Development of a data compression index for discrimination between transformer internal faults and inrush currents

This paper proposes an algorithm for transformer differential protection. The paper presents the development of an index based on discrete wavelet transform to discriminate between internal faults and inrush currents. The index is retained energy in percentage. The proposed technique consists of decomposition of differential current signals up to a specific level and compression by level thresholding. Various simulations and analysis have been performed to determine the best compression ratio for distinguishing between transformer inrush and fault currents. The performance of the index is demonstrated by simulation of different faults and switching conditions on a modeled power system. Obtained results show that the proposed scheme is robust and reliable.