Influence of Oil Temperature on Repeatability of Measurements in Frequency Response Analysis of Power Transformers

Power transformers are one of the key apparatus in electric power networks. To ensure the reliability of power transformers, the diagnosis of mechanical faults, such as the deformation and displacement of windings, is important. Frequency response analysis (FRA) has attracted considerable attention as a diagnostic method for mechanical faults of power-transformer windings. In FRA, the frequency responses of power transformers, namely, transfer functions (TFs), are measured in a wide frequency range. Faults are detected by detecting changes in TFs compared with reference data measured when the power transformers were operating normally. Therefore, it is important to ensure the repeatability of measurements. In this study, the TFs of two power transformers are measured shortly before and after a temperature-rise test and a subsequent cooling stage. Changes in TFs due to a change in oil temperature are found at frequencies higher than 400 kHz. Furthermore, the influence of the oil temperature on the repeatability of the measurements in FRA is objectively evaluated by employing a numerical index, which represents the identification degree of two compared TFs. It is found that the influence of the oil temperature is negligible when the average oil temperature is less than 50 degrees C.