A Novel Methodology for the Transformer Winding Equivalent Ladder Network Circuit Parameters Identification by Employing the Frequency-Domain and Population Based Method

Abstract

Frequency response analysis (FRA) is the widely used technique for diagnosis of power transformer winding. Previous works in the field of FRA have steer to the standardization of its measurements procedure. To date, there is no reliable standard code for the interpretation of FRA results. In this context, this paper proposes a new parametric method to synthesis the High frequency electrical ladder network circuit (HF-ELNC) of the transformer winding. Initially, a nondestructive process is applied to extract three main winding parameters. Mainly, the shunt-series capacitances and the total equivalent inductance basing on the frequency response data from the winding terminal. To this end, the proposed continuous-time state–space model is converted into a gain-numerator-denominator form. After that, the derived matrix equations are iteratively estimated by population-based algorithm using enhanced logistic choatic marine predator algorithm (ELCMPA), from where, three objective functions are considered in the model of identification including the capacitances, the inductances and the resistances. As well, classical methods such as finite elements or analytical formulas require design winding specifications. In this study, instead of winding geometry knowledge, the proposed method certifies a unique, physically and mutually coupled HF-ELNC. The accuracy of the proposed method is validated through case study of a real transformer winding.