Key Gases in Transformer Oil – An Analysis using Self Organizing Map (SOM) Neural Networks

This work examines the application of machine learning (ML) algorithms to evaluate dissolved gas analysis (DGA) data to quickly identify incipient faults in oil-immersed transformers (OITs). Transformers are pivotal equipment in the transmission and distribution of electrical power. The failure of a particular unit during service may interrupt a massive number of consumers and disrupt commercial activities in that area. Therefore, several monitoring techniques are proposed to ensure that the unit maintains an adequate level of functionality in addition to an extended useful lifespan. DGA is a technique commonly employed for monitoring the state of OITs. The understanding of DGA samples is conversely unsatisfactory from the perspective of evaluating incipient faults and relies mainly on the proficiency of test engineers. In the current work, a multi-classification model that is centered on ML algorithms is demonstrated to have a logical, precise, and perfect understanding of DGA. The proposed model is used to analyze 138 transformer oil (TO) samples that exhibited different stray gassing characteristics in various South African substations. The proposed model combines the design of four ML classifiers and enhances diagnosis accuracy and trust between the transformer manufacturer and power utility. Furthermore, case reports on transformer failure analysis using the proposed model, IEC 60599:2022, and Eskom (Specification—Ref: 240-75661431) standards are presented. In addition, a comparison analysis is conducted in this work against the conventional DGA approaches to validate the proposed model. The proposed model demonstrates the highest degree of accuracy of 87.7%, which was produced by Bagged Trees, followed by Fine KNN with 86.2%, and the third in rank is Quadratic SVM with 84.1%.

Section: Materials and Protocolsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Discernment of transformer oil stray gassing anomalies using machine learning classification techniques

Ngwenyama,

Gitau

2024

Transformer fault diagnosis method based on SMOTE and NGO-GBDT

Wang,

Chi,

Ding

et al. 2024

In order to improve the accuracy of transformer fault diagnosis and improve the influence of unbalanced samples on the low accuracy of model identification caused by insufficient model training, this paper proposes a transformer fault diagnosis method based on SMOTE and NGO-GBDT. Firstly, the Synthetic Minority Over-sampling Technique (SMOTE) was used to expand the minority samples. Secondly, the non-coding ratio method was used to construct multi-dimensional feature parameters, and the Light Gradient Boosting Machine (LightGBM) feature optimization strategy was introduced to screen the optimal feature subset. Finally, Northern Goshawk Optimization (NGO) algorithm was used to optimize the parameters of Gradient Boosting Decision Tree (GBDT), and then the transformer fault diagnosis was realized. The results show that the proposed method can reduce the misjudgment of minority samples. Compared with other integrated models, the proposed method has high fault identification accuracy, low misjudgment rate and stable performance.

Research on transformer fault diagnosis method based on ACGAN and CGWO-LSSVM

Guan,

Wu,

Yang

2024

This paper proposes a transformer fault diagnosis method based on ACGAN and CGWO-LSSVM to address the problem of misjudgment and low diagnostic accuracy caused by the small number and uneven distribution of some fault samples in transformer fault diagnosis. Firstly, generate adversarial networks through auxiliary classification conditions, The ACGAN method expands a small and imbalanced number of samples to obtain balanced and expanded data; Secondly, the non coding ratio method is used to construct the characteristics of dissolved gases in oil, and kernel principal component analysis is used, KPCA method for feature fusion; Finally, using the improved cubic gray wolf optimization algorithm, CGWO for least square support vector machines, optimize the parameters of the LSSVM model and construct a transformer fault diagnosis model. The results show that the proposed method has a low false alarm rate and a diagnostic accuracy of 97.66%, compared to IGOA-LSSVM the IChOA-LSSVM and PSO-LSSVM methods improved accuracy by 0.12, 1.76, and 2.58%, respectively. This method has been proven to solve the problems of misjudgment and low diagnostic accuracy caused by small sample sizes and uneven distribution. It is suitable for multi classification fault diagnosis of transformer imbalanced datasets and is superior to other methods.