The article describes the possibility of assessing technical state oil-filled power transformers based on technical diagnostics using Bayes’ method. The analysis is based on expert judgment using the membership functions, based on which the calculation of the probability of appearance the different states of power transformers. In this paper we propose to use a mathematical method based on the synthesis of fuzzy linguistic variables and quantitative characteristics of diagnosed equipment. The accuracy of the proposed method is confirmed by the corresponding calculated results, which demonstrate adequate behavior of the model as applied to power transformers.