Multi-domain feature fusion and machine learning-based SFRA diagnosis of transformer winding faults using a novel fault deviation index

Creative Commons License

Keleş C., Mamiş M. S., Arkan M., Köseoğlu M., Yalçınöz Z.

Scientific Reports, cilt.1, sa.1, ss.1-19, 2026 (SCI-Expanded)

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 1 Sayı: 1
  • Basım Tarihi: 2026
  • Doi Numarası: 10.1038/s41598-026-72481-3
  • Dergi Adı: Scientific Reports
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED)
  • Sayfa Sayıları: ss.1-19
  • İnönü Üniversitesi Adresli: Evet

Özet

Sweep Frequency Response Analysis (SFRA) is one of the commonly used techniques for evaluating transformer

winding integrity. However, the interpretation of the SFRA results remains qualitative and depends

on expert evaluation. In this study, a quantitative, measurement-oriented diagnostic framework based on the

advanced normalized Multi-Domain Fault Deviation Index (MFDI) is proposed. Within this method, magnitude,

phase, impedance, and resonance-based descriptors are integrated in a unified formulation. The proposed

method combines frequency band decomposition with multi-domain feature fusion to capture both local and

global spectral deviations. This framework is experimentally validated under both intra-phase (U–N) and interphase

(V–W) fault scenarios using SFRA measurements. The results demonstrate that the proposed MFDI

captures the obvious difference between healthy and faulty states. For intra-phase fault scenarios, the proposed

method provides quantitative differentiation among the investigated fault conditions despite the high similarity

of their corresponding spectral signatures. In contrast, inter-phase faults generate more pronounced global

spectral deviations, leading to highly separable feature representations for Support Vector Machine (SVM) and

Random Forest (RF) classifiers. These findings demonstrate that the proposed approach advances SFRA from

a predominantly qualitative interpretation technique toward a quantitative, interpretable, and machine-learningcompatible

diagnostic methodology suitable for intelligent transformer condition monitoring.