New SFRA measurement interpretation methodology for the diagnosis of power transformers

Abstract

The analysis of the frequency response by the Sweep Frequency Response Analysis method (SFRA) is a diagnosis technique that can detect displacements, deformations and other mechanical and electrical failures in power and distribution transformers. One of the main disadvantages of the method is the lack of an international agreement regarding the methodology for measurement analysis, which is usually done by experts in the field, who normally make the diagnosis with the aid of statistical parameters such as correlation coefficient and standard deviation, or parameters derived from modeling the frequency response as a complex transfer function, represented by poles and zeros, or poles and residues. This paper presents a new methodology for SFRA measurement analysis, which makes use of the properties of various types of parameters and expert knowledge through the application of fuzzy causal diagnosis.