论文信息 - Statistical Treatment of Insulation Failure Data With The Application of Simplified Regression Method in The High-Voltage Laboratory

Statistical Treatment of Insulation Failure Data With The Application of Simplified Regression Method in The High-Voltage Laboratory

Abstract In this article, we propose a new and simplified regression method for estimating the parameters of the Weibull distribution for high-voltage insulation failure data. This method is simpler to apply than the widely used maximum likelihood method and has some interesting properties suitable for small sample size data and short lifetime test data. This fact makes the method appealing for practical purposes. A simulation is applied for comparison of the introduced simplified regression method with the maximum likelihood method. Comparison is also realized by using experimental data obtained from lifetime recordings during stressing of pressboard and XLPE specimens with 50Hz a.c. voltages of constant value in the high-voltage laboratory.

[1] John C. Fothergill,et al. Estimating the cumulative probability of failure data points to be plotted on Weibull and other probability paper , 1990 .

[2] L. Dissado,et al. Weibull Statistics in Dielectric Breakdown; Theoretical Basis, Applications and Implications , 1984, IEEE Transactions on Electrical Insulation.

[3] E. Occhini,et al. A Statistical Approach to the Discussion of the Dielectric Strength in Electric Cables , 1971 .

[4] H. Meyer. Voltage endurance of electrical insulation , 1967 .

[5] P. McCullagh,et al. An outline of generalized linear models , 1983 .

[6] Hideo Hirose,et al. A Method to Estimate the Lifetime of Solid Electrical Insulation , 1987, IEEE Transactions on Electrical Insulation.

[7] G. C. Montanari,et al. A Probabilistic Insulation Life Model for Combined Thermal-Electrical Srresses , 1985, IEEE Transactions on Electrical Insulation.