Frequency Response Function of Short Circuit Vibration for Power Transformer

Offline method cannot make online diagnosis and assessment of transformer winding's mechanical condition once short-circuit occurs. A new method for characterizing the mechanical features is proposed via acquiring the frequency response function (FRF) of winding vibration under short-circuit impulse. The vibration signal from transformer oil tank and the short-circuit current from recorder are synthesized to establish a single-input and multiple-output vibration system, and the FRF of winding vibration is acquired in terms of the auto-power spectral density function. The correlation analysis determines the correlation coefficient (CC) of FRF under multiple short-circuit impulse. The short-circuit test for a 220 kV power transformer indicates that under multiple short-circuit impulse, the CC of FRF and the deviation of phase reactance obey the negative correlation, namely, the CC decreases with the increasing phase reactance deviation, and the larger the deviation of phase reactance, the smaller the CC. Compared with the traditional offline methods, this method takes advantages of the recorder information in the accident, and employs the FRF of winding vibration as well as CC between two short-circuit impulses to characterize the mechanical features of transformer winding.