Transformer winding looseness identification method based on local mean decomposition and support vector machine

Abstract

The invention discloses a transformer winding looseness identification method based on local mean decomposition and a support vector machine. The transformer winding looseness identification method comprises the following steps: step 1, respectively acquiring vibration signals of a normal state and a winding looseness state at the moment of closing a transformer; step 2, performing variational mode decomposition on the acquired vibration signals, so that each PF component can be obtained; step 3, calculating the energy and singular value of each PF component and the permutation entropy and singular spectrum entropy of the reconstructed signal; step 4, selecting features with relatively high precision through a Fisher-Score method to form a feature vector group; step 5, training the simulated annealing optimized support vector machine model by using the training sample set; and step 6, using the obtained support vector machine model as a classifier to carry out classification and identification on the test sample set so as to realize fault diagnosis. According to the invention, the loosening state of the transformer winding can be identified at the moment when the transformer is switched on, early warning of the transformer is realized, and a novel method is provided for transformer vibration signal feature extraction and fault diagnosis.

Description

Technical field [0001] The present invention belongs to the field of mechanical fault diagnosis of power transformers, and more particularly to a transformer winding loosening recognition method based on local mean decomposition and support vector machine. Background technique [0002] Transformers are complicated in power systems, high prices, high prices, key effects on the safety stability of the power system, and once the transformer is faulty, it will lead to major economic losses. Therefore, monitoring and troubleshooting the transformer operation, discovering the risk of malfunction in advance, improving the reliability of transformer, is of great significance to ensure reliable operation of transformers. [0003] The transformer is prone to various faults due to long operating time, such as iron core loose, winding deformation, deformation, and the like. At present, the mechanical fault state monitoring method of power transformer windings, iron cores is: short circuit im...

AI Technical Summary

Problems solved by technology

[0004] In view of the above problems, the present invention provides a transformer winding loose identification method based on local mean decomposition and support vector machine, which solves the problem of misjudgment in the diagnostic method in the prior art

Images