Skewness and Kurtosis Analysis of High Impedance Fault Currents

Skewness can be used in HIF classification with its statistical strength to identify the asymmetry introduced to the current stream by a HIF. Kurtosis is effective in analyzing the tails of a distribution reflecting its outliners. This paper introduces the concept of relative skewness and relative kurtosis as superior computational approaches for detecting HIF induced statistical changes. Relative skewness produced the most conclusive result in identifying a distinct decay in the index of the stream, subsequent to HIFs as low as 0.5 A. The relative kurtosis shows a clear decline in the index of the data stream as soon as fault occurs with the magnitude of the disturbance a factor on how far and how fast the index decays. While skewness and kurtosis are powerful techniques, a blend of techniques is recommended to ensure a secure HIF classification performance incorporating multiple strategies for their discrimination from other long-lasting network transients.