Expertise method to diagnose transformer conditions

Abstract: In this paper we have deduced a new method combining the classic dissolved gases analysis (DGA) techniques for diagnosis of fault transformers. Based on the interpretation of DGA, a prototype of an expertise system for diagnosis of suspected transformer faults and their maintenance procedures is proposed. The requested estimation is based on historical information provided by the electric company Iberdrola S.A. and its employees An expertise method has been developed in order to diagnose transformer conditions, model its aging process, and finally estimate its residual life. In the process of information collecting it is necessary to know transformer data and characteristics, the results of gases analyses as well as the results of the physical-chemical analysis, the records of the machine (carried out revisions, damage repaired, oil degassed and/or filtered, location changes, etc.). This method is being tested by the utility Iberdrola S.A. to demonstrate its feasibility