Diagnosis Method of Transformer Winding Deformation Based on Discrete Wavelet Transform

Abstract

The invention provides a transformer winding deformation diagnosis method based on discrete wavelet transform, which applies the discrete wavelet transform method to frequency response analysis and diagnosis of transformer winding deformation. The present invention uses discrete wavelet transform to decompose and process the measured waveform and reference waveform obtained by frequency response analysis at multiple levels, then uses the processed smooth data to perform mathematical statistical analysis, and diagnoses the degree of deformation of transformer windings according to corresponding criteria. This diagnosis mode can effectively reduce the error of the traditional diagnosis mode caused by the white noise interference and some local relatively strong interference points during the frequency response measurement, and can accurately diagnose the deformation degree of the transformer winding. The estimation of the development trend provides a new judgment method.

Description

technical field [0001] The invention relates to the field of transformer winding deformation diagnosis, in particular to a transformer winding deformation diagnosis method based on discrete wavelet transform. Background technique [0002] According to incomplete statistics of the State Grid Corporation of China, during the five-year period from 2002 to 2006, there were 162 transformer accidents with a voltage level of 110kV and above in the State Grid System, which has become a major challenge for the power system to achieve safe and stable operation. serious problem. Especially when the main transformer is directly connected to the generator, its damage will force the generator to stop generating power. An accident occurs when a large power transformer is running in the system, which may cause a large-scale power outage. It is very difficult to repair the damaged coil. The maintenance period generally takes more than half a year, which costs a lot and affects a wide range ...

AI Technical Summary

Problems solved by technology

The short-circuit impedance method is a traditional method for judging winding faults, which has the advantages of short time consumption, but the sensitivity is not high; the vibration signal analysis method judges the state of the transformer based on the vibration signal on the oil tank, which cannot fully reflect the short-circuit fault of the winding. There are many influencing factors; the frequency response analysis method is a method to measure and analyze the frequency response characteristics of windings in a wide frequency band, and judge the winding state. It has the advantages of high sensitivity, simple wiring, and wide frequency bandwidth. It has been widely used in transformer windings. Deformation detection

[0004] However, further studies have shown that the frequency response analysis method still has the following defects or deficiencies: it is easily interfered by background noise, low-frequency resolution is low, and the excitation source energy is limited.

Images