Using Statistical Decision Methods to Correct the Maximum Permissible Values of Transformer Oils Indicators

Abstract

The paper presents the results of adjusting the maximum permissible values of transformer oils indicators using statistical decision methods. The features of practical determination of maximum permissible values of breakdown voltage of transformer oils using integral functions (survival function), minimum number of wrong decisions method, minimum risk method and Neyman-Pearson method are considered. Results of the comparative analysis of values of probabilities of correct decisions, I- and II-type errors, wrong decisions, correction of maximum permissible values of breakdown voltage and values of possible risks which accompany the use of maximum permissible values obtained with the use of these methods and regulated by international and national standards are presented. The results of the analysis show that it is more appropriate to correct the limit values of oil indicators using the minimum risk method. At the same time, the results did not allow the authors to recommend the minimum number of wrong decisions method for correction of maximum permissible values of oil indicators.