Failure analysis and prediction for transformers in the application of power systems

Abstract

Transformer failure analysis has been an important field of study in the power system as transformers amount to the significantly high capital investment in transmission and distribution systems. Among all the failure modes, a most important factor which affects the operational transformer life expectancy is the thermal degradation of the insulation. Many electrical power companies have been trying to improve transformer’s asset management to prevent failures for transformer to maintain its high reliability and efficiency throughout its lifetime. As technology advances, the companies are advancing from preventive maintenance to a predictive maintenance. The aim of the project is to study the load curve parameters which influence the change in hot spot temperature and predict the transformer failure with an online predictive program. For more accurate insulation life expectancy calculations, in this paper, a model called the cumulative moving average (CMA) will be implemented to the IEEE Standard (C57.91-2011) theoretical calculation. This project discussed the relationship between the load curve, hottest spot in the winding and the loss of life in the transformer using a simulated online predictive program.