Gas Insulated Switchgear Partial Discharge 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.

Domain-alignment multitask learning network for partial discharge condition assessment with digital twin in gas-insulated switchgear

Deep-learning-driven methods have made great progress in the condition assessment of partial discharge (PD) which including diagnosis and location in gas-insulated switchgear (GIS). However, these methods perform diagnosis and location as two separate tasks and ignore the coupling relationship. In addition, these methods all require obtaining sufficient samples to develop models, and the model becomes ineffective when there is a significant difference in sample distribution. Therefore, we propose a novel domain-alignment multitask learning network (DAMTLN) for condition assessment including diagnosis and location assisted by digital twin. Firstly, a digital virtual model is established to assist the actual condition assessment of GIS PD. Then, a novel multitask network is constructed to mine the coupling relationship between the two tasks. Finally PD condition assessment guided by a digital twin model are achieved via a combination of local-maximum-mean-discrepancy-based and adversarial -based domain adaptation, in which fine-grained information on each category is captured. Experimental results show that the proposed DAMTLN achieved a diagnostic accuracy of 98.73%, and the mean absolute error of location was 9.06 cm, which were significantly better than the results of other methods. The DAMTLN thus provides a new avenue for PD diagnosis and location driven by ‘data–physics’ coupling.

Gas Insulated Switchgear Partial Discharge Type Identification Based on Frequency Division Features and Improved Long Short-Term Memory Neural Networks

In order to ensure the safe and stable operation of gas insulated switchgear (GIS), a new method for identifying the type of partial discharge in GIS is proposed. First, the GIS partial discharge experimental platform produced four partial discharge signals. Second, the energy moments are calculated at each frequency to produce the frequency division features after the four partial discharge signals have been decomposed using variational modal decomposition (VMD). Finally, the type of GIS partial discharge is determined using the LSTM model enhanced by the quantum cuckoo search (QCS) algorithm. The experimental results show that the proposed method can successfully identify four types of partial discharge in GIS with an accuracy of 99%.

Application of SCNGO-VMD-SVM in Identification of Gas Insulated Switchgear Partial Discharge

Application of SCNGO-VMD-SVM in Identification of Gas Insulated Switchgear Partial Discharge

Deep Domain-Invariant Long Short-Term Memory Network for Partial Discharge Localization in Gas-Insulated Switchgear

Localization based on differences in the timing of an ultrahigh-frequency signal is difficult to apply in actual settings because of the strict requirement for accurate calculation of the difference in arrival time between discharge pulses. With the development of digital twinning, the operational state of equipment is mapped onto a digital virtual model, and large-scale simulation data can be obtained to assist in localization of partial discharge (PD) in gas-insulated switchgear (GIS). With this aim, this paper proposes a novel deep domain-invariant long short-term memory network (LSTM) for GIS PD localization. Firstly, an attention residual LSTM is constructed to automatically extract discriminative features and effectively obtain the temporal and spatial dependences between features. Collaborative domain adaptation is proposed to transfer the diagnostic knowledge learned from large-scale simulation data to GIS PD localization. Via maximum mean discrepancy-based domain adaptation and domain adversarial training, discrepancies between samples in the two domains are reduced so as to achieve a more perfect match. The domain-invariant LSTM achieved an average localization accuracy of less than 11 cm, and the standard deviation of the error reached 8.15. The domain-invariant LSTM thus provides a novel solution for the high-precision and robust localization of PD in GIS.

A PRPD-Based UHF Filtering and Noise Reduction Algorithm for GIS Partial Discharge.

The online detection of partial discharge (PD) in gas-insulated switchgear (GIS) is a crucial and powerful tool for maintaining their reliability. However, extracting weak discharge signals from strong disturbances is a significant challenge. The presence of noise can hamper the identification and localization of PD types, making the extraction of pure PD signals the focus of current research. This paper proposes a PRPD-based PD filtering algorithm that analyzes interference using the output information from PRPD and sets threshold parameters for noise reduction processing. This method is mainly used for secondary noise reduction at a later stage, without analyzing the noise source, to achieve effective signal acquisition while retaining the characteristics of the PD signals, thereby improving the system's sensitivity and the signal's purity.

Partial discharge diagnosis via a novel federated meta-learning in gas-insulated switchgear.

Several deep learning partial discharge (PD) diagnostic approaches have been developed in recent years to guarantee the security and stability of gas-insulated switchgear (GIS). The centralized training method requires multiple clients to jointly obtain as much data as possible to train the model to achieve excellent performance, which is impractical due to conflicts of interest and privacy protection. Furthermore, because of differences in the distribution of client data and the presence of a small sample, achieving high-precision and robust diagnosis for each client is an urgent problem. To that end, a novel personalized federated meta-learning (FML) is proposed in this paper to address the aforementioned issues. It develops reliable and personalized PD diagnosis models by collaborating with multiple clients and solves the problem of small sample diagnosis through scenario training under the premise of protecting data privacy. The experimental results show that the FML proposed can diagnose GIS PD with high precision and robustness for each client while maintaining privacy. The diagnostic accuracy of the FML proposed in this paper, especially for on-site unbalanced small sample clients, is 93.07%, which is significantly higher than that for other methods. It serves as a model for the collaborative development of an effective GIS PD diagnostic model.

Measuring Point Localization and Sensor Pose Control for Gas Insulated Switchgear Partial Discharge Detection Robot

Measuring Point Localization and Sensor Pose Control for Gas Insulated Switchgear Partial Discharge Detection Robot

External UHF Sensor for GIS PD Insulation Defect from Resin Hole

In view of the effective perception demand of PD (Partial Discharge) insulation defect of in-service GIS (Gas Insulated Switchgear), the external UHF (Ultra High Frequency, 0.3−3 GHz) sensing technology for GIS PD insulation defect from resin hole is studied in this paper. Firstly, rectangular waveguide theory is used to analyze the electromagnetic wave propagation characteristics from rectangular resin hole in GIS disc insulator. It is found that the energy of leaked electromagnetic wave mainly concentrates in the frequency band above 1 GHz. Then based on this frequency band characteristics, a cavity-backed bowtie sensor for GIS PD external detection is designed optimized by HFSS finite element software, in band of 1–3 GHz, its maximum VSWR is 7, the minimum is 1.1, the average is 2.9, and in band of 0.3–1 GHz, the maximum and minimum VSWR is 42.8 and 7, the average is 20.6. Finally, a sensor prototype is made, its sensitivity experiment is carried out on 220 kV GIS, experimental results show that the designed sensor can stably detect the PD electromagnetic wave signal leaked from the resin hole, and the signal energy in frequency band range is consistent with the theoretical analysis results.

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.

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.

A novel method for small and unbalanced sample pattern recognition of gas insulated switchgear partial discharge using an auxiliary classifier generative adversarial network

AbstractConvolutional neural networks (CNNs) have been widely used for gas‐insulated switchgear (GIS) partial discharge (PD) pattern recognition due to their powerful feature extraction ability. However, there is commonly a scarcity of fault samples due to low insulation failure rate of GIS equipment, which degrades the diagnostic performance of these CNN networks when directly applied to small and unbalanced datasets. Therefore, we propose a novel auxiliary classifier generative adversarial network for GIS PD pattern recognition for small and unbalanced samples. First, we propose using synchrosqueezed wavelet transform to extract time‐frequency characteristics of PD pulses and obtain a time‐frequency image with high energy aggregation and time‐frequency distribution rate. Then, we propose an improved generative adversarial network with an auxiliary classier and self‐attention mechanism, which can generate high‐quality PD samples for situations with few classes. Experiments show that our proposed method can reach 95.75% recognition accuracy for small datasets, which is the highest among several comparable methods. Furthermore, the proposed method has excellent and stable recognition performance for various unbalanced datasets.

GIS Partial Discharge Pattern Recognition Based on Time-Frequency Features and Improved Convolutional Neural Network

Different types of partial discharge (PD) in gas-insulated switchgear (GIS) cause different damage to GIS insulation, correctly identifying the PD type is very important for evaluating the insulation status of GIS. This paper proposes a PD pattern recognition method based on an improved feature fusion convolutional neural network (IFCNN) to fully use the time-frequency features of PD pulses to realize PD pattern recognition. Firstly, the one-dimensional time-domain feature sequence of the PD pulse and the corresponding wavelet time-frequency diagram are applied as inputs. Secondly, the convolutional neural network (CNN) with two parallel channels is used for feature extraction, the extracted fault information is fused, and the shallow features of the wavelet time-frequency diagram are fused to prevent feature loss caused by pooling operation. Finally, the extracted features are sent to the classifier to recognize different types of PD. The discharge data of different types of PD are obtained for testing by experiments and simulation. Compared with 1-D CNN and 2-D CNN under the same specification, the proposed method can mine more potential local features of discharge pulses by fusing the time-frequency features of PD pulses in different dimensions, and improves the recognition accuracy to 95.8%.

A Domain Adaptive Deep Transfer Learning Method for Gas-Insulated Switchgear Partial Discharge Diagnosis

Intelligent fault diagnosis methods, especially convolutional neural network (CNN), have made significant progress in gas-insulated switchgear (GIS) partial discharge (PD) diagnosis, which are attributable to two reasons: 1) the training and testing samples come from the identical distribution; 2) there are massive labeled data with PD information. However, owing to the specific operating conditions of GIS, collecting massive samples from the same distribution is difficult in field conditions. With the purpose of resolving the data dilemma of conventional diagnosis methods in the field, we propose a domain adaptive deep transfer learning (DADTL) CNN for small samples GIS PD diagnosis. First, we adopted a CNN to automatically extract transferable features from PD samples. Then, a multilayer DADTL is developed to reduce the marginal and conditional distribution of learned transferable features, and Sliced Wasserstein distance (SWD) is employed as a penalty to reduce the negative migration. The experimental verification was performed on experimental and on-site GIS PD datasets. The results show that DADTL CNN can effectively achieve a high-accuracy diagnosis of GIS PD for small samples. Compared with other methods, the DADTL CNN can achieve more accurate and robust GIS PD diagnosis for on-site small samples.

A novel federated deep learning framework for diagnosis of partial discharge in gas-insulated switchgear

In recent years, many different deep learning methods have been developed to ensure the safe and stable operation of gas-insulated switchgear (GIS). However, the use of these methods to achieve excellent results depends on obtaining as much training data as possible, which is difficult to accomplish because of conflicts of interest among different clients and privacy concerns. To address this issue, this paper proposes a novel federated deep learning (FDL) for the diagnosis of partial discharge (PD) in GIS. A federated learning (FL) based on an improved federated averaging algorithm is proposed, which allows different clients to collaboratively participate in model training and preserves data privacy. In addition, a novel subtractive attention Siamese network is introduced for feature extraction and classification, which achieves the high-precision classification of unbalanced data. Experimental results showed that the diagnostic accuracy of the proposed FDL reached 95.61%, which was significantly higher than that achieved by other methods. The proposed FDL can also achieve excellent performance in the case of unbalanced samples and small samples. As a distributed learning, FL does not require clients to share data, and clients can collaboratively develop an effective global diagnostic model, which provides a novel solution for GIS PD diagnosis.

GIS partial discharge pattern recognition via a novel capsule deep graph convolutional network

Deep learning (DL) has attracted significant attention in partial discharge (PD) pattern recognition of gas-insulated switchgear (GIS). However, the existing DL method extracts numerical information and ignores spatial correlation information. In addition, once the sample size is reduced, model performance is severely degraded. To address these issues, a novel capsule deep graph convolutional network (CDGCN) is employed for GIS PD pattern recognition. First, the GIS PD data is transformed to graph data. Then, the CDGCN is established for GIS PD pattern recognition. The introduction of a ChannelSortPooling layer enriches the graph representation information and filters out the feature values with higher information content. In addition, the multi-view graph representation is converted to a capsule form, and a capsule network with stronger feature extraction capability is introduced to improve the accuracy of the final graph representation. The results reveal that the proposed CDGCN can effectively diagnose the GIS PD and is more accurate than the traditional DL method. The proposed CDGCN can handle small samples pattern recognition problems and provides a reliable reference for GIS PD pattern recognition.

A novel differentiable neural network architecture automatic search method for GIS partial discharge pattern recognition

Convolutional neural network (CNN) has been extensively used in pattern recognition of partial discharge (PD) in gas-insulated switchgear (GIS) because of its powerful feature extraction ability. However, at this stage, manual trial and error is needed to construct the CNN. Moreover, the model is designed for specific datasets, which will cause domain bias when applied to a new dataset. Therefore, a novel differentiable neural architecture search method is proposed to automatically construct a GIS PD pattern recognition model. First, a factorized hierarchical search space is used to design the CNN architecture. Then, a discrete search space is relaxed into a continuous search space through a search strategy based on Gumbel–softmax. Experiments show that the recognition accuracy of the proposed method can reach 97.625%. Furthermore, the proposed method has strong robustness and high precision against noise and strong tolerance for unbalanced datasets.

Partial discharge fault diagnosis based on zernike moment and improved bacterial foraging optimization algorithm

Feature extraction and classifier optimization are important parts in the diagnosis of partial discharge (PD). Traditional feature extraction methods generally target not the images but structured data. Traditional classifier optimization methods have many disadvantages such as the many parameters that need to be adjusted, difficulty in determining the best parameters, and the ease in falling into local optimal solutions. To eliminate these defects, a PD diagnosis method of gas-insulated switchgear (GIS) based on the Zernike moment and improved support vector machine (SVM) is proposed. Firstly, five typical PD models of corona discharge, metal particle discharge, floating electrode discharge, void discharge, and surface discharge were designed. Then, the UHF sensor was used to collect discharge signals of different defects and different types of interference signals to build a PD phase-resolved pulse sequence (PRPS) patterns library. After that, the Zernike moment of the PRPS images were extracted as the PD features. Finally, the SVM model based on the differential evolution adaptive bacterial foraging optimization (DEABFO) algorithm was used to fulfill the GIS PD diagnosis. The results show that the Zernike moment feature extraction method is suitable for PD PRPS images, and DEABFO improves the convergence speed of the SVM parameter optimization process and can effectively avoid the probability of falling into the local optimum. Compared with other image feature extraction methods and other SVM optimization algorithms, this method has high accuracy in GIS PD fault diagnosis, with the recognition rate reaching 91.23%.

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.

GIS Partial Discharge Pattern Recognition Based on a Novel Convolutional Neural Networks and Long Short-Term Memory.

Distinguishing the types of partial discharge (PD) caused by different insulation defects in gas-insulated switchgear (GIS) is a great challenge in the power industry, and improving the recognition accuracy of the relevant models is one of the key problems. In this paper, a convolutional neural network and long short-term memory (CNN-LSTM) model is proposed, which can effectively extract and utilize the spatiotemporal characteristics of PD input signals. First, the spatial characteristics of higher-level PD signals can be obtained through the CNN network, but because CNN is a deep feedforward neural network, it does not have the ability to process time-series data. The PD voltage signal is related to the time dimension, so LSTM saves and analyzes the previous voltage signal information, realizes the modeling of the time dependence of the data, and improves the accuracy of the PD signal pattern recognition. Finally, the pattern recognition results based on CNN-LSTM are given and compared with those based on other traditional analysis methods. The results show that the pattern recognition rate of this method is the highest, with an average of 97.9%, and its overall accuracy is better than that of other traditional analysis methods. The CNN-LSTM model provides a reliable reference for GIS PD diagnosis.