Gas Content In Transformer Oil Research Articles

Thermodynamic Computational Method of Thermal Decomposition Gas Content in Transformer Oil Considering Temperature and Competitive Reaction

Thermodynamic Computational Method of Thermal Decomposition Gas Content in Transformer Oil Considering Temperature and Competitive Reaction

Data Restoration of dissolved gas content in transformer oil based on the CS-SVR model

Accurate monitoring of the dissolved gas content in transformer oil is crucial for transformers’ safe and stable operation. The early identification for detecting potential power transformer failures is necessary for the stability of an electrical grid. Dissolved gas analysis is an essential technology in transformers diagnosing insulation faults. Missing dissolved gas data can directly impact the reliability of monitoring results of a transformer. This study presents a data plug-in model based on support vector regression (SVR) to restore missing dissolved gas data. To further improve the accuracy of data restoration, the cuckoo search algorithm (CS) is used for optimizing SVR parameters. By verifying H2 and C2H4, the CS-SVR model demonstrates superiority over other plug-in procedures in repairing dissolved gas data.

GRU-AGCN model for the content prediction of gases in power transformer oil

Aiming at high accuracy of dissolved gas content prediction in transformer oil, a novel method based on gated recurrent unit and adaptive graph convolution network (GRU-AGCN) is proposed. For gated recurrent unit (GRU) can selectively choose the feature of time series, it is used to extract time series information of the gas content. Correlation among gases are extracted to improve the accuracy. The original adjacency matrix of the model is constructed according to the grey relational analysis (GRA), and the dynamic relation information between gases is extracted by adaptive graph convolution network (AGCN). The experimental result shows that the GRU-AGCN model can efficiently extract the temporal features and perceive the dynamic relationship of gases. The predictions error of the proposed method is lower than that of RNN, LSTM network and GRU network. The proposed method provides a reliable and accurate result for the prediction of dissolved gas content in transformer oil.

Prediction for Dissolved Gas in Power Transformer Oil Based On TCN and GCN

Dissolved gas analysis (DGA) is an important method to identify internal faults of transformers. A novel predictive method for dissolved gas content in transformer oil based on temporal convolutional network (TCN) and graph convolutional network (GCN) is proposed in this article. First, a TCN based on dilated causal convolution algorithm is designed to extract feature information from both current data and historical data of each gas content. Since the correlation coefficients among different gases may affect the accuracy of the prediction, a topological structure diagram is constructed to describe the relationship among different gases. Then, a GCN is designed to predict the gas content in the desired time horizon by using the information obtained from the topological structure diagram. The elements in the adjacency matrix of GCN are replaced by Pearson correlation coefficients to improve the accuracy of the prediction. The test results show that the proposed method in this article achieves high accuracy for the prediction of dissolved gas content in transformer oil. Compared with back propagation neural network, long short-term memory network and TCN, the mean absolute error of the predicted result by the proposed method is reduced by 12.23, 11.18, and 3.81%, respectively.

A Prediction Model for Time Series of Dissolved Gas Content in Transformer Oil Based on LSTM

By combining with dissolved gas analysis, time series prediction of dissolved gas content in oil provides a basis for transformer fault diagnose and early warning. In the view of that, a prediction model based on long short time memory (LSTM) network for time series of dissolved gas content in oil is proposed, which takes advantage of LSTM network’s ability to deal with long-sequence prediction problems. Five characteristic gas concentrations are used as input to the model, and the hyper parameters of the model is optimized by Bayesian optimization algorithm to further improve prediction accuracy, then a LSTM prediction model is constructed. By case study, it is verified that the proposed model can precisely predict time series of dissolved gas content. Compared with gray model, BP neural network and support vector machine, the proposed model has higher prediction accuracy and can better track the trend of time series of dissolved gas content in oil.

An Interval Forecasting Model Based on Phase Space Reconstruction and Weighted Least Squares Support Vector Machine for Time Series of Dissolved Gas Content in Transformer Oil

Transformer state forecasting and fault forecasting are important for the stable operation of power equipment and the normal operation of power systems. Forecasting of the dissolved gas content in oil is widely conducted for transformer faults, but its accuracy is affected by data scale and data characteristics. Based on phase space reconstruction (PSR) and weighted least squares support vector machine (WLSSVM), a forecasting model of time series of dissolved gas content in transformer oil is proposed in this paper. The phase spaces of time series of the dissolved gas content sequence are reconstructed by chaos theory, and the delay time and dimension are obtained by the C-C method. The WLSSVM model is used to forecast time series of dissolved gas content, the chemical reaction optimization (CRO) algorithm is used to optimize training parameters, the bootstrap method is used to build forecasting intervals. Finally, the accuracy and generalization ability of the forecasting model are verified by the analysis of actual case and the comparison of different models.

Anomaly detection for condition monitoring data using auxiliary feature vector and density‐based clustering

High-quality condition monitoring data can provide vital information on power equipment condition assessment and fault diagnosis. However, data quality is difficult to guarantee because valid and invalid anomalies, and different normal data patterns inevitably occur in datasets. This study presents a method for anomaly detection based on auxiliary feature vector and density-based spatial clustering of applications with noise (DBSCAN). The auxiliary feature vectors of each condition variable are constructed for clustering to recognise normal data patterns and different types of anomalies. Furthermore, a heuristic method based on the ‘number of clusters–Eps’ curve is proposed for the parameter selection of DBSCAN in an unsupervised setting. Different application examples are implemented on data of dissolved gas content in transformer oil. Compared with state-of-the-art anomaly detection techniques, the proposed method shows an ability to identify and distinguish normal data patterns and valid and invalid anomalies accurately, provided that the condition monitoring data satisfy the assumption of stationarity.

Outlier detection for on-line monitoring data of transformer based on wavelet transform and weighted LOF

Power transformers are one of the core equipment in the power grid, so it is of great significance to guarantee transformers’ normal operation. By analyzing the dissolved gas content in transformer oil, we can monitor the operation of the power transformer. However, usually there are outliers in the data generated using the on-line monitoring system. In this paper, we propose a new outlier detection method based on wavelet transform and local outlier factor (LOF) algorithm. Using wavelet transform, we get the high dimensional representation of the original data in frequency domain, and by adding the weighted LOF (WLOF), we can identify outliers in high dimensional data set. Furthermore, we use the sliding window method to improve the efficiency of the algorithm, and achieve transformer oil on-line outlier detection efficiently. The experimental results on transformer data from several power transformers indicate that this algorithm can identify the outliers that exceed the threshold value, as well as the oscillations due to fluctuations in gas content. This can help achieve initial diagnosis of transformer oil on-line monitoring system rapidly.

Proposed three ratios technique for the interpretation of mineral oil transformers based dissolved gas analysis

Dissolved gas analysis (DGA) of oil is one of the major diagnostic tools that are used to detect the incipient faults of oil-immersed transformers through the correlation between the dissolved gases content in transformer oil and a particular malfunction. There are several techniques of DGA to interpret the incipient transformer faults such as Doernenburg ratio, Rogers's ratio, IEC ratio and Duval triangle techniques. This study presents a new proposed technique for transformer fault diagnosis using a new three gas ratios concentration of DGA to overcome the conflict that takes place in the traditional interpretation techniques. The accuracy of the proposed diagnosis technique is verified using practical DGA data obtained from 688 samples that have been collected from different transformers of different rating and different lifespan reported by Egyptian Electricity Holding Company, IEC TC10 and related databases surveyed from actual incident cases. The comparison with traditional DGA techniques showed that the proposed DGA ratio technique has good diagnosis accuracy. It is found that the suggested three ratios technique method has 99.86% accuracy as compared with 85.67% for Duval, 75.08% for Doernenburg, 47.34% for IEC and 39% for Rogers when testing 688 cases.

Fiber Bragg Gratings Hydrogen Sensor for Monitoring the Degradation of Transformer Oil

In this paper, two fiber Bragg gratings (FBGs) sensors, which are based on different palladium (Pd) coatings ratios, was proposed to detect the dissolved hydrogen gas content in transformer oil. The FBG samples were prepared using a physical vapor deposition technique. Two samples were prepared with different Pd coating compositions. The first sample was coated with 100% Pd (Pd100), and the second sample was coated with mixture of Pd chromium (Pd58Cr42) that has a ratio of 58:42. Pd100 sensor showed faster response to dissolved hydrogen than the Pd58Cr42 sensor. In addition, the Pd100 sensor exerted higher strain at lower dissolved hydrogen content to the FBG compared with the Pd58Cr42 sensor. The Pd100 sensor has higher sensitivity, and the response range is limited to lower hydrogen content. This paper has showed the potential of using alloy like chromium to the Pd metal to enhance the dissolved hydrogen detection range.

Application of Extreme Learning Machine in Fault Classification of Power Transformer

Reliability of power system is very essential for every nation to generate and transmit power without interruption. Power transformer is one of the most significant electrical apparatus and hence it must be kept in good health. Identification and classification of faults in power transformer is a major research area. Conventional method of fault classification in transformer uses gas concentrations data and interprets them using international standards. These standards are not able to classify the faults correctly under certain conditions. To overcome this limitation, several soft computing tools namely artificial neural network (ANN), Support Vector Machine (SVM) etc. are used to automate the process of classification of faults in transformers. However, there is a scope exists to improve the classification accuracy. Hence, this research work focuses to design Extreme Learning Machine (ELM) method for classifying fault very accurately using enthalpy of dissolved gas content in transformer oil as an input feature. The ELM method is tested with two databases: one based on IEC TC10 database (DB1) and the other one based on data collected from utilities in India (DB2). The application of ELM to Power Transformer fault classification based on enthalpy as input feature outperforms over the conventional classification based on gas concentration as input feature.

Particle swarm optimization-least squares support vector regression based forecasting model on dissolved gases in oil-filled power transformers

This paper presents a forecasting model based upon least squares support vector machine (LS-SVM) regression and particle swarm optimization (PSO) algorithm on dissolved gases in oil-filled power transformers. First, the LS-SVM regression model, with radial basis function (RBF) kernel, is established to facilitate the forecasting model. Then a global optimizer, PSO is employed to optimize the hyper-parameters needed in LS-SVM regression. Afterward, a procedure is put forward to serve as an effective tool for forecasting of gas contents in transformer oil. The application of the proposed model on actual transformer gas data has given promising results. Moreover, four other forecasting models, derived from back propagation neural network (BPNN), radial basis function neural network (RBFNN), generalized regression neural network (GRNN) and support vector regression (SVR), are selected for comparisons. The experimental results further demonstrate that the proposed model achieves better forecasting performance than its counterparts under the circumstances of limited samples.

Improved BP Neural Network for Transformer Fault Diagnosis

The back propagation (BP)-based artificial neural nets (ANN) can identify complicated relationships among dissolved gas contents in transformer oil and corresponding fault types, using the highly nonlinear mapping nature of the neural nets. An efficient BP-ALM (BP with Adaptive Learning Rate and Momentum coefficient) algorithm is proposed to reduce the training time and avoid being trapped into local minima, where the learning rate and the momentum coefficient are altered at iterations. We developed a system of transformer fault diagnosis based on Dissolved Gases Analysis (DGA) with a BP-ALM algorithm. Training patterns were selected from the results of a Refined Three-Ratio method (RTR). Test results show that the system has a better ability of quick learning and global convergence than other methods and a superior performance in fault diagnosis compared to convectional BP-based neural networks and RTR.

Evolving neural nets for fault diagnosis of power transformers

This paper proposes evolving neural nets (ENNs) for fault diagnosis of power transformers. Based on the proposed evolutionary algorithm, the ENNs automatically tune the network parameters (connection weights and bias terms) of the neural nets to achieve the best model. The ENNs can identify complicated relationships among dissolved gas contents in transformer oil and corresponding fault types, using the global search capabilities of the evolutionary algorithm and the highly nonlinear mapping nature of the neural nets. The proposed ENNs have been tested on the Taipower Company diagnostic records and compared with the fuzzy diagnosis system, artificial neural networks, and the conventional method. The test results confirm that the proposed ENNs are much more diagnostically accurate and require less learning time than the existing approaches.

Fault identification of power transformers using genetic-based wavelet networks

The paper presents genetic-based wavelet networks (GWNs) for fault identification of power transformers. GWNs are three-layer structures, which contain wavelet, weighting and summing layers. Using a genetic-algorithm (GA) based optimisation process, the GWNs automatically tune the network parameters, translation and dilation in the wavelet nodes and the weighting values in the weighting nodes. The GWNs, with global search abilities of the GA and the multiresolution and localisation natures of the wavelets, can identify the complicated relations of dissolved gas contents in transformer oil to corresponding fault types. The proposed GWNs have been tested on the Taipower Company diagnostic records and compared with the fuzzy diagnosis system, artificial neural networks as well as the conventional method. The experimental results reveal that the GWNs have remarkable diagnostic accuracy and require far less construction time than conventional methods.

Evolving wavelet networks for power transformer condition monitoring

This paper proposes a novel model for power transformer condition monitoring using evolving wavelet networks (EWNs). The EWNs are three-layer structures, which contain wavelet, weighting and summing layers. The EWNs automatically adjust the network parameters, translation, and dilation in the wavelet nodes and the weighting values in the weighting nodes, through an evolutionary based optimization process. Global search abilities of the evolutionary algorithm as well as the multiresolution and localization natures of the wavelets enable the EWNs to identify the complicated, numerical-knowledge relations of dissolved gas contents in transformer oil to corresponding fault types. The proposed EWNs have been tested on the Taipower Company diagnostic records and compared with the fuzzy diagnosis system, artificial neural networks as well as the conventional method. The test results reveal that the EWNs possess far superior diagnosis accuracy and require less constructing time than the existing methods.