IEEE 930 : 2004

Introduction
1 Scope
2 References
3 Steps required for analysis of breakdown data
   3.1 Data acquisition
   3.2 Characterizing data using a probability function
   3.3 Hypothesis testing
4 Probability distributions for failure data
   4.1 The Weibull distribution
   4.2 The Gumbel distribution
   4.3 The lognormal distribution
   4.4 Mixed distributions
   4.5 Other terminology
5 Testing the adequacy of a distribution
   5.1 Weibull probability data
   5.2 Use of probability paper for the three-parameter Weibull
        distribution
   5.3 The shape of a distribution plotted on Weibull probability
        paper
   5.4 A simple technique for testing the adequacy of the Weibull
        distribution
6 Graphical estimates of Weibull parameters
7 Computational techniques for Weibull parameter estimation
   7.1 Larger data sets
   7.2 Smaller data sets
8 Estimation of Weibull percentiles
9 Estimation of confidence intervals for the Weibull function
   9.1 Graphical procedure for complete and censored data
   9.2 Plotting confidence limits
10 Estimation of the parameter and their confidence limits of the
   log-normal function
   10.1 Estimation of lognormal parameters
   10.2 Estimation of confidence intervals of log-normal parameters
11 Comparison tests
   11.1 Simplified method to compare percentiles of Weibull
        distributions
12 Estimating Weibull parameters for a system using data
   from specimens
Annex A (informative) Least squares regression
Annex B (informative) Bibliography
Annex C (informative) List of participants

Describes statistical methods to analyze times to breakdown and breakdown voltage data obtained from electrical testing of solid insulating materials, for purposes including characterization of the system, comparison with another insulator system, and prediction of the probability of breakdown at given times or voltages.