Diagnostic statistical characteristics of dielectric strength of power transformer oil insulation

Advancement of methods for diagnosing internal insulation of oil-filled transformers has led to the use of statistical characteristics of the dielectric strength of oil to achieve this. These characteristics are determined based on the results of tests in a standard arrester. The effectiveness of utilizing such diagnostic characteristics depends on the extent to which they are explored and their standard parameters established. Therefore, it is essential to carry out studies that will facilitate informed selection of the diagnostic statistical characteristics of dielectric strength of transformer oils and their standard indicators that accurately reflect the operational features of the transformers. The study relied on the findings from the operational tests of transformer oil carried out on a standard arrester connected to operating transformers, using probability theory and mathematical statistics for processing the data. The diagnostic statistical characteristics of dielectric strength for operating oils in a standard arrester were determined using the traditional methods and the Gnedenko-Weibull distribution parameters. The limits of change in the values of these characteristics and their compliance with the standard values were established. The study revealed that the standard value of the coefficient of variation in the average dielectric strength decreases the requirements for the transformer oil quality. As evidenced by the findings, the values of the lower limit of the dielectric strength of the oil exhibit the presence of impurity particles in the oil. It was established that the obtained values of the diagnostic statistical characteristics of the dielectric strength of transformer oils, along with their variation limits, could be used to select more informed standard indicators of the diagnostic characteristics for assessing the internal insulation of transformers to enhance their operational reliability.