Measurement of surface vibration signal of 500 kV transformer and analysis of its frequency characteristics

Abstract By measuring and analyzing the vibration signals of the transformer core and windings, the operating conditions of the transformer can be diagnosed. In this paper, the surface vibration signals of 500 kV autotransformer with different structures are measured, and the frequency domain characteristics of the surface vibration signals are analyzed. The blind source separation algorithm is used to separate the surface vibration signals and obtain the vibration signals of core and winding respectively, which provides strong data support for further condition evaluation and fault diagnosis of the core and winding using their vibration signals. Firstly, the surface vibration signal measuring points are set according to the specific arrangement of the core and winding of the transformer; secondly, the vibration signals of different vibration measuring points of the transformer under the same load and different loads of the same vibration measuring point are measured respectively; finally, the frequency characteristics of the measured vibration signals are compared and analyzed. Then, use blind source separation algorithm for transformer surface vibration signal processing, the two signals obtained are close to the vibration characteristics of core and winding respectively, and the separated, it is proved that the accurate vibration signals of core and winding can be obtained by blind source separation, it is helpful for further analysis of the operation state of the core and winding.