Transformer winding and iron core vibration signal separation method based on RBF neural network

Abstract

The invention discloses a transformer winding and iron core vibration signal separation method based on an RBF neural network. The method comprises the following steps: when a transformer is under different working conditions, collecting a vibration signal of a transformer winding, a vibration signal of a transformer iron core and a mixed signal of the transformer winding and the iron core; a mixed signal of a transformer winding and an iron core being used as an input layer of a radial basis function neural network; a vibration signal of a transformer winding and a vibration signal of a transformer iron core being used as an output layer of a radial basis function neural network; training the radial basis function neural network; separating the mixed signal of the transformer winding andthe iron core to be separated by using the trained radial basis function neural network. By means of the method, the vibration signals of the transformer winding and the vibration signals of the transformer iron core can be accurately separated, and therefore the vibration characteristics of the winding and the vibration characteristics of the iron core can be accurately reserved.

Description

technical field [0001] The invention belongs to the technical field of separation of transformer vibration signals, and relates to a method for separating vibration signals of transformer windings and iron cores based on an RBF neural network. Background technique [0002] Transformer is one of the most important equipment in the power system, and its safe operation has been closely related to the national economic development. If the transformer fails, it will cause a large-scale power outage, which will not only affect the production of the factory, but also affect the lives of the people. Therefore, in order to discover potential accidents of transformers in time, avoid sudden accidents, and improve the reliability of transformer operation, it is of great significance to carry out research on transformer fault diagnosis methods. [0003] Based on the current research status of transformer vibration signal separation at home and abroad, the blind source separation algorit...

AI Technical Summary

Problems solved by technology

[0003] Based on the current research status of transformer vibration signal separation at home and abroad, the blind source separation algorithm represented by the independent component analysis algorithm (ICA) has a high requirement for the independence of the source signal, but due to the vibration signal of the core and the vibration signal of the winding The similarity in amplitude and frequency spectrum is very high, and the reliability of the mixing matrix is ​​doubtful. Therefore, the signals separated by blind source separation algorithms such as mainstream blind source separation cannot accurately retain the vibration characteristics of windings and cores.

Images