An EKF?SVM machine learning?based approach for fault detection and classification in three?phase power transformers

In this paper, a hybrid approach for effective diagnosis of power transformers is proposed. the proposed method, extended Kalman filter used estimation three-phase currents in primary windings transformer. Three residual signals are defined as differences between measured and estimated currents. When transformer healthy, EKF perfectly estimates hence, nearly zero. However, when faulty, cannot suitably estimate due to large model mismatch resulting from internal faults. Consequently, generated, which key signatures discriminating faults energisation conditions. Besides, method uses entries covariance matrix error locate classify For these purposes, support vector machine classifiers used. The effectiveness demonstrated by number simulation test cases obtained using PSCAD/EMTDC software. Also, hardware-in-the-loop experiments conducted dSPACE1104 development platform real-time feasibility authenticated.