Dariush Keihan Asl

Diagnosis of internal faults in power transformer windings is crucial for improving system reliability and reducing operational costs in modern power networks. Among these faults, inter-turn short-circuit (ITSC) faults are particularly critical, as they can cause severe equipment damage, reduce transformer lifespan, and widespread outages. Hence, this paper proposes an accurate, efficient, and non-destructive method for detecting and locating ITSC faults using time-domain reflectometry (TDR) and traveling wave analysis. The method operates offline using widely accessible and reasonably priced equipment like a waveform generator and a digital oscilloscope. By applying a step voltage waveform with a restricted rising edge within an appropriate frequency bandwidth and analyzing the reflected waves in the time domain, fault location is precisely identified using a simple and effective model. Experimental validations on a real winding under controlled laboratory conditions show high accuracy, with an average localization error below 1%. The method reliably detects and locates ITSC faults involving as little as 1% of total turns and high fault impedances (up to 100 Ω) at the mid-point of the winding, demonstrating its high sensitivity and robustness.