A methodology for transformer fault diagnosis based on the feature extraction from DGA data

Issue title: 20th International Symposium on Applied Electromagnetics and Mechanics

Guest editors: Theodoros Theodoulidis, Christos Antonopoulos, Nikolaos Kantartzis, Ioannis Rekanos and Theodoros Zygiridis

Article type: Research Article

Authors: Rao, Shaowei^a | Yang, Shiyou^a; | Zou, Guoping^b

Affiliations: [a] College of Electrical Engineering, Zhejiang University, Hangzhou, China | [b] College of Mechanical and Electrical Engineering, China Jiliang University, Hangzhou, China

Correspondence: [*] Corresponding author: Shiyou Yang, Zhejiang University, College of Electrical Engineering, 310027, Hangzhou, China. E-mail: [email protected]

Abstract: A methodology to diagnose transformer faults based on dissolved gas analysis (DGA) is proposed. Since a fault is more sensitive to the ratios of gas contents, a general ratio feature extraction framework is proposed to generate a feature subset as the input of the diagnosis model. The feature subset is evaluated by the improved case-based reasoning (CBR) and optimized by using a proposed new algorithm called the k-optimal algorithm (k-OA). The comparison results between the k-OA and the genetic algorithm (GA) show that the k-OA is more efficient in solving such a combinatorial optimization problem. The obtained optimal feature subset is used to diagnose a public transformer fault dataset, and a 92.6% diagnosis accuracy is observed as compared to that of only 85.1% diagnosis accuracy by using the original features.

Keywords: Case-based reasoning, combinatorial optimization, fault diagnosis, feature extraction, transformer

DOI: 10.3233/JAE-220169

Journal: International Journal of Applied Electromagnetics and Mechanics, vol. 71, no. S1, pp. S313-S320, 2023