Transformer Winding Deformation Classification Method Based on Scanning Impedance Method and Support Vector Machine

Abstract

Based on the scanning impedance method and support vector machine classification method for transformer winding deformation, the short-circuit impedance is measured and calculated in a wide frequency domain under normal operating conditions. Construct the scanning impedance amplitude-frequency curve and phase-frequency curve of the normal transformer from the amplitude and phase data of the obtained short-circuit impedance as the fingerprint track; when the transformer has a winding deformation fault, under the same test conditions, use the same method to construct this state The amplitude-frequency curve and phase-frequency curve of the scanning impedance; under the complex condition of the scanning impedance, the eigenvector is determined based on the statistical index and the characteristics of the scanning impedance waveform, and the eigenvector library is constructed according to the obtained characteristic curves of different winding deformations; according to different types of The eigenvectors of the winding deformation feature extraction are classified by the support vector machine optimized by the particle swarm optimization algorithm, and the deformation type of the winding is finally determined according to the classification results, which provides a new way for the detection of the small deformation of the transformer winding.

Description

technical field [0001] The invention relates to the field of detection of winding states of power transformers, in particular to a method for classifying deformation of transformer windings based on a scanning impedance method and a support vector machine. Background technique [0002] As the key equipment of the transmission line, the transformer is responsible for transforming the voltage to ensure better transmission of electric energy. When it fails in operation, it will cause a major impact on the local or even the overall power system. During the transportation, installation and operation of the transformer, the winding deformation will be caused by irresistible factors such as human misoperation, external force and other natural events. Therefore, the power department attaches great importance to the state detection of the transformer, especially the prevention and control of gradual faults such as winding deformation detection is particularly important. According to...

AI Technical Summary

Problems solved by technology

However, the current research based on the scanning impedance method is still limited to the detection method itself, and there are few studies on the fault classification using the distinguishable feature vectors extracted from the data obtained by the scanning impedance method.

Images