Transformer protection using MLE approach

This paper proposed a new wavelet based method to identify inrush currents and to distinguish it from power system faults. The proposed algorithm extracts fault and inrush generated transient signals using DWT. Transient current signals at both sides of a transformer are firstly captured. The wavelet transform is a relatively new and powerful tool in the analysis of power transformer transient phenomena because of its ability to extract information from the transient signals simultaneously in both time and frequency domain. These currents are analyzed by wavelet transform from which the detail coefficient of each phase is derived. In this proposed method, the wavelet transform is firstly applied to decompose the differential current signals of power transformer. The MLE (Maximum Likelihood Estimate) values of each of detail coefficient is calculated and compared with threshold values to detect and identify the type of faults. The extracted information from transient signals is simultaneously in both time and frequency domains. In this study, Daubechies 6 wavelets are used to construct first level filter bank to extract the transients. To discriminate the Inrush current and external faults like Line –Line fault, Line –Ground Fault, 3-Φ Fault using wavelet algorithm. And to prevent protective devices to mal operate for inrush, and external faults.