Fault Line Selection Method Based on Improved HHT Small Current Ground

In the analysis of nonlinear, non-stationary signal is better than the traditional wavelet signal analysis method, commonly used in this kind of signal singular value point detection. Empirical mode decomposition (EMD) is the core of the analysis method, but EMD in the original HHT method exists end effect in decomposition process, the decomposition of intrinsic mode functions IMF serious distort, greatly affected the accuracy of the fault line selection, by extension on the endpoint can inhibit the effect of the end point. In this paper, we compare the different endpoint extension methods, and use the method of endpoint extension based on adaptive waveform matching to improve the algorithm. MATLAB simulation results show that the proposed method is effective and accurate in the fault location of small current grounding system.