Abstract
Frequency Response Analysis (FRA) is considered as a well-established technique to evaluate transformer winding and core structures. This evaluation would be reported in a wide frequency range spectrum. FRA is quite accurate technique; however, it is very sensitive to be affected via the external parameters. In this study, to analyze the external parameters influences on FRA spectrum, different topologies of external lumped parameters, namely series impedance, shunt impedance (ground resistance and capacitance to ground), and parallel to winding impedance are emulated, practically implemented and examined. In order to do that, four statistical indices such as Correlation Coefficient (CC), Euclidean distance, Root Mean Square Error (RMSE) and Absolute Sum of Logarithmic Error (ASLE) are calculated to explore the effect of external resistance and capacitance on Statistical Indicators (SIs). It is revealed that the most significant influence is caused by the impedance parallel to the winding. Moreover, individual analysis of statistical indices assists to identify a healthy region for each transformer unit.