A Wavelet Packet Transform-Based Technique for the Discrimination of Inrush Currents from Faults in Three-Phase Transformer

Abstract

The inrush current occurs primarily when a transformer is energized. The magnitude of inrush current may be as high as ten times of transformer rated current which causes malfunction of differential protection system. Therefore, it is necessary to discriminate inrush current from faults. In this paper, a new approach of discrimination between inrush current and internal fault currents is proposed based on wavelet packet transform (WPT). The selection of optimal mother wavelet and the optimal number of level of resolution is carried out using the minimum description length (MDL) criteria. The proposed approach is tested for signal tripping using data collected from a simulink model of power three phase transformer for different cases that included inrush, internal fault currents and external fault currents. The simulated results clearly show that the proposed approach facilitates the accurate discrimination between inrush and fault currents in differential transformer protection.