Data driven model for the prediction of the expected lifetime of transformers

Abstract

Power transformer is one of the most expensive equipments in the electrical power grid. Transformer outages may lead to substantial economical losses. One of the most important parameters governing a transformer's life expectancy and reliability is transformer aging condition and loading condition. At the same time, it is found that electric consumption is highly correlated with weather condition, therefore, in this study, we present a solution that firstly, predict the probability that a transformer Accelerated Aging Event (AAE) happens under the given weather condition; secondly, to predict the severity of the AAE, i.e., the scale of the accelerated aging factor for the event. Finally, we computed the expected life expectancy of the transformer under possible weather conditions trajectories during transformer's remaining life span using Monte Carlo simulation.