A Bayesian Method for Transformer Life Estimation Using Perks´ Hazard Function

This paper introduces PJM Interconnection´s new method to model and predict transformers´ service lives. A simplified Perks´ equation is used to model the hazard rates, which will be used synonymously with retirement rates in this paper, of transmission transformers. A Bayesian Method is used to model the uncertainty in the three parameters of Perks´ equation. This method is different from the traditional Iowa Curve method in that: 1) it gives a continuous family of transformer life distributions as well as the extinction likelihoods, rather than a limited number of distribution patterns and 2) a closed-form distribution can also be deduced with only three parameters, which can assume all the three distribution patterns of Iowa Curves: right, left, and symmetrically modal. The method has been successfully applied to the statistics on hundreds of large electric transformers in PJM. Example problems are provided and solved with this new method