Fuzzy Logic Expert System for Optimum Maintenance of Power Transformers

Condition monitoring of power transformers improves the reliability and the safety of an electrical power network. It protects the transformers from failures, and avoids huge revenue loss to utilities and customers. The health status of transformers is decided by several influencing factors. An accurate health assessment of transformers has been a challenging task for the researchers as well as the transformer diagnostic experts. In the present paper, a new fuzzy logic model is proposed to assess the impact of all such factors on the overall health index of the transformers, and in turn their health status. It incorporates the information obtained from various attributes which have been categorized into four different groups namely, transformers tests and measurements, visual inspection, maintenance and operation history, and design and fabrication. These attributes are initially assigned appropriate scores based on their present status/condition. Further, the proposed fuzzy logic model is developed based on the total scores attained by the four design categories. The present analysis avoids the over impact of the attributes thereby resulting in a more reliable and accurate health index of transformers. It overcomes the shortcomings of the conventional health assessment methods of transformers.