Statistical analysis of impulsive flashover voltages across solid-air interfaces

Within the pulsed power industry, a key factor determining the achievable output voltage of a HV system is the flashover voltage of the insulating parts. Statistical analysis of the breakdown voltages associated with solid-gas interfaces can reveal useful information to aid system designers in the selection of solid materials. However, it is important to test the applicability of the distribution being applied, to ensure that the fitting parameters obtained are truly representative of the distribution of the data. Normal, lognormal, 2-parameter Weibull and 3-parameter Weibull cumulative distribution functions (CDF) were plotted, to enable extraction of the specific fitting parameters associated with each distribution. The CDF for each statistical method has been plotted alongside the empirical cumulative distribution function (ECDF), found from the flashover voltages recorded during experimental testing. The distribution of best fit was then analysed by using the Kolmogorov-Smirnov (K-S) test, in order to determine the CDF that best represented the ECDF. Maximum values have been compared to the α = 0.05, K-S critical value, in order to reject or accept the null hypothesis based on how the data fits the specified distributions. This will facilitate a comparison between different statistical distributions, applied to experimental data on breakdown/flashover voltages of gas-solid interfaces, generated at a fixed pressure, and different levels of RH