Transformer Winding Fault Analysis Based on Oscillatory Wave Binarization Image Processing

Frequency response method is easy to interfere with transformer winding fault identification in the field, and experienced engineers are needed to evaluate the winding correctly. In this paper, a method of obtaining transformer winding oscillatory wave by high voltage transient excitation is presented, and the feasibility of the method is verified by field test. Then the oscillatory waves under different fault degrees and different fault positions are obtained by using the oscillatory wave test platform, the characteristics of the oscillatory waves are extracted based on the binarization image processing method. The results show that the oscillatory wave can effectively diagnose the transformer winding fault, and the fault degree and position of the axial shift of the winding can be further identified by extracting the characteristic parameters, which provides a reference for the state evaluation of the transformer winding.