Adaptive management of technical condition of power transformers

Abstract

Ensuring reliable operation of power transformers as part of electric power facilities is assigned to the maintenance and repair system, whose important components are diagnostics and monitoring of the technical condition. Monitoring allows you to answer the question of whether the transformer abnormalities and how to do they manifest, while diagnostics allow determining the nature, the severity of the problem, determine the cause and possible consequences. The article presents the results of the authors ' research on creating an algorithm for adaptive control of the technical condition of power transformers using diagnostic and monitoring data. The developed algorithm implements the decision-making procedure for ensuring the reliable operation of oil-filled transformer equipment as part of the substations of electric power facilities. The decision-making procedure is based on the method of statistical Bayesian identification the states of a transformer based on the results of dissolved gas analysis (DGA) in oil. The method is characterized by high reliability of recognizing defects in the transformer and the ability to adapt the probabilities of the obtained solutions to the newly received diagnostic information. These results illustrate the effectiveness of the developed approach and the possibility of its application in the operation of oil-filled transformer equipment.