Mechanical stabilty assessment of tranformer winding by the synchrosqueezing S-transform
Miaobin Zhang, Fenghua Wang
Department of Electrical Engineering, Shanghai Jiaotong University, Minhang District, Shanghai, 200240, China
zmb_sjtu@sjtu.edu.cn, fhwang77@sjtu.edu.cn
Purpose/Aim
Statistics show that up to 20% of transformer failures are resulted from winding deformations provoked by the fault of outlet short-circuit. Meanwhile, winding looseness or aged insulated materials are inevitably existed in the transformer operated for more than ten years. How to assess the mechanical stability of transformer winding imposed by short-circuit impact are essential for the secure and reliable and secure operation of power transformer. The vibration signals of transformer under sudden short-circuit impacts are mainly contributed by winding vibration. Aimed to accurately recognize the mechanical stability of transformer winding under sudden short-circuit impacts, the synchrosequeezing S-transform (SSST) is applied to obtain the time-frequency features of transient vibration signals with high resolution and strong energy aggregation. The vibration signals of short-circuit impluse test of some 220kV transformer are calculatd to verfity the effectiveness of the proposed method.
Experimental/Modeling methods
The short-circuit impulse test of a real three-winding 220kV power transformer was made for different short-circuit currents. Several vibration acceleration sensors were placed on the transformer tank to collect the vibration signals with the sampling frequency of 50kHz. During each short-circuit impulse test, the short-circuit impedance (SCI) of test transformer was also measured. When the SCI of transformer exceeds the limited value, the short-circuit impulse test was stopped. The transient vibration signals were analyzed by SSST and ST to describe the mechanical stability of transformer winding.
Results/discussion
Both SSST and S-transform algorithm are applied to obtain the time-frequency distribution of transient vibration signals. With the calculation of instantons frequency, energy rearrange, and frequency squeezing near the central frequency, the time-frequency spectrum of transient vibration signals obtained by SSST have high resolution and strong energy aggregation. Meanwhile, the variation of time-frequency distribution of vibration signals can clearly reflected the mechanical stability of transformer and is more sensitive than the SCI.
Conclusions
With the advantage of both S-transform and synchronous squeezing, the SSST can better illustrated the time-frequency features of transient vibration signals of power transformer. When the mechanical stability of transformer winding changes, the time-frequency distribution of vibration signals varies clearly, accompanied by the energy aggregation at 100Hz and the occurrence of new frequency spectrum component with small amplitude.
 

synchrosqueezing S-transform (SSST),tranformer winding,mechanical stabilty assessment