Design of an oil immersed power transformer monitoring and self diagnostic system integrated in Smart Energy Management System

Abstract

The power transformers are one of the most important components in the electrical grid, where the failures are highly critical and can impact all sources and terminals, including production and distribution systems. This paper presents the interactions of the power transformer failures classification algorithms and health index calculation with the smart energy management system in an electrical grid and proposes a new monitoring approach of this component. This system can classify the different power transformer failures using predictive analytics. The integrated models calculate important characteristics such as loss of life, health index of oil quality, dissolved gas analysis and insulation paper. Since each fault of power transformer has an impact on the grid, the objective is looking for new correlations between different monitoring techniques of the power transformer with the power quality data acquired in the primary and secondary. These correlations results can reduce the cost of the instrumentation and the calculation time. A list of the requirements to design an optimized intelligent and online predictive maintenance system for power transformer is also proposed, integrating instrumentation, important KPIs and data to be monitored, stored, and analyzed. These data are used to classify the critical defects, using probabilistic models and machine learning algorithms to propose maintenance rescheduling and an effective predictive maintenance plan to avoid blackouts.