Application of adaptive neuro fuzzy inference system to support power transformer life estimation and asset management decision

Abstract

Power transformer is a critical asset in electrical transmission and distribution networks that need to be carefully monitored during its entire operational life. Considering the fact that a significant number of worldwide in-service power transformers have approached the end of their expected operational life, utilities have adopted various transformer condition-based maintenance techniques to avoid any potential catastrophic failure to the equipment. The extent of transformer insulation system ageing can be quantified through measuring several diagnostic indicators such as interfacial tension number of the insulating oil which has a strong correlation with the number of transformer operating years. Moisture and furanic compounds generated due to paper insulation degradation are indicators for solid insulation ageing. This paper introduces a new adaptive neuro fuzzy logic model to estimate the life of mineral oil-filled power transformers based on the values of insulating oil interfacial tension number, furan content in oil and the moisture content within the cellulose insulation. Also, an integrated asset management decision model is proposed. Results of the proposed model are validated against practical data collected from utility and industry mineral oil-filled power transformers of different ratings, designs, operating conditions and lifespans.