Equivalent statistics based inrush identification method for differential protection of power transformer

Abstract

Second harmonic restraint as a traditional inrush identification method is widely used in power transformer protection. However, it may fail in the case of inrush current with a high remanence in transformer and may restrain the protection relay for an unnecessarily long time in the cases such as an internal fault during current transformer saturation or an internal fault contaminated by inrush current. Therefore, this paper proposes an equivalent statistics-based inrush identification method to improve the reliability and sensitivity of transformer differential protection. Firstly, the waveform of differential current is intercepted and reshaped so that it can be abstractly seen as a probability density function. Then, the equivalent statistics are used to quantify the features of the function’s waveform. On this basis, a group of criteria is presented to discriminate disturbances from internal fault. A large number of simulation studies and experimental tests have been conducted to verify the effectiveness of the proposed method, and the results indicate that it outperforms the second harmonic restraint method.