STOCHASTIC PROPERTIES OF PARTIAL-DISCHARGE PHENOMENA

Prebreakdown, pulsating, partial-discharge (PD) phenomena that occur in dielectric media are inherently complex stochastic processes that exhibit significant statistical variability in such properties as pulse amplitude, shape, and time of occurrence. Previously published work concerned with the theory and measurement of the stochastic behavior of PD is reviewed. The types of PD phenomena considered in this review include ac and dc generated electron avalanches, pulsating positive and negative corona in gases, and PD that occur in liquid media and in the presence of solid dielectric surfaces. The basic physical mechanisms of discharge initiation, growth, and memory propagation that determine the probability distributions for pulse occurrence times and pulse amplitudes are discussed. Consideration is also given to special problems associated with the measurement and interpretation of data on the various statistical properties of PD phenomena.