A New Wavelet Transform Approach to Discriminate Magnetizing Inrush Current and Fault Current

A new approach for transformer protection based on wavelet transformation is proposed in this paper. In the proposed approach, transient current signals at both sides of a transformer are firstly captured and then analyzed by wavelet transform, from which the modular maxima (MM) are derived and subsequently used to discriminate between magnetizing inrush current and fault currents. Various types of internal faults and magnetizing inrush are modelled using real-time digital simulator (RTDS). Extensive RTDS simulations were carried out to cover various transformer operational conditions such as magnetizing inrush with various closing angles of circuit breaker, various fault types on either side of transformer, fault currents under CTs saturation during server fault conditions. The sampled current signals of both sides of a transformer are then analyzed by wavelet transform. A series of obvious characteristics are observed by decomposing the transient current signals into scale 5 (j=5) using Daubechies4 wavelet function. The multiplication of the MM on either side of transformer is then utilized to distinguish between inrush current, internal fault current and external fault current. Extensive simulation results proved that the new algorithm is simple, reliable, accurate, immune to CT's saturation, and is able to perform correctly for the worst cases where an internal fault occurs when a transformer is energized without load