance of CNN-based partial discharge pattern classifier. First of all, through GIS simulation, one-dimensional partial discharge data is obtained as five cases such as corona discharge, floating discharge, insulator discharge, free particle discharge, and steady state (noise) in an environment with a noise signal by using a UHF sensor. After that, by applying the two conventional partial discharge preprocessing methods such as PRPS(Phase Resolved Pulse Sequence), PRPD(Phase Resolved Partial Discharge) and the proposed partial discharge preprocessing method, one-dimensional partial discharge data is transformed into 3 data types such as image set of PRPS, image set of PRPD, and image set of Projection. Finally each image set is used as inputs to the designed CNN-based partial discharge pattern classifier. Through the comparative analysis of the feature maps of each layer in CNN as well as the performance accuracy of partial discharge pattern classification for each image set, the superiority of the proposed preprocessing method is demonstrated.

Quasi‐phase resolved partial discharge pattern analysis of surface discharges arising on the tar polluted porcelain bushing of an electrostatic precipitator unit under high voltage DC

AbstractThis paper investigates on understanding the DC electrical discharges arising on the surface of feed‐through type indoor bushing with tar depositions. Such information not only provides diagnostic status but also characterizes the multitude of species forming tar depositions which is currently unavailable in literatures. The current practice on analyzing surface discharges initiated under DC voltage employs primitive methods that accounts for the individual discharge events over a larger span rendering it less sensitive and time consuming. The usual practice is to count the discharge events initiated under DC voltage as PD pulses and count the events, repetition rates, time of occurrence, and so forth, to estimate the severity of the fault condition. On contrary, synchronizing these discharge pulses to the phase of the AC voltage, input to the rectifier unit, manifests a pattern, which is time independent and is sensitive in revealing diagnostic status of the bushing. These objectives are initially verified on a sample extracted from a tar polluted bushing. A half‐wave bridge rectifier that produces an uncontrolled positive and negative DC voltage is selected to initiate surface discharges. Pertinent pulses are then correlated to the phase of the AC voltage input to the rectifier and a quasi‐phase resolved partial discharge pattern (QPRPD) is developed. Using this QPRPD pattern, the significance of positive and negative DC voltages applied on the surface of the polluted sample is studied and the inception of discharges events and their correlation to the defect conditions are evaluated. Following this, the pertinent findings are validated on an actual bushing installed in an electrostatic precipitator unit applied for cleaning producer gas generated in a biomass gasification plant.

Development of Hankel Singular-Hypergraph Feature Extraction Technique for Acoustic Partial Discharge Pattern Classification

Different types of classifiers for acoustic partial discharge (PD) pattern classification have been widely discussed in the literature. The classifier performance mainly depends on the measurement conditions (location and type of the PD, acoustic sensor position and frequency response) as well as extracted features. Recent research posits that features extracted by singular value decomposition (SVD) can exhibit the natural characteristics and energy contained in the signal. Though the technique by itself is not novel, in this paper, SVD is employed for PD classification in a revised way starting from data arrangement in Hankel form, to embedding the hypergraph-based features and finally to extracting the required set of optimal features. The algorithm is tested for various measurement conditions that include the influences of various PD locations and oil temperatures. The robustness of the algorithm is also tested using noisy PD signals. Experimental results show the proposed feature extraction method supremacy.

Noise invariant partial discharge classification based on convolutional neural network

Partial discharge (PD) pattern recognition is essential since it can help to identify the nature of the insulation defect. Numerous machine learning models have been utilized for PD classification applications in the past. However, traditional machine learning models rely on manual feature extraction to obtain training data. They are usually trained using clean PD data measured in the laboratory but are expected to work on-site where some degree of interference or noise is expected. When tested using clean PD data, most machine learning models can easily achieve above 90% accuracy. However, when tested using PD data overlapped with noise, classification accuracy reduces significantly. In this work, the development of a convolutional neural network (CNN)-based PD classification system using transfer learning was proposed. In order to achieve a more practical performance evaluation, a modified 10-fold cross-validation procedure was used where the CNN-based PD classifier was trained using clean PD data but tested using PD data that has been overlapped by noise. The results showed that CNN-based PD classifier was able to achieve up to 16.90% higher classification accuracy under noise contamination compared to traditional machine learning with manual feature extraction. This shows that the proposed method was able to retain higher classification accuracy in the presence of noise.

A Convolutional Neural Network-Based Model for Multi-Source and Single-Source Partial Discharge Pattern Classification Using Only Single-Source Training Set

Classification of the sources of partial discharges has been a standard procedure to assess the status of insulation in high voltage systems. One of the challenges while classifying these sources is the decision on the distinct properties of each one, often requiring the skills of trained human experts. Machine learning offers a solution to this problem by allowing to train models based on extracted features. The performance of such algorithms heavily depends on the choice of features. This can be overcome by using deep learning where feature extraction is done automatically by the algorithm, and the input to such an algorithm is the raw input data. In this work, an enhanced convolutional neural network is proposed that is capable of classifying single sources as well as multiple sources of partial discharges without introducing multiple sources in the training phase. The training is done by using only single-source phase-resolved partial discharge (PRPD) patterns, while testing is performed on both single and multi-source PRPD patterns. The proposed model is compared with single-branch CNN architecture. The average percentage improvements of the proposed architecture for single-source PDs and multi-source PDs are 99.6% and 96.7% respectively, compared to 96.2% and 77.3% for that of the traditional single-branch CNN architecture.

A Convolutional Neural Network-based UHF Partial Discharge Atlas Classification System for GIS

In order to solve the difficult problem of partial discharge pattern recognition caused by large amount of partial discharge detection data and complex multi-source, a partial discharge pattern recognition algorithm based on VGG-16 convolution neural network is proposed. The parameters of VGG-16 network model are optimized in convolution layer, pool layer and connection layer by means of migration learning. The VGG-16 model is superior to LeNet-5 model and has higher recognition accuracy.

The importance of particle dispersion in electrical treeing and breakdown in nano-filled epoxy resin

The addition of nano-fillers has been widely proposed as a method to enhance the dielectric properties of high voltage polymeric insulation, though there are mixed reports in the literature. Here the potential of silica nano-particles to extend the time to failure specifically through resistance to electrical tree growth in epoxy resin is determined. The benefit of silane treating the nano-particles before compounding is clearly established with regard to slowing tree growth and subsequent time to failure. The growth of trees in needle-plane samples is measured in the laboratory with loadings of 1, 3 and 5 wt% nano-filler. In all cases the average times to failure are extended, but silane treatment of the nano-particles prior to compounding yields much superior results. The emergence of a pronounced inception time before tree growth is also noted for the higher-filled, silane-treated cases. The average time to failure of silane-treated 5 wt% filled material was 28 times that of the unfilled resin. The improvement in performance between the nanocomposites with untreated and treated fillers is attributed to fewer agglomerations and improved dispersion of the filler in the treated cases. Measurements of Partial Discharge (PD) indicated significant differences in PD patterns during the growth of trees in the treated and untreated cases. This distinction may provide a quality control method for monitoring materials. In particular, long periods in which PDs were not measured were observed in the silane-treated cases. Visual imaging of tree growth in the unfilled material allowed the changing nature of the tree from fine to tree to dark tree to be observed as it grew. Corresponding PD measurements suggest the dark tree is gradually becoming conductive, and that growth of maximum PD measured is dependent on the relative rates of the growth of the tree and its carbonization. X-ray computer tomography identified significant differences in average tree channel diameters (a reduction from 2.8 µm to 2.0 µm for 1 wt% and 3 wt% cases). This implies that in addition to tree length changes, evaporated tree volumes also change and may explain the change in partial discharge characteristics observed.

WITHDRAWN: Analysis of partial discharge patterns of natural ester oil and mineral oil used in power transformer

Abstract The liquid insulation used for power transformer over the years is petroleum based mineral oil because of it availability and good electrical, chemical insulation properties. However because of its poor biodegradability, low flash point and concern over environment nowadays mineral oil has been replaced with biodegradable oil. Natural ester oil is made from natural substances like soyabean, which are environment friendly, almost fire proof and have great electrical insulation properties. Partial discharge is considered to be one of the main reasons for degradation of insulations in power transformers. In this paper comparative study of experimental results of partial discharge patterns of mineral and natural ester oil are discussed. The experimental setup was designed as per IEC Standard. Power transformer insulation is modeled as pigtail specimen and tested for partial discharge behavior by applying stress more than inception voltage of mineral and natural ester oil. Phase resolved partial discharge pattern are used to analyze the results. It was found that Natural ester oil has better inception voltage compared to mineral oil and also number of partial discharge pulses and PD magnitudes are less compare to mineral oil.

Partial Discharges Characteristics Analysis in Power Transmission Lines Artificial Defects

The paper describes the research on partial discharges (PD) characteristics in various types of power transmission lines insulation defects. Thirteen types of insulation defects were considered. PD measurements were performed using the AC high voltage source and commercial PD registration device. Phase-resolved partial discharge (PRPD) patterns, average apparent charge, and PD intensity were obtained for each defect type. The research results showed that most overhead lines insulators defects had relatively high PD intensity with small average apparent charge. Several insulators’ defects showed a low average charge and intensity, but insulators’ bodies in these cases were cut and could be destructed by mechanical loads. Power cable samples showed relatively high intensity and average charge in case of the cable termination defect. However, the needle grounding electrode defect showed considerable average charge and intensity. The obtained results will be useful for the defect type recognition tools development and PRPD patterns base extension.