NIST logo

Publication Citation: Statistical Analysis of Partial Discharge Phenomena - Time of Occurrence Distributions

NIST Authors in Bold

Author(s): X. Han; Nathanael A. Heckert; James J. Filliben; Yicheng Wang;
Title: Statistical Analysis of Partial Discharge Phenomena - Time of Occurrence Distributions
Published: October 01, 1998
Abstract: This paper presents time-of-occurrence (phase) distributions of individual pulsating partial discharges (PDs) which occur in a point-dielectric gap in air for ac voltage conditions. It is determined that the pulse phase distribution is adequately modeled by a normal (Gaussian) distribution. Based on such best-fit normal distributions, estimated mean values of the phase and the standard deviations are given for the individual PD pulse distributions. For the alternating voltage case, the spread of time-of-occurrence distributions of succeeding pulses are universally broader than that of preceding pulses. PD pulse separations are non-uniform in phase. Further, if one assumes the usual sinusoidal waveform, then the voltage separation of PD pulses are also seen to be non-uniform.
Proceedings: Proc., IEEE Intl. Symp. on Electrical Insulation and Dielectric Phenomena (CEIDP)
Volume: II
Pages: pp. 407 - 411
Location: Atlanta, GA
Dates: October 25-28, 1998
Keywords: ;dielectric insulation;normal distribution;partial discharge;statistical analysis;
Research Areas: Electrical Metrology
PDF version: PDF Document Click here to retrieve PDF version of paper (3MB)