Partial Discharge Source Location Research Articles

A novel meta‐learning network for partial discharge source localization in gas‐insulated switchgear via digital twin

AbstractDue to the requirement for highly precise synchronous sampling and the substantial reliance on time difference calculations, the current partial discharge (PD) localization based on the time difference of arrival is only applicable in certain situations. As digital twin technology has advanced, it is possible to employ virtual models to support gas‐insulated switchgear (GIS) PD localization. To do this, we propose a meta‐learning (ML) network with the aid of digital twin for actual GIS PD localization. Firstly, a GIS digital twin model was established to acquire an auxiliary simulated sample library. Then, a temporal convolutional network is established to extract the discriminable features, effectively obtain the time dependence between features, and improve the accuracy of localization. Next, ML is adopted to quickly learn meta‐knowledge that can be applied across tasks, and the model's sensitivity to task changes is improved. Finally, the model is fine‐tuned through a limited number of samples from the target task, and high precise PD localization is achieved. The experimental results demonstrate that the ML has an average localization error of only 9.25 cm and a probability density rose to 93% within 20 cm, which is clearly superior to previous methods.

Partial Discharge Monitoring of Power Transformers by Calibrated UHF Measurements

Partial discharge (PD) measurement is an established technique to detect local defects in the oil/paper insulation. They can be measured by electrical PD measurements according to IEC 60270 or by electromagnetic measurements in the ultra-high frequency range (UHF: 300 MHz – 3 GHz). The electromagnetic emissions of PD are recorded using an UHF sensor which is inserted into the transformer tank. The importance of PD measurements is reflected in the standard for charge-based electrical measurements (IEC 60270). Because of the standard, apparent charge, QIEC, became an essential factor for determining insulation quality in transformers. To be an accepted factor for both, quality testing in factory acceptance tests (FAT) or site acceptance tests (SAT) and continuous PD monitoring, the UHF technology must be as reliable and reproducible as the electrical method. To achieve this, a standardized calibration procedure is needed: It makes UHF measurement results from different systems comparable. Because this calibration procedure was lacking in the past, the UHF technology does not provide comparability between measurement systems and different sensors, yet. However, it shows advantages for on-site acceptance tests like SAT, continuous monitoring, and diagnostic purposes. E.g., regarding signal to noise ratios, the possibility for localization of PD sources and also for practical reasons like preparation times and accessibility on-site. This contribution presents a PD monitoring system which can be calibrated according to the calibration process described recently in Cigré TB 861 and two different types of UHF PD sensors for power transformers. One sensor type is for new transformers using dielectric windows and the other is for retrofitting transformers using their DN50/80 drain valves. A recommendation or strategy on where to place window-type UHF sensors at the tank of new power transformers is also provided in accordance with TB 861. Additionally, this contribution provides a description of the calibration process for UHF PD measurement systems consisting of two steps: 1) Calibration of the measurement instrument analogous to the calibration of the electrical PD measurement circuit using a defined reference signal. 2) Calibration of a UHF sensor using its characteristic antenna factor (AF). Furthermore, it shows three case studies from different power transformers equipped with UHF sensors and UHF PD monitoring systems.

Application of Pathfinding Algorithms in Partial Discharge Localization in Power Transformers.

The introduction of artificial intelligence (AI) to ultra-high-frequency (UHF) partial discharge (PD) monitoring systems in power transformers for the localization of PD sources can help create a robust and reliable system with high usability and precision. However, training the AI with experimental data or data from electromagnetic simulation is costly and time-consuming. Furthermore, electromagnetic simulations often calculate more data than needed, whereas, for localization, the signal time-of-flight information is the most important. A tailored pathfinding algorithm can bypass the time-consuming and computationally expensive process of simulating or collecting data from experiments and be used to create the necessary training data for an AI-based monitoring system of partial discharges in power transformers. In this contribution, Dijkstra's algorithm is used with additional line-of-sight propagation algorithms to determine the paths of the electromagnetic waves generated by PD sources in a three-dimensional (3D) computer-aided design (CAD) model of a 300 MVA power transformer. The time-of-flight information is compared with results from experiments and electromagnetic simulations, and it is found that the algorithm maintains accuracy similar to that of the electromagnetic simulation software, with some under/overestimations in specific scenarios, while being much faster at calculations.

Gas-insulated switchgear partial discharge acoustic–electric joint localisation method based on the Salp Swarm Algorithm and least squares estimation

Partial discharge (PD) is an early warning of gas-insulated switchgear (GIS) insulation faults. Consequently, accurate localisation of PD is crucial for the safe operation of GIS. The time difference of arrival method based on acoustic emission technology is more favorable for the engineering application of PD source localisation owing to its noninvasive characteristics and the small amount of required data. However, because PD ultrasonic signal propagation in GIS will introduce a variety of errors, current localisation algorithms suffer from problems of low accuracy and poor stability. Therefore, this study simulate and analyse the ultrasonic propagation characteristics as well as the sources of errors in GIS, based on which a joint acoustic–electric localisation method based on the Salp Swarm Algorithm is proposed, and an error-correction method based on least squares estimation is introduced. The localisation results show that the proposed method can effectively reduce the localisation error caused by multiple sources of error, and the mean localisation error can be reduced to within 20 mm in different structural gas chambers of GIS, which is more accurate and more stable compared with other methods, and it has good value for engineering application.

Fault localization on power cables using time delay estimation of partial discharge signals

Precise localization of partial discharge (PD) sources on power cables is vital to prevent power line failures that can lead to significant economic losses for electrical suppliers. This study proposes four methods to estimate the time delay of PD signals under electromagnetic interference, including white Gaussian noise (WGN) and discrete sinusoidal interference (DSI), using denoised PD signals with signal-to-noise ratios ranging from 10.6 to -7.02 dB. The maximum peak detection (MPD) and cross-correlation (CC) approaches, as well as two new techniques, interpolation cross-correlation (ICC) and envelope cross-correlation (ECC), are evaluated for their effectiveness in PD source localization. The researchers employ the time difference of arrival (TDoA) algorithm to compute PD location using the double-end PD location algorithm, where the PD location precision serves as an indicator of the accuracy of the time delay estimation methods. The study concludes that CC and ICC are the most suitable methods for estimating the time delay of PD signals in the PD location algorithm, as they exhibit the lowest error rates. These results suggest that CC and ICC can be used effectively for precise PD source localization under electromagnetic interference on power cables.

Partial Discharge Localization Techniques: A Review of Recent Progress

Monitoring the partial discharge (PD) activity of power equipment insulation is crucial to ensure uninterrupted power system operation. PD occurrence is highly correlated to weakened insulation strength. If PD occurrences are left unchecked, unexpected insulation breakdowns may occur. The comprehensive PD diagnostic process includes the detection, localization, and classification of PD. Accurate PD source localization is necessary to locate the weakened insulation segment. As a result, rapid and precise PD localization has become the primary focus of PD diagnosis for power equipment insulation. This paper presents a review of different approaches to PD localization, including conventional, machine learning (ML), and deep learning (DL) as a subset of ML approaches. The review focuses on the ML and DL approaches developed in the past five years, which have shown promising results over conventional approaches. Additionally, PD detection using conventional, unconventional, and a PCB antenna designed based on UHF techniques is presented and discussed. Important benchmarks, such as the sensors used, algorithms employed, algorithms compared, and performances, are summarized in detail. Finally, the suitability of different localization techniques for different power equipment applications is discussed based on their strengths and limitations.

A Comparison of Double-End Partial Discharge Localization Algorithms in Power Cables

The double-end partial discharge (PD) measurement method is the most common method for measuring and localizing PD sources in power cables. The sensitivity of the PD sensor, the processing speed of the data acquisition unit, and the method of the PD localization algorithm are the three main keys to ensuring the accuracy of the PD source localization on power cables. A new multi-end PD localization algorithm known as segmented correlation trimmed mean (SCTM) has recently demonstrated excellent accuracy in the localization of PD sources on power cables. The algorithm, however, is only applicable to multi-end PD measurement methods. In this paper, the mathematical equation of the SCTM algorithm is customized to match the double-end PD measurement method. A MATLAB simulation was conducted to assess the performance of the SCTM algorithm in the double-end PD measurement method. The maximum peak detection (MPD) algorithm, segmented correlation (SC), and SCTM algorithm were compared as PD localization algorithms. The SC algorithms have shown that identifying the correlation bond between two cues instead of the peak of the PD signal in the MPD algorithm significantly increases the PD localization accuracy. The results show that the SCTM algorithm outperforms the MPD and SC algorithms in terms of accuracy.

UHF Sensors Based Localization of Partial Discharge Sources in Air-Insulated Electrical Substation Using TDOA and Relative RSS Information

Time-difference of arrival (TDOA) based localization of partial discharge (PD) sources using Ultra High Frequency (UHF) technique often provides uncertain accuracy because of unavoidable errors in TDOA estimation between PD pulses. This paper presents a novel localization approach utilizing TDOA and relative received signal strength (RRSS) information together for improving the localization accuracy of PD sources within air-insulated substations (AIS). A complete PD localization system governed by this approach has been developed using multiple UHF PD sensors. The effectiveness of the proposed PD source localization approach has been validated through laboratory and field-based experiments in real-life AIS.

Collaborative Domain Adaptation Network for Partial Discharge Source Localization in Gas-Insulated Switchgear

Abnormal discharge in gas-insulated switchgear (GIS) is a key cause of insulation failure, and it is also an external manifestation of insulation defects. Sensitive discharge source localization is an important goal of GIS partial discharge (PD) monitoring. However, most existing GIS PD source localization methods rely on time-delay estimation, which not only requires a high-precision synchronous sampling device but also suffers from serious interference. To this end, a collaborative domain adaptation network (CDAN) is proposed for GIS PD source localization. First, a synchronous squeezing wavelet transform is introduced to filter out the noise, eliminating the influence of noise on GIS PD source localization. Then, one-dimensional attention convolutional neural network is constructed to ensure that discriminative temporal fine-grained information is extracted. Next, a CDAN is proposed to transfer the localization knowledge learned from the simulation to actual GIS and achieve high-precision localization through domain matching and alignment. The results demonstrate that the error of the CDAN proposed is only 19.87 cm, which is considerably better than that from other methods. This work provides a reference solution for GIS PD source localization.

Localization of Partial Discharge Sources in Spatial Distributed High-Voltage Windings of Rotating Machines

This article presents the key results of comprehensive investigations and the development of an algorithm which targets the partial discharge (PD) localization in spatial distributed rotating machine windings. Limitations and accuracy of the developed tool are discussed using the case study of a 4 kV rated voltage motor winding. Artificial and real PD events, which are provoked at various positions along the insulation system, can be localized reliably within certain limits. In addition, the influence of different measurement configurations is investigated and assessed regarding their application possibilities for fault localization.

Convenient Online Approach to Multisource Partial Discharge Localization in Transformer

Partial discharge (PD) online monitoring is an essential part of transformer insulation condition based maintenance and prognostics and health management in industrial power system. In order to realize online localization of multiple PD sources conveniently, an approach that uses single-channel mixed acoustic emission (AE) signal is proposed in this article. First, the combined method based on the singular spectrum analysis and the independent component analysis is used to calculate the number of PD sources and realize the separation of mixed AE signals. Next, a difference selection algorithm is proposed to solve the problem that the inherent disorder of signal blind source separation makes the time difference of arrival technology difficult to localize the multiple PD sources directly. Then, the simulation signals and measured signals have verified that the signal separation effect of the proposed single-channel-based method is no less than that of the multichannel-based method. Finally, the proposed approach is applied to a transformer. The disordered separated signals were successfully grouped to realize multisource PD localization.

Experimental Analysis of Ultra-High-Frequency Signal Propagation Paths in Power Transformers

Ultra-high-frequency (UHF) partial discharge (PD) monitoring is gaining popularity because of its advantages over electrical methods for onsite/online applications. One such advantage is the possibility of three-dimensional PD source localization. However, it is necessary to understand the signal propagation and attenuation characteristics in transformers to improve localization. Since transformers are available in a wide range of ratings and geometric sizes, it is necessary to ascertain the similarities and differences in UHF signal characteristics across the different designs. Therefore, in this contribution, the signal attenuation and propagation characteristics of two 300 MVA transformers are analyzed and compared based on experiments. The two transformers have the same rating but different internal structures. It should be noted that the oil is drained out of the transformers for these tests. Additionally, a simulation model of one of the transformers is built and validated based on the experimental results. Subsequently, a simulation model is used to analyze the electromagnetic wave propagation inside the tank. Analysis of the experimental data shows that the distance-dependent signal attenuation characteristics are similar in the case of both transformers and can be well represented by hyperbolic equations, thus indicating that transformers with the same rating have similar attenuation characteristics even if they have different internal structures.

The Investigation of Detect Position of Partial Discharge in Cast-Resin Transformer Using High-Frequency Current Transformer Sensor and Acoustic Emission Sensor

The lifetime of a cast-resin transformer mainly depends on the condition of insulation material. Partial discharge (PD) is an important reason for insulation deterioration in cast-resin transformers. Identifying the position of PD is very necessary for damage assessment while the transformer is still operating, and the transformer is covered by housing. This paper proposes the investigation of a cast-resin transformer using an AE sensor and HFCT sensor to specify the precise source of PD. In this study, four AE sensors were used to find PD sources, and the high-frequency current transducer (HFCT) technique was used to identify the PD source and the criteria level. The experiment, in the first two parts, identified the possibility of PD, which includes the position of PD. The final part of the experiment verified the position of the PD source of a cast-resin transformer and confirmed the inspection results. AE and HFCT sensors can be used to detect the location of PD sources, confirming the position of the PD source by sensor detection. In addition, the evident partial discharge picture on the insulator surface of high voltage side. The process successfully and accurately identifies and locates the PD source.

Research on improved time difference estimation algorithm based on fourth-order cumulant in UHF PD location

Ultra-high frequency (UHF) detection technology is a fast-developing method for partial discharge (PD) diagnosis of electrical equipment in recent years. Extracting accurate time difference is the priority for PD location in electrical equipment using UHF technology. An improved time difference estimation algorithm based on fourth-order cumulants was first proposed in this article. It mainly included the following two steps, that is using the “correlation-shift-superposition” transformation method to process multiple samples collected from a PD source at first, and then utilizing the statistical characteristics of the fourth-order cumulant to calculate the time difference. In addition, the accuracy and stability of the improved algorithm and the other three common methods were compared. The calculation results show that the improved time difference estimation algorithm based on the fourth-order cumulant can obtain a more accurate and stable time difference than the other three methods. Improved time difference estimation algorithm based on fourth-order cumulant can approximately meet the requirements for reliable location of PD sources.

Location Technique based on Multiple Partial Discharge Signal in 11kV Underground Power Cable using EMTP-ATP Software

Power cables are very critical in electrical power systems as power cables failure can interrupt the electrical flow due to unexpected power failure. There are a few sorts of partial discharge (PD) estimations gadgets in the market. For instance, PD can be distinguished by utilizing Rogowski coil (RC) sensors in the disconnected procedure. The current issue PD signal does not usually occur as a single source. Thus, the analysis of multiple PD sources is required to ensure that the cable insulation is in a healthy condition. PD location technique based on multiple signals in 11kV underground power cable was conducted in this research to estimate the accurate location of the PD signal. Modelling of single power cable in a distance of 10km with the RC sensor is installed at several distances to capture the PD signal that travels along the power cable. By selecting the distance between six RC sensors and synchronous multiple PD signal, the design of the power system has been constructed by using EMTP-ATP software. Multi-point technique based on time difference of arrival (TDOA) was performed in the single line power cable to obtain the PD location. The measurement using multi-point of RC sensor technique is preferred based on the conditions due to the value of velocity elimination. Based on the results, the accurate location of PD Source 1 is detected 501 m along RC sensor A1 to RC sensor A3. In contrast, PD source 2 has been detected 2800.15 m along RC sensor A4 to RC sensor A6 with the percentage error of 0.2% and 0.0053%, respectively. The findings show that the location of multiple PD signal that occurred along the line cable can be detected accurately by using the multi-point technique and TDOA. Hence, the performance of the power system has been improved.

Defect Localization Inside Simulated MV Switchgear Based on Cumulative Energy Curve Using Transient Earth Voltage Sensors

The partial discharge (PD) leads to catastrophic failure of the medium voltage (MV) switchgear insulation. Determination of the PD source (defect) location in high voltage (HV) equipment is very important in the maintenance procedures and in isolating the root cause of PD generation. In this paper, the transient earth voltage (TEV) detection method was used to acquire defect-initiated PD signals in a simulated MV switchgear model. An array of four TEV sensors were placed on the surface walls outside an MV switchgear tank to acquire the PD signals generated from the known location(s)/coordinates of sharp needle type defect inside the tank. The time difference of arrival (TDOA) between signals that are captured by the TEV sensors array was critically analyzed. Estimating the TDOA between PD signals generated by PD source at a known location with high accuracy is of great importance for accurate defect localization. The cumulative energy method (CEM) is used to estimate the onset time point of each TEV signal. The estimated TDOA by the cumulative energy method is compared with actual and expected TDOA based on known coordinates of PD source and TEV sensors. Experimental data are used as a basis for determining the TEV method accuracy for PD source localization. Experimental results show the average error of time difference is about 1.34 ns, which is equivalent to the propagation distance of 0.4 m.

Localization and intensity calibration of partial discharge based on attenuation effect of ultrasonic sound pressure

In the insulation system of power equipment, the partial discharge (PD) of short period does not cause the insulation to produce the penetrating breakdown, however the long-term PD of is one of the important causes of local deterioration, and even breakdown in dielectric. Therefore, it is very important to study the location of PD source and the calibration of discharge intensity. To achieve this, in this paper we take the needle-plate discharge model for example and go through the following steps respectively. Firstly, combined with the positive correlation between the ultrasonic signal and the apparent discharge magnitude in the process of PD, the ultrasonic method to detect partial discharge can be implemented. Then, based on the principle of time difference of arrival method (TDOAM), the accuracy of location is analyzed by using quantum genetic algorithm (QGA), genetic algorithm (GA), simulated annealing algorithm (SAA), particle swarm optimization (PSO) and generalized cross correlation method (GCC), respectively. And thus, starting from the study of the attenuation effect of sound pressure caused by the propagation loss, reflection and refraction of acoustic wave, the calibration model of PD intensity is established for the first time after determining the location of PD source with high precision. Some important findings are extracted from simulations and experimental results. First, the localization algorithm of PD source with high precision is observed. The localization of PD source by means of QGA is the most accurate, with maximum deviation of (0.27 ± 0.13) cm. Comparing with GA, SAA, PSO and GCC, the accuracy of location is improved by 33.57%, 41.51%, 32.11% and 87.26%, respectively. Second, due to the attenuation effect of sound pressure, when the measured voltage amplitude of ultrasonic signal is the same, the apparent discharge magnitude of PD source gradually increases with the test distance increasing. When the test distance is 37.80 cm, the apparent discharge magnitude of PD source is 633.83 pC, which increases by 28.51% compared with 7.00 cm. Moreover, simulation results and measurement results are compared with each other and they are well consistent. The discharge curve almost coincides with the calibration fitting curve of PD source when the test distance is 7.00 cm. Finally, it is concluded that the discharge intensity calibration model of PD source is accurate, which is of great significance in evaluating the extent of insulation damage.

A UHF Sensor Based Partial Discharge Monitoring System for Air Insulated Electrical Substations

A Partial Discharge (PD) monitoring system for Air Insulated Substations (AIS) has been developed for monitoring of partial discharge activities of various insulators and insulating media that are distributed over an electrical substation. The proposed system utilizes four UHF sensors, which are to be placed around the area of interest so that any impulsive UHF signal originated from PD activity can be captured and subsequently processed for detection and localization of multiple PD sources. To improve the localization accuracy and to improve the detection specificity, i.e. to eliminate false impulsive UHF signals originated from other sources, a novel nearest-neighbour based approach is proposed, which utilizes the physical layout of the AIS to identify most probable PD source locations. The proposed algorithm can accurately find the PD source locations even in the presence of high time-delay-of-arrival (TDOA) estimation error by making necessary corrections in the PD source location estimation. The proposed algorithm has been field trialed in many small and medium-size distribution substations. A detailed comparative study with the other algorithms in different extreme situations has been performed with the help of simulation and laboratory-based testing. The obtained results are worthwhile for supporting widespread implementation of the proposed method.

An Acoustic Time Reversal Technique to Locate a Partial Discharge Source: Two-Dimensional Numerical Validation

In this paper, the time reversal (TR) technique is extended to localize a partial discharge (PD) source in the acoustic regime. The proposed acoustic TR method consists of three steps, a) the acoustic waves emitted by the PD source are recorded by at least one sensor, b) the recorded signal(s) are reversed in time and back injected into the considered medium via numerical simulation, and c) an appropriate criterion is applied to identify the focal spot, which corresponds to the PD source location. We present a numerical validation of the proposed method in which the acoustic wave equations are solved numerically using the two-dimensional (2D) finite-difference time-domain (FDTD) method. The minimum entropy criterion is used to localize the PD source. The numerical results reveal that the proposed acoustic TR method is very efficient in localizing the PD source in the presence of the transformer windings.

A Method for the Localization of Partial Discharge Sources in Transformers Using TDOA and Truncated Singular Value Decomposition

Ultrasonic sensor detection has been widely used in partial discharge (PD) location. However, this PD localization method based on an ultrasonic sensor is computationally expensive, inaccurate, and sensitive to the time differences. A localization method for PD based on time difference of arrival (TDOA) and truncated singular value decomposition (TSVD) regularization is proposed in this paper, which can effectively ensure the accuracy of PD location in a transformer. First, the PD signal is received by the ultrasonic sensor, and an oscilloscope is used to receive the ultrasonic signal from the operational amplifier. Second, a spherical linear transformation method is used to transform the general nonlinear localization equations. Third, a TSVD regularization method is used to solve the linear equations. Then, to eliminate the influence of coordinate selection on the matrix ${b}$ , the matrix ${b}$ is centred. By dealing with the balance of matrix ${A}$ , the ill-conditioned degree of linear equations is weakened. Finally, the optimal solution serves as a good approximation of the original problem. The proposed method is applied to the PD location in a transformer, and the localization results are compared with direct Newton iteration and direct TSVD methods. A comparison of the experimental results shows that the method in this paper can reduce the ill-conditioned degree of a linear equation, and effectively weaken the influence of time difference error on the solution. The localization results of this method are better than the other two methods.