A Novel Approach to Assess Power Transformer Winding Conditions Using Regression Analysis and Frequency Response Measurements

A frequency response analysis (FRA) is a well-known technique for evaluating the mechanical stability of power transformer’s active part components. FRA’s measuring practices have been industrialised and are codified in IEEE IEC standards. However, because there no valid coding standard, interpretation FRA data still far from being widely acknowledged authoritative approach. This study proposes an innovative fault segmentation localisation based on data. The algorithm regression to estimate repeatability relationship between fingerprint latest measured Initially, discretised into three regions narrow location fault; model current then evaluated. As benchmark, two statistical indicators employed benchmark against proposed method. Finally, scheme identifies characterises various transformer conditions, such as healthy windings, axial radial winding deformations, core deformation electrical faults. database used this consists measurements 70 mineral-oil-immersed transformers different designs, ratings manufacturers that were physically inspected faults comparable regions. results achieved corroborate efficacy recognition (RAFRA) diagnosis using FRA. Further recommendations made address reproducibility concerns induced by multiple testing conditions.