Transformer Fault Research Articles

Intelligent Diagnosis Method of Transformer Based on Oil Chromatographic Data

Abstract As one of the most important equipment in the operation of power system, ensuring the safe and stable operation of power transformer is a prerequisite for ensuring the normal supply of power grid. The failure of the power transformer will lead to the interruption of the power supply of the power grid and cause huge losses to the national economy. With the rapid development of China’s power grid scale and the continuous improvement of the intelligent construction of modern power system, the number of power equipment such as transformers is increasing. Therefore, it is necessary to make timely and accurate judgment on the status of key power transformation and distribution equipment in the power system, and grasp the operation of electrical equipment in real time, so as to ensure the reliability of power supply in the power system. This paper takes 220kV oil-immersed transformer as the research object. This paper considers the problems and characteristics of the current transformer fault diagnosis methods. By making full use of dissolved gas analysis (DGA) information in oil, a method for diagnosing faults in electrical transformers based on particle swarm optimization and generalized regression neural networks has been proposed. The simulation comparison experiment is carried out to select the most suitable transformer intelligent diagnosis method based on oil chromatographic data.

Predictive Model for Incipient Faults in Oil-Filled Transformers

Predictive Model for Incipient Faults in Oil-Filled Transformers

Analysis and Treatment of a Partial Discharge Fault in a 220kV Transformer

Abstract Partial discharge test of transformer is one of the most important detection items in the factory test of transformer, which is related to whether the transformer can operate safely and stably. This article will analyze a case of partial discharge exceeding the standard in a 220kV transformer. Based on the joint detection technology of ultrasonic partial discharge and ultra high frequency partial discharge, the defect location will be found. And the structure of the transformer will be researched to identify the cause of the defect. The diagnosis and treatment will be successfully completed. Finally, the conclusion will be verified through finite element analysis. This provides reliable guidance for partial discharge detection and rectification measures in transformers’ production and operation.

Influence of water vapor concentration on photoacoustic spectroscopy‐based characteristic gas analysis of transformer faults

AbstractThe water vapor in the ambient air affects the accuracy of the photoacoustic (PA) dissolved gas analysis system for transformer health monitoring. A laser PA system was evaluated by dry and humidified standard gases to study the influence of water vapor concentration on PA gas detection. Theoretical analysis was conducted on the effect of gas molecule relaxation on PA signal detection. A high‐frequency resonant PA cell and a low‐frequency nonresonant PA cell were used to detect acetylene (C2H2) and carbon monoxide (CO), respectively. The experimental results show that the PA signal of humidified CO is about 12 times higher than PA signal of dry gas for the resonant PA detection system, respectively. In addition, as the frequency is increased from 30 to 980 Hz, the PA signals of humidified and dry CO attenuate by 1.5 and 6.9 times, respectively.

An investigation of the Behaviour of Copper Electronic State Evolution on the Failure of Oil-paper Insulation Systems

Abstract Copper, as one of the most hazardous metals to transformer insulation systems, has attracted a great deal of interest from scholars at home and abroad. Unfortunately, the specific type of copper that causes catalytic aging of oil-paper insulation is still controversial. Therefore, in this paper, the effect of the electronic state evolution of copper in different oxidation states on the aging performance of the insulation systems was investigated by characterization techniques and performance evaluation. Cu2O has a stronger catalytic oxidizing effect on insulation system than CuO. This study elucidates the electronic state evolution of different oxidation states of copper and its influence on the insulation performance, which lays a good foundation for the advancement of transformer fault warning technology.

The Evaluation of Over Current Relay and Ground Fault Relay Settings on Power Transformers

Power transformer is an important component of the substation and has an important role in the electric power distribution system. During operation the power transformer can’t be isolated from distribution system when abnormal conditions or faults occur. A fault in the power transformer will have huge impact on the power transformer itself. The substation has an installed capacity of 60 MVA of power transformers, and during operation there is a possibility of relay failure, which may interrupt the power distribution process to the load. To avoid this failure, it is necessary to analyse the settings of Over Current Relay (OCR) and Ground Fault Relay (GFR) on the 60 MVA transformers. The analysis was carried out by simulating short circuit faults of the electrical system in the ETAP software application, and then configure the setting of the relay by referring to the magnitude of the short circuit current value for phase to phase and phase to ground. The results of this analysis are used as a reference to determine the OCR and GFR relay values in the power transformer. The results obtained show that the recommended value for OCR setting on 150 kV power transformers is 0.92 second while for GFR is 0.96 second, and for the 20 KVA power transformers the OCR and GFR setting values are 0.40 and 0.20 seconds respectively. Comparison based on the results of the setup data and the analysis data has carried out, there is difference in the working time of the relay, with a difference of 5%, where the value of the analysis results is not much different from the conditions in the field.

Adsorption sensitivity and sensing performance of Au-doped Ti3C2O2 gas sensor for fault characteristic gases in power transformer

By analyzing the compositions and concentration of dissolved gases in power transformer oil, it is possible to effectively identify and predict the types and degrees of internal faults in oil-immersed transformers. This technology is currently an important method for fault diagnosis and prevention of large oil-immersed power transformers. This study established two different adsorption systems, namely Ti3C2O2 and Ti3C2O2 doped with Au atom, for the detection of four fault characteristic gases (CO, CO2, C2H4, CH4) in power transformer oil. The research results indicate that Ti3C2O2 exhibits inadequate adsorption capabilities for the above gases. The maximum adsorption energy is only −0.440 eV, which belongs to physical adsorption. After doping with Au atoms, the adsorption sensitivity of Ti3C2O2 for the above gases was significantly improved, with a significant increase in adsorption energy and charge transfer. The average adsorption energy has increased by 435%, and the minimum adsorption energy is also −0.660 eV. The adsorption process changed from physical adsorption to chemical adsorption. These research results provide a theoretical basis for Au-doped Ti3C2O2 gas sensor in detecting fault characteristic gases in power transformer.

Analysis of interturn short circuit in regulating winding of power transformer based on field-circuit coupling

The interturn short circuit fault is one of the common faults in power transformers. At present, research on interturn short circuit faults focuses on high, medium, and low voltage windings, while there is relatively little research on interturn short circuit in regulating windings. Specifically, there is a lack of reported studies on the transient electromagnetic processes, magnetic field distribution, and electromagnetic force characteristics of interturn short circuits in regulating windings unconnected to the circuit. This study presents an actual fault scenario involving interturn short circuits occurring in the untapped portion of the regulating windings of a specific power transformer. A field-circuit coupled model was established to analyze the transient electromagnetic processes during the fault, and the model’s effectiveness was validated by comparing its results with actual fault recording data. Additionally, the magnetic field distribution and electromagnetic force characteristics during the fault were analyzed, and discussions were carried out regarding various ratios of short-circuit turns in the regulating windings. The results indicate that even when an interturn short circuit occurs in the portion of the regulating winding that is not connected to the circuit, the current in the short-circuited turns can reach several tens of times the rated value. Additionally, the leakage magnetic field and the electromagnetic force experienced by the short-circuited ring also increase significantly. The short-circuit ratio has a significant impact on the current of the short-circuited ring, leakage magnetic field intensity, and electromagnetic force. This study contributes to a better understanding of the impact of interturn short-circuit faults in the untapped portion of the regulating windings, offering crucial technical support for fault diagnosis and prevention of power transformers.

Transformer fault diagnosis based on the improved QPSO and random forest

Abstract Although dissolved gas analysis (DGA) is an effective method for transformer fault diagnosis, problems with the quality and accuracy of DGA characterization datasets often arise in practical industrial applications and face difficulties in adjusting the parameters of fault diagnosis models. To address the above problems, this paper proposes a fault diagnosis model (MD-IQPSO-RF) based on Mahalanobis distance (MD) data cleaning and improved quantum particle swarm (IQPSO) optimization of random forest (RF) parameters. Specifically, the abnormal samples of the DGA dataset are first processed by MD to improve the quality and accuracy of the dataset. Then, the RF parameters were optimized using the IQPSO algorithm to adjust the model parameters in order to improve the diagnostic performance of the RF. Finally, the optimal parameters of RF are output, and the training data are used to train the RF algorithm to construct the MD-IQPSO-RF transformer fault diagnosis model. The experimental results show that the model achieves an average accuracy of 93.631% for fault diagnosis, which is 6.92% higher than the unoptimized RF model. Comparison with other similar methods also achieved good results, which further validated the effectiveness of the fault diagnosis model.

BiPLS-RF: A hybrid wavelength selection strategy for laser induced fluorescence spectroscopy of power transformer oil

In this paper, a method for indirect diagnosis of transformer faults based on the fluorescence spectrum and characteristic wavelength screening of transformer oil has been proposed. Specifically, a hybrid strategy (BiPLS-RF) for establishing the fluorescence spectrum feature screening of transformer oil using backward interval partial least squares (BiPLS) and random forest (RF) has been proposed. Aiming at the problem of transformer fault diagnosis, the laser induced fluorescence (LIF) spectroscopy of transformer oil in different states was first collected, and it is found that the fluorescence spectrum intensity of normal transformer oil was stronger than that of faulty transformer oil. Then the characteristic bands of the original fluorescence spectra were screened by BiPLS. It is found that when the original fluorescence spectra were divided into 15 sub-intervals, the minimum root mean squares error of cross-validation can be obtained by selecting 3 sub-intervals (including 411 wavelengths). On this basis, RF was employed to further screen the characteristic wavelengths and realized the identification of the fluorescence spectrum. It is found that in the RF model composed of 54 trees, the selected 196 characteristic wavelengths of the fluorescence spectrum can minimize the analysis error (0.56%). In addition, the selected characteristic wavelength information was fed into other common classifiers to construct a fluorescence spectrum identification model, which further proved the effectiveness of BiPLS-RF for wavelength selection for LIF spectroscopy of power transformer oil. The results show that it is feasible to use BiPLS-RF to screen the characteristic wavelength of LIF spectroscopy and apply it to transformer fault diagnosis, which provides a new solution for transformer fault diagnosis.

Twins transformer: rolling bearing fault diagnosis based on cross-attention fusion of time and frequency domain features

Abstract Current self-attention based Transformer models in the field of fault diagnosis are limited to identifying correlation information within a single sequence and are unable to capture both time and frequency domain fault characteristics of the original signal. To address these limitations, this research introduces a two-channel Transformer fault diagnosis model that integrates time and frequency domain features through a cross-attention mechanism. Initially, the original time-domain fault signal is converted to the frequency domain using the Fast Fourier Transform, followed by global and local feature extraction via a Convolutional Neural Network. Next, through the self-attention mechanism on the two-channel Transformer, separate fault features associated with long distances within each sequence are modeled and then fed into the feature fusion module of the cross-attention mechanism. During the fusion process, frequency domain features serve as the query sequence Q, and time domain features as the key-value pairs K. By calculating the attention weights between Q and K, the model excavates deeper fault features of the original signal. Besides preserving the intrinsic associative information within sequences learned via the self-attention mechanism, the Twins Transformer also models the degree of association between different sequence features using the cross-attention mechanism. Finally, the proposed model's performance was validated using four different experiments on four bearing datasets, achieving average accuracy rates of 99.67%, 98.76%, 98.47%, and 99.41%. These results confirm the model's effective extraction of time and frequency domain correlation features, demonstrating fast convergence, superior performance, and high accuracy.

Adsorption and gas sensing properties of Pdn(n=1–3) cluster-modified PtSe2 on transformer fault characterizing gases under biaxial strain and electric field

In this study, the adsorption behavior of three DGA gases (C2H2, C2H4, and CO) on monolayers of PtSe2 doped with metal clusters Pdn(n=1–3) was investigated using density-functional theory (DFT). The critical influence of the Pd clusters on the sensitivity of PtSe2 gas is revealed by in-depth analysis of the geometrical structure of the adsorption system, DCD, ELF, DOS and band gap. The results show that the detection performance of intrinsic PtSe2 monolayer is poor. However, the doping of Pdn(n=1–3) fundamentally changes the adsorption behavior of PtSe2 on C2H2 and C2H4 from an adsorptive to an exothermic reaction, and endows it with excellent adsorption, desorption, and gas sensitivity properties. In addition, the introduction of external electric field and biaxial strain can significantly change the charge transfer, band gap, and adsorption energy of each adsorption system, which suggests that the Pdn(n=1–3)-PtSe2 monolayer has excellent tunability and can be strain-modulated for gas detection.

Research on anomaly detection in oil chromatography based on convolutional neural network and grey relational analysis algorithm

Abstract The power transformer is a crucial voltage conversion node and a core device in substations. Currently, dissolved gas analysis (DGA) in oil is a practical and validated method for assessing the operational status of transformers. In this study, addressing issues such as insufficient data mining from single dissolved gas measurements, we employed a combination of Convolutional Neural Network and Grey Relational Analysis algorithm to enhance the reliability of identifying abnormal dissolved gas data in transformer oil. This approach improves the sensitivity and accuracy of transformer fault detection. The study includes data processing and result analysis, indicating that the proposed algorithm effectively identifies abnormal dissolved gas data in transformer oil.

Research on Vibration Characteristic Analysis and Fault Diagnosis Method of Oil‐Immersed Transformer Based on Multi‐Physics Coupling

The oil‐immersed transformer is a crucial power equipment in power system, which is prone to abnormality or failure under long running. In order to improve the level of operation and maintenance and ensure the safe and reliable operation of the power grid, it is necessary to diagnose the fault of the oil‐immersed transformer in time. Most of the traditional fault vibration methods are off‐line detection, and are easily disturbed by mechanical deterioration. In order to solve the above problems, this paper established a 3D electromagnetic and stress coupling simulation model of oil‐immersed transformer by finite element simulation software. The vibration characteristics of oil‐immersed transformer core winding are obtained, and at the same time, this paper obtained the vibration signals at different positions on transformer core winding and oil tank. According to the vibration signal of the transformer, the best measuring point position of the tank wall is proposed to accurately reflect the vibration characteristics of the transformer. The signal data is preprocessed by CEEMDAN decomposition method, and the signal data is classified by GWO‐BP composite classification algorithm. This paper proposed an online transformer fault diagnosis method based on vibration signal, which achieves the purpose of accurately diagnosing oil‐immersed transformer faults. The test results show that in the case of oil‐immersed transformers the accuracy of the proposed transformer fault diagnosis method is more than 93%, which provides important guiding significance for the safe and stable operation of the transformer. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

A Fault Restoration Method for DC-interconnected Distribution Systems with Renewable Resources Considering Multiple Fault Scenes

Background:: The distribution systems are gradually transformed into active systems with the large-scale integration of renewable resources, which brings greater potential to the fault recovery of the distribution networks and higher requirements for the power scheduling capability under fault conditions. Power electronic devices such as soft open points (SOPs) can break the radial constraints of the traditional distribution network, provide cross-station power support in fault scenarios, and enhance the resilience of the distribution networks. Methods:: This paper focuses on the distribution systems with multi-transformer flexible interconnection, and a multi-device coordinated fault restoration method is proposed. Firstly, the characteristics of different transformer fault scenes and the fault restoration process coordinated by the bus-bar switches and SOPs in each scene are sorted out. Secondly, an optimization model of the fault restoration for DC-interconnected distribution system is established considering the network operation and the equipment constraints in the fault restoration process. Result:: The proposed model aims at the maximization of the weighted load restoration amount, and is transformed into a second-order cone programming model to be solved quickly. Lastly, the effectiveness of the proposed method is verified in a typical flexible interconnection distribution system. Conclusion:: The simulation results prove that it can effectively improve the capability of preserving power supply for important loads after transformer failure.

A Reviewed Turn at of Methods for Determining the Type of Fault in Power Transformers Based on Dissolved Gas Analysis

Since power transformers are the most important pieces of equipment in electricity transmission and distribution systems, special attention must be paid to their maintenance in order to keep them in good condition for a long time. This paper reviews the main steps in the process of diagnosing the health of power transformer insulation, which involves the science of analysing the gases dissolved in power transformer oil for effective identification of faults. An accurate diagnosis of incipient faults is favourable to sustainable development and necessary to maintain a reliable supply of electricity. The methods presented for fault diagnosis in mineral-oil-immersed power transformers are divided into analytical and graphical methods and have been found to be simple, economical and effective. After describing the methods, both their strengths and weaknesses were identified, and over the years, the methods were complemented to provide highly accurate information, validated by field inspections. This paper focuses on practical information and applications to manage maintenance based on accurate and up-to-date data. The contents of this paper will be of particular use to engineers who manufacture, monitor and/or use high-power transformers in the energy sector, as well as to undergraduate, master’s and PhD students interested in such applications.

Fault Identification Method of Transformer Winding based on Gramian Angular Difference Field and Convolutional Neural Network

Background:: As the frequency of transformer winding faults becomes higher and higher, the frequency response analysis used to detect the winding status has attracted more and more attention. At present, there is still a lack of reliable and intelligent technologies for detecting the state of transformer windings in this field. Objective:: This paper focuses on studying a high-precision method for transformer fault diagnosis, which can be easily and effectively applied to daily life. Methods:: By changing the detection method, the traditional detection method can not distinguish the problem that the detection data are highly overlapping when identifying the same fault of the head and tail symmetric points, and the problem that the phase is too similar is changed. In order to solve the problem that the fault samples of transformer frequency response curve are scarce and the one-dimensional data cannot be read by partial deep learning method, the one-dimensional data of frequency response curve is first converted into characteristic index and then into a three-dimensional image by moving window calculation method and Gramian Angular difference field transformation. The fault classification is realized by a convolutional neural network. Results:: The accuracy of the final model for slice classification reached 100%. Conclusion:: Illustrative examples show that the method is distinguishable from different fault types. The traditional method only uses the amplitude of the frequency response curve, but this method displays the two features of the amplitude-phase together in the image. Compared with the traditional method, more features and samples are added to further improve the accuracy of the method. The accuracy of diagnosis results reached 100%, which showed the feasibility of the method.

How to mine the abnormal information of power transformers: An efficient tool for quantifying the fault characteristics via multi-vibration signals

The new power system puts forward higher requirements for the reliability of power equipment. Mining the intrinsic information of equipment operation from massive data is the key challenge to determining fault early warning. The transformer fault diagnosis through vibration signal analysis has gained increasing prominence. However, the vibration signals produced by power transformers exhibit complex nonlinear and multivariate fluctuations due to the interaction of electromagnetic and mechanical factors. Unfortunately, the traditional fault diagnosis models lack the capability to capture extensive global fault information, resulting in inadequate performance in fault identification. This paper proposes the enhanced hierarchical multivariate multiscale fuzzy entropy (EHMvMFE) as an effective method for quantifying defect information in multivariate transformer signals to overcome the challenge. Firstly, the enhanced hierarchical decomposition method and time series coarse-graining theory are introduced into multivariate fuzzy entropy to develop EHMvMFE, which can comprehensively reflect the feature information of multivariate signals from the full range of frequencies. Then, EHMvMFE is utilized to extract the features via multivariate vibration signals of the transformer. Finally, the extreme learning machine is conducted to achieve effective identification of abnormal conditions. The proposed method is applied to two categories of transformer fault cases, and the diagnostic indicators are at least higher than 99.70 % and 91.82 %, which indicates a strong advantage in comparison with other popular models. The approach could be flexibly extended to the fault diagnosis of other power equipment, which is a potentially effective tool for improving the reliability of power systems.

Transformer Fault Location Classification Using FFT Based 1D-Convolutional Neural Network Model

Transformer Fault Location Classification Using FFT Based 1D-Convolutional Neural Network Model

Transformer fault diagnosis based on adversarial generative networks and deep stacked autoencoder

Establishing a deep learning model for transformer fault diagnosis using transformer oil chromatogram data requires a large number of fault samples. The lack and imbalance of oil chromatogram data can lead to overfitting, lack of representativeness of the model, and unsatisfactory prediction results on test set data, making it difficult to accurately diagnose transformer faults. A conditional Wasserstein generative adversarial network with gradient penalty optimization (CWGAN-GP) is adopted in this paper, which based on gradient penalty optimization and expand the oil chromatography fault samples of 500 sets of transformer oil chromatography data with 5 types of faults. The proposed method is used to classify transformer faults using a deep autoencoder, and the sample quality of the neural network model proposed in this paper is compared with several other variants of generative adversarial neural network models. The research results show that after using the method proposed in this paper for sample expansion, the overall accuracy of fault diagnosis can reach 93.2 %, which is 4.98 % higher than the original imbalanced samples. Compared with other sample expansion methods, the accuracy of fault diagnosis of the algorithm in this paper is improved by 1.70 %–3.05 %.