Multi-source Partial Discharge Research Articles

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.

Classification of Multi-Source Partial Discharge in GIS

Abstract Partial discharge (PD) analysis is an important method to identify electrical equipment faults. However, due to the influence of environmental electromagnetic interference and other equipment discharges in the power supply circuit during the on-site PD test, the collected PD signal is the combination of various PD and noise, which makes it challenging to identify the type or source of on-site partial discharge. A multi-source PD classification method based on the multiple threshold and K-means clustering algorithm is pro-posed to solve this problem. First, wavelet transform with an improved threshold is used to initially denoise the original signal. Then the multiple threshold method is used to filter the residual ripple between every two pulses. Finally, spectrum analysis is carried out for each pulse, and the maximum value of the spectrum is extracted as the feature quantity. Then, the K-means algorithm is used to classify the PD pulses, and the PRPD images of PD pulses in different frequency domains are obtained. The results show that the method adopted in this paper can denoise, extract and classify the PD signals collected on the site, and the PRPD diagram of PD pulse after classification is more obvious, which provides a basis for identifying the PD source on the site.

Multi-source partial discharge pattern recognition algorithm based on DCGAN-Yolov5

This study aims to overcome the complication that deep learning-based pattern recognition in partial discharge (PD) diagnosis of gas-insulated switchgear (GIS) can only identify single-source partial discharge but not multi-source partial discharge. Specifically, a GIS multi-source partial discharge detection algorithm based on Deep Convolution Generative Adversarial Networks and You Only Look Once (DCGAN-YOLOv5) is proposed. First, the Phase Resolved Partial Discharge (PRPD) features of multi-source PD are analyzed, a GIS experiment platform is established, and four typical PD defects are simulated. Besides, sample data are collected, and the DCGAN network is used for sample expansion. Then, a YOLOv5 network model is designed, and a spatial and channel attention mechanism is added to the feature extraction network with a positive sample equilibrium strategy. Finally, the effectiveness of the proposed algorithm is verified using laboratory data and field data collected from a 220 kV substation. The experimental results demonstrate that the proposed algorithm can effectively detect the multi-source PD features in PRPD patterns under complex noise and thus successfully identify the types of multi-source PD. The mean Average Precision (mAP) can reach 98.4%. The precision of single-source PD and multi-source PD can reach 95.2% and 89.3%, when testing with field data.

Multi-source partial discharge diagnosis in gas-insulated switchgear via zero-shot learning

Partial discharge (PD) diagnosis in gas insulation switchgear (GIS) has been continuously studied as a hot issue. The diagnostic model developed at this stage requires the possible combination of all single PDs, and each combination has sufficient data. However, the demand for the number of multi-source PD samples increases in the form of an index due to the accidental and concurrency of different types of PD, where it is unrealistic to collect sufficient PD samples. To this end, we proposed zero-shot learning (ZL) to diagnose multi-source PD. The ZL diagnosis multi-source PD only required single PD for training. Specifically, an attention residual network was introduced for feature extraction. Then, a novel semantic descriptor was proposed to obtain the semantic attributes, and the semantic embedded features were mapped to the visual space with a full connection network. Finally, the relationship network adopted the similarity between the visual features and semantic embedded features, and the multi-source PD diagnosis was realized. In addition, the multi-source PD experiment in GIS was constructed to verify the proposed ZL. The experimental results demonstrated that the ZL could achieve 90.25 % accuracy when only single PD participated in the training. It provided novel solutions for the diagnosis of multi-source PD in GIS.

Multi-Source Partial Discharge Fault Location with Comprehensive Arrival Time Difference Extraction Method and Multi-Data Dynamic Weighting Algorithm

The location of the partial discharge source is an important part of fault diagnosis inside power equipment. As a key step of the ultra-high frequency location method, the extraction of the time difference of arrival can generate large errors due to interference. To achieve accurate time difference extraction and further multi-source partial discharge location, a location method with comprehensive time difference extraction and a multi-data dynamic weighting algorithm is proposed. For time difference extraction, the optimized energy accumulation curve method applies wavelet transform and mode maximization calculations such that it overcomes the effect of interference signals before the wave peak. The secondary correlation method improves the interference capability by performing two rounds of correlation calculations. Both extraction methods are combined to reduce the error in time difference extraction. Then, the dynamic weighting algorithm effectively utilizes multiple data and improves the location accuracy. Experimental results on multi-source partial discharge locations performed in a transformer tank validate the accuracy of the proposed method.

Generative Zero-Shot Learning for Partial Discharge Diagnosis in Gas-Insulated Switchgear

Class imbalance exists widely in partial discharge (PD) diagnosis of gas-insulated switchgear (GIS). On the one hand, while atypical PDs occur sometimes but the sample is scarce. On the other hand, due to the contingency and concurrency of PDs, the demand for multi-source PD increases exponentially, making it difficult to obtain sufficient data. This study focuses on zero-shot diagnosis in extreme cases where only typical defects are available and where atypical and multi-source PDs are not available during training. And we propose a semantic rectifying discriminative generative adversarial network (SRDGAN) for zero-shot diagnosis. The proposed SRDGAN trains the generator from the visible samples and semantic attributes, and generates the unseen samples from the unseen semantic attributes and trains the classifier for zero-shot diagnosis. First, a semantic rectifying module is designed to correct the structure between the visual and semantic space to make the semantic features distinguishable. Then, the latent discriminative attributes are extracted from the visual features, and an attribute embedding module is designed. Finally, PD diagnosis is performed on feature generation and classification modules. The proposed SRDGAN is validated on two datasets. The experimental results illustrate that the SRDGAN solves zero-sample GIS PD diagnosis with >90% accuracy.

Research on multi-source partial discharge localization based on improved FCM-LOF fuzzy clustering algorithm

At present, the ultra-high frequency method is widely used in the single-source partial discharge (PD) location of substation sites; however, there are issues with the positioning accuracy being too low, and it is difficult to meet the needs of multi-source PD positioning. Based on the local outlier factor (LOF) outlier idea and fuzzy c-means (FCM) clustering algorithm, this paper proposes the FCM-LOF algorithm to be applied to the research of multi-source PD positioning. This algorithm removes the discrete points of the original time difference data set based on the principle of the local density threshold, reduces the clustering center error, and improves the accuracy of multi-source positioning. The main research contents of this article are as follows. Firstly, using smoothing filter processing and the energy accumulation method, we collect the time difference of dual- and triple-source PD signals as the data set to be processed, and carry out the laboratory simulation experiment. Secondly, comparing the clustering effects of the k-means and FCM algorithms, it is found that the clustering accuracy of the FCM algorithm is significantly better than that of the k-means algorithm, and the positioning error is reduced by 27.3%. Then, using the neighborhood density and outlier factor to eliminate abnormal data, combined with FCM fuzzy clustering, an improved FCM-LOF algorithm is proposed. Compared with the FCM algorithm, the positioning error of this algorithm is reduced by 11.6%. It is suitable for multi-source positioning and has a greater improvement in noisy environments. Finally, the improved algorithm is applied to a field simulation test, which verifies the accuracy of the algorithm. We also study the factors affecting the positioning accuracy of the improved algorithm. The research in this paper can provide a powerful reference for multi-source PD detection of power equipment.

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.

The pattern recognition of multisource partial discharge in transformers based on parallel feature domain

The traditional diagnosis methods for the multisource partial discharge in transformer have several problems, such as the poor effect of signal separation, the inaccuracy of the extracted signal features and the over fitting of the neural network algorithm used for pattern recognition. In order to solve the above problems, the improved algorithm of stacked auto-encoder is adopted to extract the adaptive features of partial discharge one dimensional defect signals in time domain and two dimensional defect signal in time frequency domain, and a parallel feature space combining the features in time domain with that the time frequency domain is constructed. For the first time, the transformed L1 norm is utilized as the regular term of the loss function which is solved by the proximally guided stochastic subgradient in the improved auto-encoder. The trained neural network model can be utilized for pattern recognition of the different multisource partial discharge signals. The results show that the proposed method has a higher recognition rate for the different multisource partial discharge defects than that in the traditional method, and alleviates the problem of over fitting to a certain extent.

Multi-source partial discharge detection and rapid localization of switchgear based on comprehensive diagnosis method

Partial discharge detection is an effective method to find potential insulation defects in switchgear. At present, the switchgear is mainly composed of busbars, circuit breakers, grounding knives, cables and other components, which leads to multiple faults and simultaneous discharges in the partial discharge detection process of the switchgear. In order to realize the partial discharge detection and rapid location of the switchgear, based on the introduction of the detection methods and location principles of ultrasonic, transient ground voltage and UHF partial discharge, this paper proposes to use ultrasonic, transient ground voltage and UHF detection technology to diagnose and analyse the multi-source partial discharge of the switchgear and realize the detection of partial discharge; the multi-source partial discharge of the switchgear is realized by the rapid localization method combining the UHF time difference method and the flat surface method. Through partial discharge detection and localization analysis of a 35kV switchgear, the feasibility of multi-source partial discharge diagnosis and rapid localization method is verified, which provides a basis for on-site inspection personnel to realize multi-source partial discharge detection and localization of switchgear.

Improved density peak clustering for separation of multiple source partial discharge in large generators

Multiple source partial discharge (PD) separation is indispensable in on-line PD measurements, particularly for large generators. For this purpose, an improved density peak clustering algorithm for the separation of a multi-source PD is presented. DPC is characterized by the selection of cluster centers from the decision graph labeled by local density and distance, while human involvement hinders its application toward automatic analysis. For complex PD data sets, the boundary between cluster centers and non-center points is often so fuzzy on the decision graph that the results are affected by human subjectivity. To overcome these drawbacks, an improvement to the determination method for cluster centers is presented, which is a key link in clustering. Based on the decision graph, a decision quantity is developed by multiplying the local density and distance. The local density peak points, with the decision quantity meeting specific criterion, are roughly defined as potential cluster centers, then the exact cluster centers are finally determined based on the defined density-reachable relationships between the potential cluster centers. The results of actual PD data sets demonstrate the effectiveness of the proposed algorithm and reveal the ability to separate PD data sets of arbitrary shape or uneven density, indicating the wide applicability prospects of the technique.

Separating Multi-Source Partial Discharge Signals Using Linear Prediction Analysis and Isolation Forest Algorithm

Partial discharge (PD) detection is an effective way to find defects and diagnose the insulation condition of power equipment. During manufacturing and operating, there may exist multi-source PD signals in the equipment, which could seriously affect the accuracy of subsequent defect analyses. In this paper, a universal separation methodology using linear prediction analysis (LPA) and the isolation forest algorithm (IFA) is introduced. By approximating the present waveform point by a linear combination of several points in the past, a 12-D linear prediction cepstrum coefficient (LPCC) feature space can be established which can accurately characterize PD waveforms. In this paper, we applied principal component analysis (PCA) to reduce the feature space to 2-D space. The IFA was adopted to separate multi-source PD signals which quantified the degree of clustering and added one more parameter based on the original features. Thus, the proposed method can separate multiple PD sources effectively even if the difference in the features of different types of PDs is small. In addition, the proposed algorithm easily removed the noise points during the separation process. The algorithm was applied to different types of PD signals on 35-kV transformers and a three-source PD data set on a 252-kV gas-insulated switchgear (GIS) platform. The separation results confirm that the proposed algorithm can effectively separate and distinguish various PD signals.

Partial Discharges Pattern Recognition of Transformer Defect Model by LBP & HOG Features

Partial discharge (PD) measurement and identification have great importance to condition monitoring of power transformers. In this paper, a new method for recognition of single and multi-source of PD based on extraction of high level image features has been introduced. A database, involving 365 samples of phase-resolved PD (PRPD) data, is developed by measurement carried out on transformer artificial defect models (having different sizes of defect) under a specific applied voltage, to be used for proposed algorithm validation. In the first step, each set of PRPD data is converted into grayscale images to represent different PD defects. Two “image feature extraction” methods, the Local Binary Pattern (LBP), and the Histogram of Oriented Gradient (HOG), are employed to extract features from the obtained gray scale images. Different variants of Support Vector Machine (SVM) are adjusted for optimal classification of PD sources in this process. Impact of the employed parameters in the image processing such as image resolution, random noise, and phase shift, on identification accuracy is investigated and addressed. It is shown that by using HOG-SVM method 99.3% accuracy can be achieved. This is hardly affected by various external factors. Two case studies have been conducted on multi-source PD for evaluating the performance of the proposed algorithm. A void defect is implemented into the transformer model and the resultant recorded signal is used for the study. The DBSCAN algorithm is used as the mean of PD source clustering and sub-PRPD pattern development. It is shown that HOG-SVM method has superior performance in identifying active sources, under sub-PRPD pattern application.

Multi-source Partial Discharge Identification of Power Equipment Based on Random Forest

At present, research on partial discharge type identification of power equipment mainly focused on single source discharge, and a small amount of multi-source discharge research also focused on pulse separation of multi-source signals. Signal separation could lose a lot of valid signals and wasted information resources. In this paper, a multi-source partial discharge type identification method based on Random Forest (RF) is proposed. Firstly, in the feature extraction, the statistical characteristics of the multi-source data are extracted directly instead of signal separation, and the discharge information is fully utilized. Secondly, in terms of the choice of classifier, considering that the traditional SVM method can not handle the value of 0, this paper selects the random forest strong classifier with good anti-noise ability to identify the multi-source discharge type. The test results show that this method is effective and the recognition rate of PD is as high as 98%.

Fractal‐based autonomous partial discharge pattern recognition method for MV motors

On-line partial discharge (PD) monitoring is being increasingly adopted to improve the asset management and maintenance of medium-voltage (MV) motors. This study presents a novel method for autonomous analysis and classification of motor PD patterns in situations where a phase-reference voltage waveform is not available. The main contributions include a polar PD (PPD) pattern and a fractal theory-based autonomous PD recognition method. PPD pattern that is applied to convert the traditional phase-resolved PD pattern into a circular form addresses the lack of phase information in on-line PD monitoring system. The fractal theory is then presented in detail to address the task of discrimination of 6 kinds of single source and 15 kinds of multi-source PD patterns related to motors, as outlined in IEC 60034. The classification of known and unknown defects is calculated by a method known as centre score. Validation of the proposed method is demonstrated using data from laboratory experiments on three typical PD geometries. This study also discusses the application of the proposed techniques with 24 sets of on-site PD measurement data from 4 motors in 2 nuclear power stations. The results show that the proposed method performs effectively in recognising not only the single-source PD but also multi-source PDs.

Classification of simultaneous multiple partial discharge sources based on probabilistic interpretation using a two-step logistic regression algorithm

In online condition assessment monitoring of high voltage (HV) insulators, it is often required to identify multiple, simultaneously activated partial discharge (PD) sources that happen in the insulation of the HV apparatus. Phased resolved partial discharge (PRPD) patterns are commonly used to identify PD sources. However, multiple, concurrent PD sources sometimes result in partially overlapped patterns, which make them hard to be identified. In this paper, we develop an accurate, reliable algorithm by constructing a novel two-step logistic regression (LR) model to conduct probabilistic identification of multi-source PDs. To this end, principal component analysis is applied on a database to construct a low dimensional space associated with single-source PDs. Samples of multi-source PDs are then projected onto this space and one-class kernel support vector machine is adopted to distinguish multi-source PDs from single-source ones. Finally, classification is performed by estimating the probability (degree of membership) of each PRPD pattern arising from different multi-source PDs following two rounds of LR modeling. To evaluate the performance of our proposed method, we study a number of multi-source PD models to simulate common defects of Gas-Insulated Switchgear (GIS) in small-scale laboratory test cells with realistic SF6 gas condition. Observations are obtained using fingerprints generated by a novel approach from recorded PRPD patterns. Comprehensive performance evaluation of the proposed algorithm and its advantages are conducted and the development of analytical equations is presented. The results of this paper can be used to design a solid basis for an automated multi-source classification system, which facilitates multi-source PD identification in early stages and safe operation of HV apparatus.

Research on Separation of PD Signals Originated from Different Sources in Substation

The existing system of partial discharge location in substation utilizes Omni-directional UHF sensors array which can achieve single-source partial discharge source location. But for multi-source UHF partial discharge signals, the system is inaccurate and can’t distinguish types of partial discharge. This paper proposed to the method multi-source partial discharge signals acquisition and separation based on different characteristics of different types of UHF partial discharge signals in the time domain and frequency domain. Through continuously collected and stored UHF partial discharge pulse waveform, the sampling pulse sequence is analyzed in time-frequency domain to extract characteristic parameters. And then using fuzzy cluster, it can distinguish multi-source UHF partial discharge signals and discharge signals generated by the interference to achieve multi-source UHF partial discharge signals location. And it can lay the foundation of identification of the type of partial discharge. Field tests show that the method is feasible.

Effectiveness of Partition and Graph Theoretic Clustering Algorithms for Multiple Source Partial Discharge Pattern Classification Using Probabilistic Neural Network and Its Adaptive Version: A Critique Based on Experimental Studies

Partial discharge (PD) is a major cause of failure of power apparatus and hence its measurement and analysis have emerged as a vital field in assessing the condition of the insulation system. Several efforts have been undertaken by researchers to classify PD pulses utilizing artificial intelligence techniques. Recently, the focus has shifted to the identification of multiple sources of PD since it is often encountered in real‐time measurements. Studies have indicated that classification of multi‐source PD becomes difficult with the degree of overlap and that several techniques such as mixed Weibull functions, neural networks, and wavelet transformation have been attempted with limited success. Since digital PD acquisition systems record data for a substantial period, the database becomes large, posing considerable difficulties during classification. This research work aims firstly at analyzing aspects concerning classification capability during the discrimination of multisource PD patterns. Secondly, it attempts at extending the previous work of the authors in utilizing the novel approach of probabilistic neural network versions for classifying moderate sets of PD sources to that of large sets. The third focus is on comparing the ability of partition‐based algorithms, namely, the labelled (learning vector quantization) and unlabelled (K‐means) versions, with that of a novel hypergraph‐based clustering method in providing parsimonious sets of centers during classification.

Robust Heteroscedastic Probabilistic Neural Network for multiple source partial discharge pattern recognition – Significance of outliers on classification capability

Among various insulation diagnostic techniques utilized by researchers and personnel handling power equipment, partial discharge (PD) recognition and analysis has emerged as a vital methodology since it is inherently a non-intrusive testing strategy. Of late, the focus of researchers has shifted to the identification and classification of multiple sources of PD since it is most often encountered in practical insulation systems of power apparatus. Researchers have carried out studies to recognize multi-source PD and expounded the difficulties experienced in discriminating such discharge patterns. It has also been observed that identification of such patterns becomes increasingly difficult with the degree of overlap. Review of recent research studies indicates that classification of fully overlapped patterns is yet an unresolved issue and that techniques such as Mixed Weibull Functions, neural networks (NN) and Wavelet Transformation have been attempted with reasonable degree of success for single source and partially overlapped PD patterns only. This research study focuses on extending the previous work attempted by the authors in utilizing the novel approach of Heteroscedastic Probabilistic Neural Network (HRPNN) for classification of single source PD patterns to that of multiple PD sources also. Further, a Robust Heteroscedastic Probabilistic Neural Network (RHRPNN) is implemented for the classification of multi-source PD patterns. The RHRPNN utilizes the jackknife procedure for handling problems associated with training the neural network due to the presence of outliers, thus providing a compact yet effective set of centers in terms of probability density functions. In addition to the previously utilized traditional statistical operators in the pre-processing phase, a Two Pass Split Window (TPSW) scheme has also been developed to study and compare the classification capability of the RHRPNN with that of HRPNN. Detailed analysis of the performance of RHRPNN is carried out to ascertain the influence of the smoothing parameter in classifying PD patterns, to determine the role played by the pre-processing techniques during classification and to find the significance of the parsimonious set of centers in eliminating the effect of outliers during classification. Finally, the ability of the HRPNN and the RHRPNN in classifying large dataset multiple source PD patterns obtained from varying applied voltages is analyzed for its further applicability in real-time PD pattern recognition studies.

Wavelet analysis for classification of multi-source PD patterns

Multi-resolution signal decomposition (MSD) technique of wavelet transforms has interesting properties of capturing the embedded horizontal, vertical and diagonal variations within an image in a separable form. This feature was exploited to identify individual partial discharge (PD) sources present in multi-source PD patterns, usually encountered during practical PD measurements. Employing the Daubechies wavelet, features were extracted from the third level decomposed and reconstructed horizontal and vertical component images. These features were found to contain the necessary discriminating information corresponding to the individual PD sources. Suitability of these extracted features for classification was further verified using a radial basis function neural network (NN). Successful recognition was achieved, even when the constituent sources produced partially and fully overlapping patterns, thus demonstrating the applicability of the proposed novel approach for the task of multi-source PD classification.