Background. The increase of technological violation intensity and its consequences severity is caused mainly by objectively existing aging and service life depletion of electrical equipment. The power industry liberalization exacerbates the reliable operation problem of the power system and requires identification of power system operation accompanied by the maximum emergency risk with possible cascade accidents development. Therefore, the model development task for assessment of the equipment failure risk based on the diagnostic results of technical condition in particular under short circuit in the external network is relevant today. Objective. The aim of the work is to develop a fuzzy mathematical model for probability assessment of power transformer failure in the presence of a windings defect, short circuit in external network and emergency risk assessment under power transformers out of service. Methods. The fuzzy set theory and fuzzy logic were used for developing a mathematical model of risk assessment of power trans- former failure. The problems of determining the “weak” in terms of power transformers reliability of power systems based on the results of failure risk assessment due to external short circuits were solved by methods of fuzzy logic and probabilistic-statistical simulation of electric power system modes. Results. The necessity of complex simulation of electric power system modes is substantiated for probability assessment of power transformer failure under electrical network disturbances. The simulation of technical condition of power transformer windings was carried out. The short circuit influence on operability level of power transformers of electric power system was investigated. The quantitative indicators of operational risk of electric power system were determined under power transformers out of service. Conclusions. The linguistic mathematical model for estimating the failure probability of power transformer windings in the presence of defect and short circuit in electrical network has been developed to determine the quantitative indicators of emergency risk in power system.