Simulation Research on Sweep Frequency Impedance Characteristics of Transformer Winding Based on Three-Phase Three-Winding Lumped Parameter Model

As a mature and effective off-line detection method of winding faults inside the transformer, the sweep frequency impedance (SFI) method is widely utilized in this field. In SFI test, the other windings, especially the short-circuited secondary winding, are present, and therefore ambiguity arises when diagnosing winding faults that any windings may contain the faulty parts. However, few of the current papers relevant to SFI method study focus on this important issue. In this paper, based on PSPICE software, simulation research is carried out to study the effect of faults that occurred within different windings on SFI curves. At first, a complete three-phase three-winding lumped parameter network based on a transformer with YNynD11 connection is established, and the high to medium SFI curve of the network is computed. Afterwards, based on the complete three-phase three-winding model, the variation trend of SFI curves when short-circuit faults occur on different windings is studied and the Pearson correlation coefficient (PCC) between different SFI curves is calculated. In this way, the influence of short-circuit faults within different windings on SFI curves is analyzed. The simulation results indicate that the SFI curve shows obvious change with lowest PCC value when the tested winding sustains short-circuit fault than that when failure occurs on other windings, which implied that SFI method is merely sensitive to detect the failures on the tested winding.