Transformer Inrush Current Identification Of Empirical Mode Decomposition Method

As rapid development of electric undertaking, ultra high voltage、 large capacity and long range transmitting electricity have become very popular now and the large-sized power transformer has a more significant status in power supply system than before, so, it brings forward a big project of how to protect the power transformer more precise. In protecting of transformer differential, it is always the hotspot of researching for distinguishing excitation surge current and internal fault current, which is congenital and unavoidable. The existing method of transformer differential protection has some drawbacks, the main reason of which is that people can not know the principle of transformer internal fault mechanism exactly. Based on that, there is great important meaning in the improvement of transformer relaying protection level and reduce accident based on developing and improving of existed transformer differential protection.In this paper, a detailed introduction on causes of excitation surge current has been mentioned, after which, the paper analyzes and appraises various of distinguishing methods based on electric value and modern signal process by sorts. Based on these analyses for magnetic circuit and electrocircuit of transformer, the paper finds out the differences between excitation surge current and internal fault current which are produced by transformer and sets up a model for transformer. One new algorithm based on EMD and the improved EMD (EEMD) was presented for fast discrimination between inrush current and fault current of power transformers. The method was based on that the waveform of the inrush distorts seriously, while the fault current nearly keeps sinusoid. The complicated signal can be decomposed into a finite intrinsic mode functions (IMF) by EMD and EEMD. On the basis of self-defined IMF-led extracted by its energy characteristics, both inrush current and fault current of power transformers can be distinguished by their IMF-led characteristics.Empirical mode decomposition (EMD) provides a new method for analyzing signal from a nonlinear viewpoint. Because the mode mixing phenomenon will appear, when signal frequency is too close, we often don’t get correct results of empirical mode decomposition. We present the problem of mode mixing and give a solution involving a masking signal. The masking signal method also allows EMD to be used to separate components that are similar in frequency.This method combines the EMD to reach the effect of eliminating mode mixing phenomenon. And this method applied to signal analysis of excitation inrush current of transformer, masking signal method based on the IMF to join masking signal and at the same time, combined with the EMD to eliminate the effect of mode mixing phenomenon, to be able to separate every harmonic signal.