Partial Discharge Patterns Research Articles

Automatic PRPD Image Recognition of Multiple Simultaneous Partial Discharge Sources in On-Line Hydro-Generator Stator Bars

In this study, a methodology for automatic recognition of multiple simultaneous types of partial discharges (PDs) in hydro-generator stator windings was proposed. All the seven PD sources typical in rotating machines were considered, and up to three simultaneous sources could be identified. The functionality of identifying samples with no valid PDs was also incorporated using a new technique. The data set was composed of phase-resolved partial discharge (PRPD) patterns obtained from on-line measurements of hydro-generators. From an input PRPD, noise and interference were removed with an improved version of an image-based denoising algorithm previously proposed by the authors. Then, a novel image-based algorithm that separates partially superposed PD clouds was proposed, by decomposing the input pattern into two sub-PRPDs containing discharges of different natures. From the sub-PRPDs, one extracts features quantifying the PD distribution over amplitudes and the contour of PD clouds. Those features are fed as inputs to several artificial neural networks (ANNs), each of which solves a part of the classification problem and acts as a block of a larger system. Once trained, ANNs work collaboratively to identify an unknown sample. Good results were obtained, with overall accuracies ranging from 88% to 94.8% for all the considered PD sources.

Deep Learning Based Power Transformer Monitoring Using Partial Discharge Patterns

Measurement and recognition of Partial Discharge (PD) in power apparatus is considered a protuberant tool for condition monitoring and assessing the state of a dielectric system. During operating conditions, PD may occur either in the form of single and multiple patterns in nature. Currently, for PD pattern recognition, deep learning approaches are used. To evaluate spatial order less features from the large-scale patterns, a pre-trained network is used. The major drawback of traditional approaches is that they generate high dimensional data or requires additional steps like dictionary learning and dimensionality reduction. However, in real-time applications, interference incorporated in the measured single and multiple PD patterns may reduce the identification of the exact patterns and causes inaccurate diagnosis of equipment. The residual pooling layer is proposed in this work to overcome the drawbacks and provides fast learning. The projected algorithm consists of a residual encoding module and an aggregation module for spatial information preserving and order less feature generating. The advantages of the proposed work produce low dimensional data compared to other deep learning approaches. At last, the impact of random noise in the measured PD signal on recognition rate is investigated and addressed.

Comparison of AC and DC Partial Discharge by Monitoring of Vacuum Degree in a Distribution Class Vacuum Interrupter

The vacuum degree of Vacuum Interrupter ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> ) will deteriorate because of gases emitted by arc heat, and leakage through the joints. It is important to select an appropriate maintenance criteria for a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> by monitoring the vacuum degree. However, monitoring of a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> installed inside a solid insulation apparatus in a transmission system is impossible. Only studies on the insulation deterioration because of the decrease of the vacuum degree have been conducted, and important factors such as the maintenance criteria have not been investigated. Moreover, most studies have only been performed using AC voltage, not DC voltage. Therefore, in this paper, AC and DC partial discharge were compared by monitoring of the vacuum degree in a distribution class <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> to propose maintenance criteria according to voltage type. Through this, for efficient maintenance, maintenance criteria of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> for each voltage type are presented. The vacuum degree at which the dielectric strength rapidly decreases was confirmed according to voltage type. The coupling capacitor was installed directly on the floating shield of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> . Moreover, to measure the pattern and apparent charge of partial discharge according to the vacuum degree, phase resolved partial discharge (PRPD) for AC partial discharge and pulse sequence analysis (PSA) for DC partial discharge were adopted. It is found that maintenance should be performed when the vacuum degree is in the range of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$10^{-3} \boldsymbol {\sim } 10^{-2}$ </tex-math></inline-formula> torr regardless of the voltage type. Moreover, the usefulness of the technology is higher at AC voltage than at DC voltage.

Exploration on Automatic Management of GIS Using TL-CNN and IoT

The research aims to effectively monitor the state of complex industrial equipment in real time, diagnose the internal Partial Discharge (PD) pattern of Gas Insulated Metal Enclosed Switchgear (GIS), implement effective health management, and change the traditional model optimization route of increasing training time in exchange for performance improvement. This paper studies the complex equipment-oriented Health Management System (HMS) based on Internet of Things (IoT) technology and Transfer Learning. Firstly, the principles of Transfer Learning and Deep Learning (DL) technology are introduced. Secondly, the requirements of GIS internal status recognition and management are studied. Furthermore, a GIS-oriented HMS based on Transfer Learning-optimized Convolutional Neural Networks (CNN) is proposed, and the training dataset is constructed. Finally, the proposed model is tested. The results show that the complex equipment-oriented HMS based on IoT technology, CNN, and Transfer Learning can detect the internal status of GIS in real time. Compared with the traditional DL algorithm and expert system, the proposed model has a shorter training time of only 16min, faster convergence speed, high testing recognition rate, and over 96% recognition rate. Compared with other mainstream algorithms, it has higher identification, the storage parameter volume is 408, and the storage space is 12.8MB. Moreover, the proposed Transfer Learning-optimized CNN model can accurately detect the status of GIS, identify abnormal statuses, and help prolong the service life of GIS. The proposed complex equipment-oriented HMS contributes to the intelligent manufacturing industry and provides a new direction for applying emerging Computer Technology in the intelligent industry.

Investigation of Partial Discharges in AlN Substrates under Fast Transient Voltages

The objective of this article is to characterize partial discharge (PD) inception voltages and patterns at sinusoidal and square voltages in insulating ceramic substrates with generic manufacturing defects. PD tests were conducted on active metal brazed (AMB) aluminum nitride (AlN) substrates either with cavities in the polymer coating or sharp protrusions of the metal brazing. The substrates were tested in silicone liquid instead of silicone gel. Light sensitive imaging was used to localize the PD sites on the surface. PDs at square voltage with rise time of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1.6 ~\mu \text{s}$ </tex-math></inline-formula> were measured with a high-frequency current transformer and a photomultiplier. In both substrate types, the peak–peak value of the PD inception voltage (PDIV) was higher at sinusoidal voltage compared to peak-peak value of the square wave. This difference was more pronounced in uncoated substrates, where discharges incepted at sharp edges surrounded by silicone liquid. For both types of defects, comparable PDIVs were measured at bipolar and unipolar square waves. Results for unipolar pulses are discussed considering the role of charge memory effect.

Partial discharge pattern recognition in GIS based on EFPI sensor

Partial discharge pattern recognition in GIS based on EFPI sensor

РАЗРАБОТКА АЛГОРИТМОВ ОБРАБОТКИ СИГНАЛОВ ЧАСТИЧНЫХ РАЗРЯДОВ ДЛЯ РАСЧЕТА ИХ ХАРАКТЕРИСТИК

The paper describes the process of developing algorithms for processing information about partial discharges based on their registered electrical signals. The analysis is carried out and the most important characteristics of partial discharges are determined. They allow assessing the insulation condition. Taking into account the technical principles of operation of signal recording devices, algorithms have been developed for determining the apparent charge of each partial discharge, the average and maximum apparent charge, the power of partial discharges, and also phase-resolved partial discharge patterns plotting. A solution is given for registering the phase of the supply voltage at the frequency of registration of partial discharges. The results of the work can be used to develop software for the means of registration and analysis of the characteristics of partial discharges.

Optical Detection of Partial Discharges Under Fast Rising Square Voltages in Dielectric Liquids

Partial discharge (PD) measurement of an insulation system is widely used as a diagnostic tool for the quality control of high voltage apparatuses. In ac, the conventional method of PD detection is well established. However, under square voltages such as those generated across power electronic converters, the conventional method cannot discriminate between PDs and switching transients. In this context, this paper demonstrates PD detection under fast-rising square voltages by deploying a photomultiplier tube. For this, an experimental set-up generating unipolar square voltage (USV) waveforms is developed, generating a dV/dt in the range of 10-100 kV/&#x03BC;s. The device under test (DUT) comprises of needle-barrier plane configuration immersed in dielectric liquids. Barriers differing in permittivity and thicknesses are chosen to test with two types of dielectric liquids, to assess varying conditions. The results thus obtained show a lower partial discharge inception voltage (PDIV) in USV than in bipolar ac for the same DUT specimen. The phase-resolved (PR) PD patterns exhibit a higher PD occurrence during rising and falling front of the USV waveform. Experimental results demonstrate the relevance of the PMT as a PD detector for square voltage waveforms exhibiting high dV/dt in dielectric liquids.

Planar Ultrawideband Circularly Polarized Cosine Slot Archimedean Spiral Antenna for Partial Discharge Detection

Design and experimental verification of an ultrawideband planar circularly polarized ultrahigh frequency (UHF) sensor is presented for detection of partial discharge (PD) activities in high voltage (HV) equipment. The proposed sensor is cavity-backed cosine slot Archimedean spiral (CSAS) antenna with an electromagnetic (EM) transparent radome. The dual-arm slot spiral was meandered as cosine wave for size reduction and optimized for broadband operation with circular polarization (CP) over 0.5-5 GHz. Unidirectional radiation pattern with CP was achieved over 0.5-5 GHz using an optimized cavity loaded with EM absorber. The fabricated UHF CSAS sensor has the widest operating bandwidth of 0.5-5 GHz with circular polarization and the smallest aperture of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$0.26\lambda {}_{L}$ </tex-math></inline-formula> when compared to the existing UHF PD sensors, where <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\lambda {}_{L}$ </tex-math></inline-formula> is the wavelength at the lowest operating frequency. Time domain incipient discharge signals and phase resolved PD patterns gathered for three commonly occurring defects in test cells mimicking gas insulated switchgear (GIS) indicate better sensitivity and detection capability for the fabricated CSAS UHF sensor than the reference PD sensor. PD signal measurements and comparison with other UHF PD sensors indicate that the proposed UHF CSAS sensor could be used for PD detection in HV equipment.

Discharge Characteristics and Mechanism of Bubble Flow in FC-72 under Unipolar Repetitive Pulse Voltage

This article investigates the discharge characteristic of the bubble flow in FC-72 under unipolar repetitive pulsed voltage, which supports the insulation design of two-phase cooling in an HV power electronic device. The effect of thermal bubble flow on the insulation of FC-72 refrigerant was evaluated. According to the partial discharge (PD) characteristics, it is found that there are two stages in the discharge characteristics of the bubble flow as voltage rises. Combining the PD inception voltages of FC-72 vapor under unipolar pulse voltage and dc voltage, it reveals the discharge thresholds of the vapor inside the bubble under the transient electric field and the steady electric filed, respectively, which proves the mechanism of the staged PD evolution. Based on the motion calculation of the bubble, the PD probability of the bubble flow within a voltage period is analyzed. It simulates the phase-resolved PD pattern, which reveals the PD mechanism of the bubble flow. The research provides a reference for the insulation design for the two-phase cooling technology in power electronics.

A novel image‐orientation feature extraction method for partial discharges

Partial discharge identification is of great significance in monitoring power equipment. Although substantial work has been done in partial discharge pattern recognition, most studies have been performed in noise-free environments. To improve the accuracy of partial discharge pattern recognition, this paper introduces a partial discharge image feature extraction method based on upright speeded-up robust features. Each group of phase-resolved pulse sequence data is converted into grayscale images to construct a sample library of partial discharge defects. Upright speeded-up robust features is used to extract features in grayscale images. The improved support vector machine is used to achieve the best classification of partial discharge sources. The sensitivity of local features to different phase resolutions and different image resolutions of partial discharge phase-resolved pulse sequence patterns was investigated, and the optimal phase resolution and image resolution were used to construct phase-resolved pulse sequence patterns. In addition, the recognition effect of this feature extraction method under different noise conditions was also analysed. The results prove that upright speeded-up robust features can effectively deal with noisy data. The results of this research can provide references for the storage setting and recognition of three-dimensional images of on-site partial discharge data.

CALIBRATION OF EXPERIMENTAL INSTALLATION FOR MEASURING PARTIAL DISCHARGES IN LOW CAPACITANCE INSULATION SAMPLES

Purpose. The method for calibrating an experimental installation for studying the patterns of partial discharges aris-ing in samples of paper insulation having low capacitance was developed. Methodology. In order to verify the proposed method, a physical experiment was carried out in a high-voltage laboratory to measure partial discharge parameters using a digital oscilloscope. Free software for circuit simulation was used to create high-pass filter schematics. Results. The task of calibrating the systems for measuring partial discharges of low capacitance insulation samples (for example, ranging from 1 pF to 5 pF) is characterized by high complexity, because the calibration capacitor must have a capacitance an order of magnitude smaller than the capacitance of the test sample (from 0.1 pF to 0.5 pF, accordingly), which sometimes cannot be achieved practically. Moreover, in such case the stray capacitance will obviously be of the same order as the capacitance of calibration capacitor, or even greater. In such cases traditional calibration circuits where calibration generator is connected in parallel with the test object cannot be applied. Originality. Alternate calibration circuit was proposed, implemented and tested in the work, where the calibration generator is connected in series with test object. Practical value. Studies of the proposed calibration method have shown that it can be used quite effectively in cases when test object have low capacitance. The results of recording the calibration pulses showed that they can be reliably recorded and measured with a minimum error (oscilloscope error), and errors associated with parasitic parameters of the recording circuit are automatically taken into account when calibrating by this method and do not require additional adjustments. Conclusions. There is a possibility to increase the sensitivity of the measuring circuit to 0.05 pC/V, primarily by increasing the gain of the oscilloscope at least 100 times without repeating the calibration procedure. This will make it possible to study insulation samples of fairly high quality in which the level of partial discharges is much lower than in those samples used in this work. The experimental installation can be used to analyze different ways of modeling partial discharges on a personal computer and comparing their results with a real experiment. References 15, figures 12.

Inception and Propagation of Electrical Trees in the Presence of Space Charge in HVAC Extruded Cables

This paper presents the space charge impact on the inception and propagation of electrical trees in cross-linked polyethylene (XLPE) insulation via simulations and experimentation. A 3D finite element analysis (FEA)-based modeling is proposed to simulate electrical trees via a needle embedded on the XLPE insulation. The proposed FEA model demonstrates the influence of the space charge magnitude and polarity on the initiation of partial discharges (PD). Then the critical parameters at the tip of an 8 µm. tree are to be examined. Experimentation is necessary to verify the simulation results acquired and study the mechanism of electrical trees at the inception and propagation stage. Therefore, in the experiment, a needle with a curvature radius of 5 µm. is embedded in the XLPE as a method of simulating defects, such as protrusions, voids, or cracks found in cable insulation. The power source supplies several high AC voltages on the 20 kV (rated voltage) cable under test for observing the inception and propagation behavior of the electrical trees. Phase-resolved partial discharge (PRPD) patterns are extracted at the inception and propagation stages of the electrical tree to determine the severity of the PDs occurring. A relationship is obtained for the PD magnitude with respect to the applied AC voltage while considering the positive and negative polarities of the PDs. This relationship then allows for the acquisition of the maximum electric field-tree length characteristics. The goal of the simulation of electrical trees is to provide the reader with a better understanding about the effect of space charge magnitude and polarity to comprehend the behavior of the treeing inception and propagation mechanisms. The injection and extraction of space charges is a complex phenomenon that requires further visualization through simulations. Therefore, the results of both the simulation and experiment are compared with the aim of establishing a clear relationship between space charge distribution and tree initiation.

Detection of Electrical Tree Formation in XLPE Insulation through Applying Disturbed DC Waveforms

Initiation and growth of electrical trees can be linked to the development of partial discharge (PD) activity within an insulation system. We present in this paper changes in PD characteristics that appear during the electrical tree growth in a crosslinked polyethylene (XLPE) based insulation at different stages of its degradation. A DC voltage with superimposed high frequency component is applied to the material specimens for the detection and characterization of PD traces observed during the degradation. The method is flexible and offers a possibility to vary DC voltage level independently of the superimposed disturbance, its frequency and rise time, including rapid voltage steps. It also enables to create a phase-locked phase-resolved PD pattern (PRPD) facilitating various analyses of the characteristics. It is postulated that measurements of the changes in PD characteristics found under such excitations can be used for performing diagnostics and prognostics of material condition. This approach also implies new possibilities to evaluate different material concepts.

A novel 1DCNN and domain adversarial transfer strategy for small sample GIS partial discharge pattern recognition

Recently, convolutional neural networks (CNNs) have made certain achievements in gas-insulated switchgear (GIS) partial discharge (PD) pattern recognition. However, these methods rely on the availability of massive PD samples and how to apply the CNN constructed in the laboratory to the field GIS PD pattern recognition has become an urgent problem. To solve these problems, we propose a small sample GIS PD pattern recognition using one-dimensional CNN (1DCNN) and domain adversarial transfer learning (DATL). First, a novel 1DCNN is constructed to achieve high-accuracy classification using unbalanced samples, where the problem of traditional two-dimensional CNN’s dependence on sample size is solved. Second, DATL is used to realize on-site GIS PD pattern recognition using small samples containing some unlabeled samples. In the domain adversarial training, two domain classifiers are introduced to align the domain of the decision boundary, which achieves a suitable features migration and accurate classification of target domains. Through the construction of multiple experiments, we verified that the proposed method achieves 98.67% and >92% recognition accuracy in the source domain and target domain, respectively. Compared with the existing methods, the proposed method can realize satisfactory pattern recognition, which can provide strong support for the subsequent pattern recognition of GIS PD.

GAN and CNN for imbalanced partial discharge pattern recognition in GIS

The convolutional neural network (CNN) achieves excellent performance in pattern recognition owing to its powerful automatic feature extraction capability and outstanding classification performance. However, the actual samples obtained are unbalanced, and accurate diagnoses are difficult for the existing methods. A classification method for partial discharge (PD) pattern recognition in gas-insulated switchgear (GIS) that uses a generative adversarial network (GAN) and CNN on unbalanced samples is proposed. First, a novel Wasserstein dual discriminator GAN is used to generate data to equalise the unbalanced samples. Second, a decomposed hierarchical search space is used to automatically construct an optimal diagnostic CNN. Finally, PD pattern recognition classification in GIS of the unbalanced samples is realised by the GAN and CNN. The experimental results show that the GAN and CNN methods proposed in this study have a pattern recognition accuracy of 99.15% on unbalanced samples, which is significantly higher than that obtained by other methods. Therefore, the method proposed in this study is more suitable for industrial applications.

Assessment of dielectric strength and partial discharges patterns in nanocomposites insulation of single-core power cables

Nanoparticles succeeded to enhance the dielectric properties of industrial insulation but the presence of voids inside the power cable insulation still leads to formation high electrical stress inside power cable insulation material and collapse. In this paper, the dielectric strength of new design nanocomposites has been deduced as experimental work done to clarify the benefit of filling nanoparticles with different patterns inside dielectrics. Also, it has been studied the effect of electrical stress distribution in presence of air, water and copper impurities with different shapes (cylinder, sphere and ellipse) inside insulation of single core. In simulation model, it has been used finite element method (FEM) for estimating the electrostatic field distribution in power cable insulation. It has been applied new strategies of nanotechnology techniques for designing innovative polyvinyl chloride insulation materials by using nanocomposites and multi-nanocomposites. Finally, this research succeeded to remedy different partial discharges (PD) patterns according to using certain types and concentrations of nanoparticles.

Effect of O2 on AC Partial Discharge and Decomposition Behavior of C4F7N/CO2/O2 Gas Mixture

This paper focus on the influence of oxygen on the partial discharge (PD) and decomposition characteristics of C4F7N/CO2/O2 gas mixture as eco-friendly gas insulating medium. The PD characteristic parameters and phase resolved partial discharge (PRPD) patterns of C4F7N/CO2/O2 gas mixture are, respectively, extracted and constructed. The influences of oxygen content and applied voltage on the PD and decomposition characteristics of C4F7N/CO2/O2 gas mixture are investigated based on gas chromatography-mass spectrometry (GC-MS). We find that the addition of 2-4% O2 to C4F7N/CO2 gas mixture can significantly inhibit the negative half-cycle PDs and the generation of most by-products (CF4, C2F6, C3F6, CO, C2F6O3, CF3CN, C2F5CN, C2N2). With the oxygen content higher than 6%, the accumulation discharge magnitude and discharge repetition rates of C4F7N/CO2/O2 gas mixture increase sharply, accompanying the intensification of decomposition and deterioration of insulation strength. There is a positive correlation between the mean discharge magnitude and gas mixture decomposition intensity. To avoid the serious impact of PD fault on gas insulated equipment (GIE), the optimal oxygen addition in C4F7N/CO2/O2 gas mixture should not exceed 6%.

Partial Discharge Pattern Recognition of Transformers Based on MobileNets Convolutional Neural Network

The power system on the offshore platform is of great importance since it is the power source for oil and gas exploitation, procession and transportation. Transformers constitute key equipment in the power system, and partial discharge (PD) is its most common fault that should be monitored and identified ın a timely and accurate manner. However, the existing PD classifiers cannot meet the demand for real-time online monitoring due to their disadvantages of high memory consumption and poor timeliness. Therefore, a new MobileNets convolutional neural network (MCNN) model is proposed to identify the PD pattern of transformers based on the phase resolved partial discharge (PRPD) spectrum. The model has the advantages of low computational complexity, fast reasoning speed and excellent classification performance. Firstly, we make four typical defect models of PD and conduct a test in a laboratory to collect the PRPD spectra as the data sample. In order to further improve the feature expression ability and recognition accuracy of the model, the lightweight attention mechanism Squeeze-and-Excitation (SE) module and the nonlinear function hard-swish (h-swish) are added after constructing the MCNN model to eliminate the potential accuracy loss in PD pattern recognition. The MCNN model is trained and tested with the pre-processed PRPD spectrum, and a variety of methods are used to visualize the model to verify the effectiveness of the model. Finally, the performance of MCNN is compared with many existing PD pattern recognition models based on convolutional neural network (CNN), the results show that the proposed MCNN can further reduce the number of parameters of the model and improve the calculation speed to achieve the best performance on the premise of good recognition accuracy.

Anomaly Detection, Trend Evolution, and Feature Extraction in Partial Discharge Patterns

In the resilient and reliable electrical power system, the condition of high voltage insulation plays a crucial role. In the field of high voltage insulation integrity, the partial discharge (PD) inception and development trends are essential for assessment criteria in diagnostics systems. The observed trend to employ more and more sophisticated algorithms with machine learning features and artificial intelligence (AI) elements is observed everywhere. The classification and identification of features in PD images is perceived as a critical requirement for an effective high voltage insulation diagnosis. In this context, techniques allowing for anomaly detection, trends observation, and feature extraction in partial discharge patterns are important. In this paper, the application of few algorithms belonging to image processing, machine learning and optical flow is presented. The feature extraction refers to image segmentation and detection of coherent forms in the images. The anomaly detection algorithms can trigger early detection of the trend changes or the appearance of a new discharge form, and hence are suitable for PD monitoring applications. Anomaly detection can also handle transients and disturbances that appear in the PD image as an indication of an abnormal state. The future monitoring systems should be equipped with trend evolution algorithms. In this context, two examples of insulation aging and application of PD-based monitoring are shown. The first one refers to deep convolutional neural networks used for classification of deterioration stages in high voltage insulation. The latter one demonstrates application of optical flow approach for motion detection in partial discharge images. The motivation for the research was the strive to machine-controlled pattern analysis, leading towards intelligent PD-based diagnostics.