Method for predicting the remaining life of oil–paper insulation system based on stochastic degradation process

Abstract

The service life of a transformer depends on the ageing degree of the internal oil–paper insulation. In order to reliably predict the remaining service life of oil–paper insulation for a single transformer, this study proposes a method based on monitoring the performance degradation data of oil–paper insulation. In this study, the remaining life prediction model of oil–paper insulation is established according to the Wiener stochastic process. The parameters of the model are iteratively updated by using the Bayesian rule and maximum expectation algorithm. The objective of remaining service life prediction of transformer oil–paper insulation is achieved. In the laboratory, multiple groups of oil–paper insulation samples undergo an ageing test at 130°C. Certain amount of performance degradation data is obtained by continuously monitoring the oil–paper insulation. Along with the prediction method of remaining life, the accuracy of the remaining life prediction of transformer oil–impregnated paper is verified when the model parameters are updated. This is a significant guiding factor for predicting the remaining life through the performance degradation data of the continuously operating transformer.