A Method for Identifying External Short-Circuit Faults in Power Transformers Based on Support Vector Machines

• DOI: 10.3390/electronics13091716
• Publication Date: 2024-04-29T10:13:22Z
• Authors: Hao Du; Linglong Cai; Zhiqin Ma; Zhangquan Rao; Xiang Shu; Shuo Jiang; Zhongxiang Li; Xianqiang Li
• Keywords: 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology

Being a vital component of electrical power systems, transformers significantly influence the system stability and reliability supplies. Damage to may lead significant economic losses. The efficient identification transformer faults holds paramount importance for security grids. existing methods identifying include oil chromatography analysis, temperature assessment, frequency response vibration characteristic examination, leakage magnetic field analysis. These suffer from limitations such as limited sensitivity, complexity in operation, high demand specialized skills. In this paper, we propose method identify external short-circuit based on fault recording data currents. It involves analyzing current signals various windings during faults, extracting appropriate features, utilizing classification algorithm support vector machine (SVM) determine types locations. different kernel functions accuracy SVM is discussed. results indicate that can proficiently type location transformers, achieving an rate 98.3%.