A Novel Fault Location Method for Distribution Networks Based on Transient Energy Relative Entropy of FDM Transform

Abstract

Aimed at the problem that the transient zero sequence current of each point in distribution networks is characterized by nonlinearity and non-stationarity when a single phase to ground (SPG) fault occurs, a fault diagnosis methodology based on Fourier decomposition method (FDM) and multi-scale transient energy relative entropy is proposed. First, the zero sequence current after a SPG fault is expanded based on the generalized Fourier expansion method, and a series of Fourier intrinsic band functions (FIBFs) which are adaptive, local, orthogonal and uncorrelated are obtained. Subsequently, the Fourier-Hilbert spectrum (FHS) based on FIBFs is utilized to define the instantaneous energy (IE) density of the fault. After that, the difference between the signals on both sides of various sections is represented by energy relative entropy. Through calculation and comparison, the fault section can be accurately determined. This methodology is suitable for different transition resistances, fault angles and noise interference. Finally, numerical results verify the effectiveness of the proposed methodology.