Transformer Winding Condition Assessment Using Feedforward Artificial Neural Network and Frequency Response Measurements

Tahir, Mehran; Tenbohlen, Stefan

doi:10.3390/en14113227

Cited by 22 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The adaptive frequency division algorithm is employed, which successfully identifies low-, medium-, and high-frequency regions in open-circuit FRA measurement depending upon transformer size, rating, winding type, and connection scheme. The details of the adaptive frequency division algorithm are described in the author's previous work [26]. It is important to note that the presented frequency division structure is applicable to open-circuit FRA measurements.…”

Section: Feature Generationmentioning

confidence: 99%

Transformer Winding Fault Classification and Condition Assessment Based on Random Forest Using FRA

Tahir¹,

Tenbohlen

2023

Energies

Self Cite

View full text Add to dashboard Cite

At present, the condition assessment of transformer winding based on frequency response analysis (FRA) measurements demands skilled personnel. Despite many research efforts in the last decade, there is still no definitive methodology for the interpretation and condition assessment of transformer winding based on FRA results, and this is a major challenge for the industrial application of the FRA method. To overcome this challenge, this paper proposes a transformer condition assessment (TCA) algorithm, which is based on numerical indices, and a supervised machine learning technique to develop a method for the automatic interpretation of FRA results. For this purpose, random forest (RF) classifiers were developed for the first time to identify the condition of transformer winding and classify different faults in the transformer windings. Mainly, six common states of the transformer were classified in this research, i.e., healthy transformer, healthy transformer with saturated core, mechanically damaged winding, short-circuited winding, open-circuited winding, and repeatability issues. In this research, the data from 139 FRA measurements performed in more than 80 power transformers were used. The database belongs to the transformers having different ratings, sizes, designs, and manufacturers. The results reveal that the proposed TCA algorithm can effectively assess the transformer winding condition with up to 93% accuracy without much human intervention.

show abstract

Section: Feature Generationmentioning

confidence: 99%

Transformer Winding Fault Classification and Condition Assessment Based on Random Forest Using FRA

Tahir¹,

Tenbohlen

2023

Energies

Self Cite

View full text Add to dashboard Cite

show abstract

“…The above method only reflects the health condition of the transformer and cannot identify specific fault types, therefore the proposed method cannot be applied for accurate fault identification. An intelligent monitoring and classification algorithm based on Frequency Response Analysis (FRA) is proposed in the literature [17] for detecting transformer winding faults. The model is able to accurately identify various winding states in transformers.…”

Section: Introductionmentioning

confidence: 99%

“…Synthesizing the above research methods in the literature [8]- [17], there are difficulties in solving multiclassification problems with SVM, while ANN requires a large amount of sample data, the model validity is not strong, and the stability and robustness of fuzzy theory is poor. Compared with other neural network architectures, SCN not only has good generalization characteristics of stochastic learning, but also the number of nodes in its implicit layer is generated gradually based on the supervision mechanism, which is a good solution to the problem that the number of nodes in the implicit layer of RVFL model is difficult to determine, and also has more extensive applications in the field of fault diagnosis.…”

Section: Introductionmentioning

confidence: 99%

Fault Diagnosis Method for Power Transformers Based on Improved Golden Jackal Optimization Algorithm and Random Configuration Network

Shi

et al. 2023

IEEE Access

View full text Add to dashboard Cite

The problem of the low accuracy of Dissolved Gas Analysis (DGA) in diagnosing transformer faults is addressed by proposing an Improved Golden Jackal Optimization (IGJO) based Stochastic Configuration Network (SCN) method. The method of transformer fault diagnosis based on IGJO optimized SCN is proposed. Firstly, Kernel Principal Component Analysis (KPCA) is used to reduce the dimensionality of the gas data and extract the effective feature quantities. Secondly, the L2 parametric penalty term is introduced into the SCN to improve the generalisation ability of SCN in practical applications. The elite backward learning and golden sine algorithms are incorporated into the golden jackal algorithm, and the IGJO performance is tested using 13 typical test functions, demonstrating that the IGJO has greater stability and merit-seeking capability. The penalty term coefficient C of the SCN is optimised using the IGJO to develop a transformer fault diagnosis model with an Improved Golden Jackal algorithm optimised Random Configuration Network (IGJO-SCN). Finally, the feature quantities extracted by KPCA are used as the input set of the model and the different transformer fault diagnosis models are simulated and validated. The results show that the IGJO-SCN has higher transformer fault diagnosis accuracy compared to other models.

show abstract

“…In recent studies of FRA interpretation, there has been an increase in the use of machine learning algorithms to help in developing objective interpretations and to reduce dependency on expert analyses [5][6][7]. The main challenge now is building a sufficient database to train and test these algorithms.…”

Section: Introductionmentioning

confidence: 99%

A Machine-Learning Approach to Identify the Influence of Temperature on FRA Measurements

et al. 2021

View full text Add to dashboard Cite

Frequency response analysis (FRA) is a powerful and widely used tool for condition assessment in power transformers. However, interpretation schemes are still challenging. Studies show that FRA data can be influenced by parameters other than winding deformation, including temperature. In this study, a machine-learning approach with temperature as an input attribute was used to objectively identify faults in FRA traces. To the best knowledge of the authors, this has not been reported in the literature. A single-phase transformer model was specifically designed and fabricated for use as a test object for the study. The model is unique in that it allows the non-destructive interchange of healthy and distorted winding sections and, hence, reproducible and repeatable FRA measurements. FRA measurements taken at temperatures ranging from −40 °C to 40 °C were used first to describe the impact of temperature on FRA traces and then to test the ability of the machine learning algorithms to discriminate between fault conditions and temperature variation. The results show that when temperature is not considered in the training dataset, the algorithm may misclassify healthy measurements, taken at different temperatures, as mechanical or electrical faults. However, once the influence of temperature was considered in the training set, the performance of the classifier as studied was restored. The results indicate the feasibility of using the proposed approach to prevent misclassification based on temperature changes.

show abstract

Transformer Winding Condition Assessment Using Feedforward Artificial Neural Network and Frequency Response Measurements

Cited by 22 publications

References 18 publications

Transformer Winding Fault Classification and Condition Assessment Based on Random Forest Using FRA

Transformer Winding Fault Classification and Condition Assessment Based on Random Forest Using FRA

Fault Diagnosis Method for Power Transformers Based on Improved Golden Jackal Optimization Algorithm and Random Configuration Network

A Machine-Learning Approach to Identify the Influence of Temperature on FRA Measurements

Contact Info

Product

Resources

About