Power transformers are among the most vital components of electrical power systems. Outage of a transformer can isolate a portion of the power network and disrupt the transmission of electrical energy. Moreover, the high cost of maintenance and replacement further emphasizes the necessity of reliable protection schemes. Among various transformer faults, the turn-to-turn fault in the winding is one of the most destructive types. It typically initiates between two adjacent turns and rapidly propagates to neighboring turns. If not detected in its early stages, it can cause severe and irreversible damage to the transformer. Conventional differential relays, designed to detect such faults, often lack the required sensitivity for identifying them at their early stage. Furthermore, during inrush current and current transformer (CT) saturation, differential protection is usually blocked to prevent false tripping, leaving the transformer momentarily unprotected. In this paper, a new method is proposed for the early detection of inter-turn faults even under inrush current and CT saturation conditions. The approach is based on analyzing the primary, secondary, and differential currents, as well as their rate of change. By examining the derivative of the differential current, the onset of an inter-turn short-circuit fault can be accurately detected. The proposed technique has been modeled and tested using the MATLAB/Simulink to evaluate its performance. The simulation results confirm the effectiveness and robustness of the proposed method in detecting inter-turn short-circuit faults in power transformer windings.

Keywords