摘要

This paper presents a new time-frequency analysis technique based on Hilbert transform for extracting the feature of high impedance faults (HIFs) detection in electrical distribution feeders. Based on residual current of distribution feeder, energy of intrinsic mode decomposition component and standard deviation of magnitude and phase of each IMF are considered as feature inputs to the least square support vector machine (LS-SVM) classifier, thus, HIF are distinguished from low impedance faults (LIFs) and normal operation events. The results obtained have validated the effectiveness of the proposed methodology to detect HIFs and discriminate them from LIFs and normal transient operations.

  • 出版日期2012-4