Improved Genetic Algorithm and XGBoost Classifier for Power Transformer Fault Diagnosis

Abstract

Power transformer is an essential component for the stable and reliable operation of electrical power grid. The traditional transformer fault diagnostic methods based on dissolved gas analysis are limited due to the low accuracy of fault identification. In this study, an effective transformer fault diagnosis system is proposed to improve identification accuracy. The proposed approach combines an improved genetic algorithm (IGA) with the XGBoost to form a hybrid diagnosis network. The combination of the improved genetic algorithm and the XGBoost (IGA-XGBoost) forms the basic unit of the proposed method, which decomposes and reconstructs the transformer fault recognition problem into several minor problems IGA-XGBoosts can solve. The results of simulation experiments show that the IGA performs excellently in the combined optimization of input feature selection and the XGBoost parameter, and the proposed method can accurately identify the transformer fault types with an average accuracy of 99.2%. Compared to IEC ratios, dual triangle, support vector machine and common vector approach the diagnostic accuracy of the proposed method is improved by 30.2, 47.2, 11.2, and 3.6%, respectively. The proposed method can be a potential solution to identify the transformer fault types.

Transformer faults identification via fuzzy logic approach

“…It is observed that fuzzy logic approach is highly reliable over other methods, due to its operational speed of faults identification in modern power system transformers within 5.3 ms, with an accuracy ranging from 96%-99% depending on the sensitivity the fault. Differential protection 80%-90% Fast 50 ms-100 ms [21] Sequential kalman filter 85%-90% Moderate 500 ms-1 sec [17] Electrical transient analyser 70%-80% Slow 5 secs-10 secs [3] Dissolved gas analysis 60%-80% Slow 15 mins-30 mins [18] Buchholz relay 70%-80% Moderate 100 ms-500 ms [22] GA-XG boost classifier 90%-95% Fast 50 ms-100 ms [Proposed]…”

“…However, the challenge in terms of computational complexity, resource demands, system overhead, and increased cost makes the approach questionable. Wu et al [22], used GI-XGBoost in transformer fault identification, the approach optimizes parameters for the robust search, improving fault diagnosis accuracy. However, it is challenging in interpreting1 features that leds to transformers' fault.…”

MI-YOLO: more information based YOLO for insulator defect detection

“…However, this process is intricate and requires further accuracy improvement. On the other hand, ML-based methods utilize similar features but focus on classifiers such as XGBoost, 18 , 19 random forest, 20 , 21 and support vector machine 22 , 23 . Although the accuracy of ML-based methods has improved substantially compared with IP-based methods, further accuracy improvement is necessary considering the development of more advanced learning methods.…”

Research on the Transformer Intelligent Operation and Maintenance System Based on the Graph Neural Network

“…He et al (2021) proposed the use of multiple inputs and multiple output back-propagation neural networks with multiple hidden layers to extract a variety of electrical and non-electrical features for state monitoring and fault diagnosis. Wu et al (2021) proposed an improved genetic algorithm (IGA) combined with XGBoost to form a hybrid diagnosis network for in-depth exploration of transformer fault data. Kari et al (2016) established a transformer evaluation system based on the D-S theory by using dissolved gas and electrical test data.…”