Estimation of simulated transfer function to discriminate axial displacement and radial deformation of transformer winding

Abstract

Purpose -The transfer function (TF) method is a reliable tool for the detection of the transformer winding deformation and displacement. One of the most important problems is the discrimination between these mechanical defects using a measured TF. The purpose of this paper is to suggest a new method based on the TF estimation, to detect the type of the mechanical defects in high voltage inhomogeneous winding of a power transformer, i.e. the winding is not homogeneous and has interleaved and inverted disks. Design/methodology/approach -The methodology of this paper is based on the estimation of the TF and in normal and defected cases can be properly achieved using vector fitting method. To study the mechanical defects, the detailed model of the transformer is in the frequency range of kHz , f. Findings -In this paper, comparison of Nyquist diagrams of the estimated TF curves result in a new discrimination method between mechanical defects and localization of the deformed section in the transformer winding. The interpretation of the TF changes is based on comparison of the Nyquist diagrams in a selected frequency range. Originality/value -This paper suggests new discrimination method between mechanical defects and localization of the deformed section in the transformer winding which is based on the TF estimation.

On the accuracy of detailed model inductance matrix estimation for air core winding

“…Considering the constraints mentioned in (4) and (5), the error function [13] is defined as (10) and is used to find the parameters of the equivalent of the 19-section circuit model. Although the equivalent circuits have been extracted accurately in a limited search space due to the small dimensions of the detailed model [11,13,2223], Fig.…”

Sensitivity Analysis on Ladder Network Equivalent Circuit Parameters of Power Transformer

“…Also, in spiral type transformers, it can represent the number of turns of windings [6,13]. Although this model is simpler than models based on the traveling-wave theory [14] and multidisciplinary transmission theory [15], it has feasible physical aspects [16]. It should be noted that to obtain satisfactory results from the ladder network, the number of parts should be equal to or greater than the number of peaks in the frequency response diagram [17,18].…”

“…The error function between two curves-the reference FRA test, denoted by r i , and the simulated FRA result, denoted by s i -is defined as follows [16,20]:…”

Intelligent Classifiers in Distinguishing Transformer Faults Using Frequency Response Analysis

“…In [14], the measured TFs from the transformer are mapped to a two-dimensional space, and the fault type is detected based on a vector-based approach. In [15], the transformer's TFs are first estimated, and then a distinction is made between the faults by plotting the Nyquist diagram. Nonetheless, only two faults, AD and RD, have been considered in these studies.…”