Fault Classification on Power Transmission Lines Using DC Component Extraction

abstract–This paper aims to present an innovative automated approach for classifying diverse faults on electrical power transmission lines. To accomplish this goal, we employ the matrix pencil method, a model-based parameter estimation algorithm that aids in extracting DC components from three-phase currents observed at one end of the line. These extracted DC features are then utilized by the k-nearest neighbors algorithm with the Hamming distance metric for performing fault classification tasks effectively. The analysis of simulated data demonstrates that the proposed methodology can accurately detect and classify various types of faults occurring on power transmission lines.