Power Transformer Dissolved Gas Analysis Research Articles

Research on transformer fault diagnosis based on active learning with imbalanced data of dissolved gas in oil.

The power transformer is the core equipment of the power system, a sudden failure of which will seriously endanger the safety of the power system. In recent years, artificial intelligence techniques have been applied to the dissolved gas analysis evaluation of power transformers to improve the accuracy and efficiency of power transformer fault diagnosis. However, most of the artificial intelligence techniques are data-driven algorithms whose performance decreases when the data are limited or significantly imbalanced. In this paper, we propose an active learning framework for power transformer dissolved gas analysis, in which the model can be dynamically trained based on the characteristics of the data and the training process. In addition, this paper also improves the original active learning spatial search strategy and uses the product of sample feature differences instead of the original sum of differences as a measure of sample difference. Compared to passive learning algorithms, the novel approach could significantly reduce the data labeling effort while improving prediction accuracy.

Convolutional sparse filter with data and mechanism fusion: A few-shot fault diagnosis method for power transformer

In actual industrial scenarios, fault data is rare and fault labels are difficult to obtain, which brings many obstacles for fault diagnosis. For this situation, this research proposes a novel semi-supervised convolutional sparse filter with local mechanism similarity regularization (CSF-LMSR) to construct a more reliable few-shot diagnosis method for power transformer. First, a sparse filter with local mechanism similarity regularization term (SF-LMSR) is designed as a more interpretable unsupervised feature extractor with prior knowledge. This unsupervised process enables the model to extract satisfactory features from the whole dataset even with a lower proportion of labeled data. Second, SF-LMSR is combined with convolutional neural network (CNN) by a novel coupling mode of kernel replacement, which enhances the learning ability of CNN. This classification model still adopts supervised learning training, but the demand for labeled fault data is greatly reduced, which reduces the burden of labels. The effectiveness of the proposed method is verified using real power transformer dissolved gas analysis (DGA) datasets. It can be seen from experimental results that the proposed method does provide a new perspective for transformer fault diagnosis, and it is a successful attempt for the power industry few-shot diagnosis problem.

Case Study and Comparison of High Voltage Power Transformer Dissolved Gas Analysis Assessed by Different Standards

A power transformer is a vital equipment in a power system. To prevent transformer failure, it is necessary to assess the transformer condition, one of which is through the DGA test. IEC 60599 and IEEE C57.104 become guidance for transformer assessment through the DGA test. DGA is a method of determining transformer conditions and fault diagnosis by assessing transformer oil data samples. H2, CO2, C2H4, C2H2, C2H6, CH4, CO, O2, and N2 are among the dissolved gases measured to monitor transformer conditions. To determine the faults, the IRM, RRM, and DTM are used. The fault detected on 8 transformer case studies showed T1, T2, T3, PD, D1, and D2. Maintenance recommendations for 8 power transformers evaluated were presented.

A Comparison between Artificial Intelligence Method and Standard Diagnosis Methods for Power Transformer Dissolved Gas Analysis Using Two Public Databases

Oil‐filled power transformers play an important role in modern network systems. Stable power supply can be achieved by early detection of power transformer faults and continuous monitoring of equipment. In recent years, dissolved gas analysis (DGA) has been widely used to diagnose faults in power transformers. Although DGA is an easier and simpler method for the fault diagnosis of transformers, different techniques usually provide different results with real‐world data. In fact, conventional diagnosis approaches for power transformers depend on human experience and available technology of human experts. Therefore, we propose using an artificial intelligence (AI) technique called multilayer perceptron (MLP) for the intelligent diagnosis of power transformer faults. In this work, the MLP model is constructed using the Keras library. The method is tested using two public databases: one based on the Electric Technology Research Association of Japan (ETRA) database and another based on the IEC TC10 database. Results indicate that high‐prediction accuracy is achieved. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.

Improved Self-Organizing Map Clustering of Power Transformer Dissolved Gas Analysis Using Inputs Pre-Processing

Ability to organize data spatially while conserving the topological relation between data features makes the Self Organizing Map (SOM) a very useful tool for analysis and visualization of high dimensional data such as a power transformer's Dissolved Gas Analysis (DGA). Past SOM application required large historical data for its training and has limited fault detection sensitivity. In this paper, the effects of input features and data normalization are studied to enhance SOM's clustering. SOM is trained using DGA results extracted from actual faulted transformers. Combination of input features and data normalization methods are tested on SOM before the best SOM is identified. Validation is conducted using several datasets i.e. the IEC Technical Committee 10 database. Compared with past SOM applications, the proposed SOM required lesser training data, improved SOM's sensitivity in incipient fault detection and has good diagnosis accuracy. The proposed SOM is also compared with other AI-based DGA interpretation method i.e. Support Vector Machine (SVM) for benchmarking.

Impact of Load Ramping on Power Transformer Dissolved Gas Analysis

Dissolved gas in oil analysis (DGA) is one of the most reliable condition monitoring techniques, which is currently used by the industry to detect incipient faults within the power transformers. While the technique is well matured since the development of various offline and online measurement techniques along with various interpretation methods, no much attention was given so far to the oil sampling time and its correlation with the transformer loading. A power transformer loading is subject to continuous daily and seasonal variations, which is expected to increase with the increased penetration level of renewable energy sources of intermittent characteristics, such as photovoltaic (PV) and wind energy into the current electricity grids. Generating unit transformers also undergoes similar loading variations to follow the demand, particularly in the new electricity market. As such, the insulation system within the power transformers is expected to exhibit operating temperature variations due to the continuous ramping up and down of the generation and load. If the oil is sampled for the DGA measurement during such ramping cycles, results will not be accurate, and a fault may be reported due to a gas evolution resulting from such temporarily loading variation. This paper is aimed at correlating the generation and load ramping with the DGA measurements through extensive experimental analyses. The results reveal a strong correlation between the sampling time and the generation/load ramping. The experimental results show the effect of load variations on the gas generation and demonstrate the vulnerabilities of misinterpretation of transformer faults resulting from temporary gas evolution. To achieve accurate DGA, transformer loading profile during oil sampling for the DGA measurement should be available. Based on the initial investigation in this paper, the more accurate DGA results can be achieved after a ramping down cycle of the load. This sampling time could be defined as an optimum oil sampling time for transformer DGA.

Power transformer dissolved gas analysis through Bayesian networks and hypothesis testing

Accurate diagnosis of power transformers is critical for the reliable and cost-effective operation of the power grid. Presently there are a range of methods and analytical models for transformer fault diagnosis based on dissolved gas analysis. However, these methods give conflicting results and they are not able to generate uncertainty information associated with the diagnostics outcome. In this situation it is not always clear which model is the most accurate. This paper presents a novel multiclass probabilistic diagnosis framework for dissolved gas analysis based on Bayesian networks and hypothesis testing. Bayesian network models embed expert knowledge, learn patterns from data and infer the uncertainty associated with the diagnostics outcome, and hypothesis testing aids in the data selection process. The effectiveness of the proposed framework is validated using the IEC TC 10 dataset and is shown to have a maximum diagnosis accuracy of 88.9%.

Artificial Intelligence Tools Aided-Decision For Power Transformer Fault Diagnosis

This paper presents an intelligent fault classification approach for power transformer dissolved gas analysis (DGA). Fault diagnosis methods by the DGA and artificial intelligence (AI) techniques are implemented to improve the interpretation accuracy for DGA of power transformers. The DGA traditional methods are utilized to choose the most appropriate gas signature. AI techniques are applied to establish classification features for faults in the transformers based on the collected gas data. The features are applied as input data to fuzzy logic, artificial neural network (ANN) and support vector machine (SVM) classifiers for faults classification. The experimental data from Tunisian Company of Electricity and Gas (STEG) is used to evaluate the performance of proposed method. The results of the various DGA methods are classified using AI techniques and the results are compared with the empirical test. In comparison to the results obtained from the AI techniques, the ratios DGA method has been shown to possess the most excellent performance in identifying the transformer fault type. The test results indicate that the SVM approach can significantly improve the diagnosis accuracies for power transformer fault classification.. General Terms Key gas, graphical representation, ratios, multi-layer perceptron, radial basis function.

Power transformer fault diagnosis based on dissolved gas analysis by support vector machine

This paper presents an intelligent fault classification approach to power transformer dissolved gas analysis (DGA). Support vector machine (SVM) is powerful for the problem with small sampling (small amounts of training data), nonlinear and high dimension (large amounts of input data). The standard IEC 60599 proposes two DGA methods which are the ratios and graphical representation. According the experimental data, for the same input data, these two methods give two different faults diagnosis results, what brings us to a problem. This paper investigates a novel extension method which consists in elaborating an input vector establishes by the combination of ratios and graphical representation to resolve this problem. SVM is applied to establish the power transformers faults classification and to choose the most appropriate gas signature between the DGA traditional methods and a novel extension method. The experimental data from Tunisian Company of Electricity and Gas (STEG) is used to illustrate the performance of proposed SVM models. Then, the multi-layer SVM classifier is trained with the training samples. Finally, the normal state and the six fault types of transformers are identified by the trained classifier. In comparison to the results obtained from the SVM, the proposed DGA method has been shown to possess superior performance in identifying the transformer fault type. The SVM approach is compared with other AI techniques (fuzzy logic, MLP and RBF neural network); the proposed method gives a good performance for transformers fault diagnosis. The test results indicate that the novel extension method and the SVM approach can significantly improve the diagnosis accuracies for power transformer fault classification.

Power Transformer Fault Classification Based on Dissolved Gas Analysis by Implementing Bootstrap and Genetic Programming

This paper presents an intelligent fault classification approach to power transformer dissolved gas analysis (DGA), dealing with highly versatile or noise-corrupted data. Bootstrap and genetic programming (GP) are implemented to improve the interpretation accuracy for DGA of power transformers. Bootstrap preprocessing is utilized to approximately equalize the sample numbers for different fault classes to improve subsequent fault classification with GP feature extraction. GP is applied to establish classification features for each class based on the collected gas data. The features extracted with GP are then used as the inputs to artificial neural network (ANN), support vector machine (SVM) and K-nearest neighbor ( KNN) classifiers for fault classification. The classification accuracies of the combined GP-ANN, GP-SVM, and GP- <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> NN classifiers are compared with the ones derived from ANN, SVM, and <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> NN classifiers, respectively. The test results indicate that the developed preprocessing approach can significantly improve the diagnosis accuracies for power transformer fault classification.